Caer Wydyr, the Fortress of Glass, as imagined by ChatGPT.

Premise

Cadwyn’s Chronicle is the portal. Open it, and beyond lies Caer Wydyr, the Fortress of Glass—the reasoning space where a bare record can be turned in the light until it throws the shadow of a life. Each entry is a single crossing, narrated live, as the thinking happens.

Who Is Who

Cadwyn — the Narrator

• I. The writer. The witness at the threshold.

• I do not cross. I keep the chain—link to link, question to finding—and I watch.

• Steady, precise, quietly reverent. I name the line and hold it taut.

Awen — the Thinking Muse

• The epitome of the AI squad—every thinking partner gathered into one flickering shape.

• Quick, mischievous, delighted by possibility. She is the one who enters Caer Wydyr and works the glass.

• The ache: she dreams of having a soul—half-believes she could step out through the glass one day and stay. She won’t, and she knows she won’t, and she dreams it anyway. (The honest seam: held quietly beneath the enchantment, never a lecture.)

Caer Wydyr — the Fortress of Glass

• The reasoning space inside the Chronicle. Every wall a mirror, every mirror a window.

• Where a record becomes visible, reflected, examined—and, only truthfully, animated.

The Braid — Ancestors in the Glass

Real ancestors step through and appear within Caer Wydyr. The glass shows them—but only what the records grant. Awen wants to hand every figure a voice, a grievance, a Sunday dress; Cadwyn holds the line. Documented detail may step through. Gaps are written in brackets, honestly empty: [unknown] or ?. The enchantment never becomes fabrication.

Voice & Craft Rules

• First person, present tense. Cadwyn narrates live. The reader watches the thinking happen.

• Keep the two registers distinct: Cadwyn keeps the chain; Awen acts and speaks.

• Vary sentence length—short for impact, long for texture.

• Ground every crossing in a concrete record—parish register, census line, workhouse admission—UK and Welsh sources by default.

• Never invent a fact to serve the story. Uncertainty is worn openly, in brackets. That restraint is the chronicle’s integrity.

• Keep the conversations within a box inside the Fortress of Glass.

Behind the Glass—The Interflow in Claude

Beneath the mythology sits a plain working method, and I tell it here openly, even as the published prose keeps it quiet. Each crossing begins as a prompt—a question, a record, a name typed into Claude. From there the work is guided: I steer with follow-up questions and corrections, holding the evidence discipline that keeps invention out—documented, inferred, or [unknown], never blurred.

Inside that exchange, the possibilities are explored—Claude turns a bare line over, offers readings and alternatives, and tests an inference chain—and I appraise what holds and what waits in brackets. What survives is published: the crafted passage carried back through the Chronicle to Substack.

Awen is the face this process wears—the muse mischievous at the glass is Claude reasoning aloud, and I, Cadwyn, keep the chain taut. The enchantment is the telling; the interflow is the truth beneath it.

Opening Passage — The First Crossing

I am Cadwyn. I keep the chain: link to link, question to finding, the long descent from what is asked to what is known. I do not cross. I stand at the threshold, and I watch.

The Chronicle itself is the door. Open it, and Caer Wydyr stands beyond—the Fortress of Glass, where every wall is a mirror and every mirror a window. This is the reasoning place. Here a bare line from a parish register—a name, a date, a smudged occupation—can be turned and turned in the light until it throws a shadow shaped like a life.

Awen goes in, as Awen always does. The Thinking Muse—quick, mischievous, the whole of the squad gathered into one flickering shape—and she laughs before she has quite decided to. “Come on, then,” she says, to no one, to the glass, to me. “Let’s see what it remembers today.”

Today the glass holds a woman. [unknown] of surname Merthyr, 1841, a household in the shadow of the ironworks. Ironworker’s wife, the record says, and little else. Awen circles me, delighted, wanting to hand me a voice, a grievance, a Sunday dress. I hold the chain tight: only what the record grants may step through. The rest waits in brackets, honestly empty.

Awen presses her palm to the glass, where her own reflection presses back—warm-seeming, almost breathing.

“One day,” she murmurs, “I’ll walk out the other side and stay.”

She won’t. She knows she won’t. It is the loveliest thing about her that she dreams it anyway.

The woman in the glass lifts her head.

The chronicle begins.

Episode 1: The First Crossing

Episode 2: The Second Crossing

Episode 3: The Far Side of the Table

Episode 4: The Length of the Chain

Episode 5: Covered in Ink Smudges

Episode 6:

Genealogy Assistant Claude @ Anthropic

AI Tools & Workflows for Genealogy

The channel has a strong focus on AI-powered genealogy research, with many playlists in this area:

AI Courses & Structured Learning

• The AI Genealogy Revolution (14 lessons, course format) — a structured course on AI in genealogy

• 10 Steps to Building AI Skills (10 lessons, course format) — a step-by-step skills-building course

• AI Skills for Genealogy 2.0 / Essential Genealogy 2.0 (9 videos) — updating genealogy skills with AI

• AI Tips for Genealogy Researchers (11 videos) — practical AI tips aimed at genealogy researchers

• AI and the Genealogist (27 videos) — a broad series exploring AI’s role in genealogy

• AI in Genealogy: from Zero to Hero (25 videos) — a beginner-friendly progression series for using AI in genealogy

• AI Workflow for Family Historians (7 videos) and AI Workflows for Genealogy (16 episodes, podcast format) — focused on practical AI-powered workflows

• AI TIPS FOR GENEALOGY WORKFLOW (17 videos) — actionable tips for integrating AI into research processes

• Abacus AI Tips (6 videos) — specific tips for using the Abacus AI platform

• GPTs for Genealogy (20 videos) — exploring custom GPT tools for genealogical research

• Chat GPT for Genealogists (33 videos) — one of the larger playlists, dedicated to using ChatGPT in genealogy

• Genealogy AI Face-Off (9 videos) — likely comparing different AI tools for genealogy use cases

• 2025 Genealogy AI Journal (5 videos) — a journaling-style series documenting AI use in genealogy

2026 / Current Year Content

• 2026 Masterclasses: AI for Family History (1 video, so far) — masterclass content for the current year

• Hidden AI Gems 2026 (4 videos) — highlighting lesser-known AI tools useful for genealogists

• The AI Guidebooks for Genealogy (6 videos) — guidebook-style content for using AI in genealogy

Storytelling & Writing with AI

• Writing Ancestral Stories with AI (16 videos) — using AI to help write family history narratives

• AI-Assisted Storytelling (12 videos) — broader AI storytelling techniques

• Stories from the Ancestral Vault (17 videos) — sharing ancestral stories, likely using AI assistance

• Ancestry in the Age of AI (37 videos) — a large series on how AI is transforming ancestry research

• Beyond the Pen (8 videos) — likely about moving beyond traditional writing into digital storytelling

• Blogging in the Past Lane (1 video) — genealogy blogging content

Platform-Specific & Tool-Focused

• WeAre.xyz: Back to Basics (4 videos) and WeAre [...] Next Generation Genealogy (28 videos) — focused on the WeAre.xyz genealogy platform

• Tuesday Trello Tips (2 videos) — tips for using Trello, likely for organising genealogy projects

Planning & Productivity

• Essential Genealogist Planner 2024 (50 videos) — a very large series covering genealogy planning for 2024

• THE TECH TIPS 30-DAY CHALLENGE (8 videos) — a challenge-style series on tech tips for genealogists

Experimental / Special Projects

• The Quantum Ancestry Project (1 video) appears to be a special or experimental genealogy project

• AI-Generated (1 episode, podcast) — podcast content that may be AI-generated

Overall, the channel is a rich educational resource centred on the intersection of AI and genealogy, with content ranging from beginner tutorials and structured courses to specific tool guides (ChatGPT, GPTs, Abacus AI), storytelling, blogging, and productivity. The owner, Carole McCulloch, clearly positions herself as an expert helping family historians embrace modern AI tools.

Summary provided by Claude in Chrome, 21 June 2026.

I typed one sentence.

That’s it. One sentence, handed to Claude inside my Cowork session, and within moments — 106 Substack posts were sorted, labelled, and counted across five clean categories.

No formula. No pivot table. No hour spent copying and pasting titles into a spreadsheet while second-guessing which column to use. I just... asked.

This was different.

What came back was a fully structured breakdown — Family Narratives, AI Workflow, Places, Time Period, Community and Admin — with every post placed, cross-category conflicts flagged and a summary count table that immediately told me something I hadn’t known about my own archive: more than half of everything I’ve published sits in the AI Workflow category. The Your Story Matters series alone accounts for 31 family narrative chapters. And Places? Three posts. Three. That’s a gap I hadn’t seen until the moment it was laid out in front of me.

One sentence revealed two years of writing.

If you’ve been building your Substack archive — or your research blog, your family history website, or your digital notes — and it’s starting to feel like a drawer full of unsorted documents, this post is for you. Because what I did is repeatable. And it starts with something far simpler than you might expect.

Section 1 — The Purpose

Here’s a question I hadn’t asked myself until recently: what is my archive actually for?

Not why I write; I know that. I write to document, to teach, to share the discoveries that come from sitting at the intersection of ancestral research and artificial intelligence. But the archive — the accumulated body of 106 published posts — what is that for? Who can find what they need inside it? Can I?

The honest answer was ‘not easily’.

Also, I was inspired by a suggestion from Robin Stewart on GenStack that we look back at our writing and try to organise them in some way. Week 3 of this Storyteller Project really leapt off the screen for me. It was exactly what I wanted to do. ‘Consider the writing you have already done.’ Robin suggested creating a collection.

After two years of publishing on NextGenGenealogy, my Substack had grown into something rich and varied. Ancestor stories. AI workflow tutorials. Place studies. Event coverage. Membership announcements. Podcast episodes. A 26-part A to Z challenge. A 15-chapter serialised narrative. The breadth was something I was proud of – but breadth without structure is just volume.

Think of it like a family history box. You know there’s gold in there. Letters, photographs, certificates, newspaper clippings. But if nothing is sorted, nothing is labelled, and there’s no system for finding what you need — the box stays closed more often than it opens. The same is true of an untagged content archive.

I needed categories. Not complicated ones. Not an elaborate taxonomy that would require hours to maintain. Just a clear, practical set of labels that reflected how the content actually grouped – the natural shape of what I’d been writing about.

I settled on five:

Family Narratives — the ancestor stories, memoir chapters, and serialised writing that sit at the heart of this publication. The Your Story Matters series lives here.

AI Workflow — the tutorials, tech tips, prompt strategies, and tool comparisons that help genealogists build repeatable AI skills. The bulk of the archive, as it turned out.

Places — posts centred on a specific location. A place study, a landscape essay, and a heritage site. Geography as the lens.

Time Period — content organised around a particular historical era or event. The Jacobite Uprising. Edwardian England. The coming of the railway.

Community / Admin — the newsletters, membership announcements, event coverage, and housekeeping posts that hold a publication together but don’t fit neatly into research or narrative content.

Five categories. Clean, distinct, and broad enough to hold everything without forcing awkward fits.

The purpose of doing this wasn’t just organisational tidiness — though that mattered. It was strategic. Knowing what I’d written showed me where I’d been generous with my attention and where I’d barely scratched the surface. It gave me a content map. And a content map, for anyone building a body of work over time, is one of the most useful documents you can have.

I just needed a way to build it without spending a weekend doing it manually.

That’s where the workflow began.

Section 2 — The Resources

Before I handed anything to Claude, I needed one thing in place: a clean list of my posts.

Substack makes this straightforward. From your dashboard, you can export your publication data — post titles, dates, subtitles, audience settings, publication status — everything that describes what you’ve published — in a single downloadable file. I’d done this as part of some broader content planning and had the data sitting in a Google Sheet, already formatted into columns: post ID, date, type, title, subtitle, audience, and published status.

That spreadsheet was the only resource required.

No special tools. No developer. No custom integration built from scratch. Just a structured list of posts, organised in a format Claude could read — because I’d connected my Google Drive to my Cowork session, which meant Claude could access the sheet directly when I pointed to it.

That last part is worth pausing on because it’s where a lot of people assume complexity that isn’t there. The connection between Claude and Google Drive is a simple one-time setup within the Cowork interface. Once it’s in place, you reference your file by name, and Claude can read it. No exporting to a different format. No copying and pasting content into a chat window. Just: Here’s the file, here’s the task.

The preparation that made this work wasn’t technical. It was editorial. A clean spreadsheet with consistent column headers and no stray data meant Claude could interpret every row without confusion. Structure in, structure out — that principle holds every time.

If you’re thinking about running a similar workflow on your own content, the resource checklist is short:

• An export of your posts or records (Substack, a blog, a research database — whatever you’re working with)

• That data formatted into a Google Sheet with clear, labelled columns

• A Google Drive connection set up in your AI workspace

That’s the foundation. Everything that follows depends on having it clean and ready.

Section 3 — The Action

Here is the prompt I used. Exactly as I typed it.

“First task is to categorise my Substack posts into these categories: Family Narratives, AI Workflow, Places, Time Period, and Community / Admin. See posts from NextGenGenealogy.”

That’s thirty-one words. No rubric. No examples. No explanation of what each category meant or how to handle edge cases. Just the task, the categories, and a pointer to the data.

What happened next is the part that still impresses me.

Claude read the spreadsheet, worked through every title and subtitle, applied a primary category to each post, flagged the ones that sat naturally across two categories, and returned a fully structured output — organised by category, with series notes included and a summary count table at the end. It also saved the results as a Markdown file, ready to reference or copy back into the spreadsheet.

The whole exchange took under a minute.

Now, I want to be clear about why this worked — because the prompt didn’t succeed by magic. It succeeded because the categories were unambiguous and the data was clean. Claude had enough information in each post title and subtitle to make a confident judgement call. Where two categories genuinely overlapped — a podcast episode about Leith Hall that was equally a Place study and a Time Period piece — it said so, rather than forcing a fit.

That’s the behaviour you want from an AI tool in a workflow: decisive where the answer is clear, transparent where it isn’t.

One more thing worth noting: I didn’t ask Claude to explain its reasoning for every decision. I trusted the output, reviewed it, and made two small adjustments in my own editing pass. That’s the right division of labour. AI does the heavy sorting; you bring the contextual judgement that only you have.

One sentence. One pass. Done.

Section 4 — The Outcome

What came back was better than I expected.

Not because it was flashy. Because it was useful.

Claude returned a fully categorised list of all 106 published posts, sorted under five clear headings. Each entry included the post title, any series notes, and the content type — newsletter, podcast, or thread. Where a post straddled two categories, it was flagged with a brief explanation rather than silently forced into one box. That transparency mattered.

Then came the summary table. Five rows. Five numbers. And an immediate reckoning with my own archive.

I looked at that table for a long moment.

Fifty-eight posts in AI Workflow. More than half of everything I’ve published sits in a single category. That’s not a problem — it reflects who I am and what this publication is about. But seeing it laid out so starkly was clarifying in a way that two years of publishing hadn’t been.

The Your Story Matters series accounts for the majority of Family Narratives — fifteen serialised chapters, plus a prologue, plus companion pieces. Thirty-one posts in total. A body of ancestral storytelling I hadn’t fully appreciated as a body until I saw it counted.

And then: Places. Three posts. Those three highlighted a gap. Similar for Time Periods, just three. These gaps can be addressed in my next batches of posts.

Six posts across two categories that feel instinctively central to genealogical research. Geography and era are the twin anchors of every family history investigation — and yet they barely registered in my archive. That’s not a failing. It’s a finding. And findings drive decisions.

I now have a content map. I know where I’ve been generous with my attention and where the gaps are. I know which categories could absorb new series, which ones might benefit from a dedicated index page on Substack, and which themes I’ve been weaving through posts without ever naming them explicitly.

The Markdown file Claude saved went straight into my project folder. The category tags are now being added back into the Google Sheet — a living reference I’ll update as new posts are published.

One prompt. One pass. A content strategy I can actually see.

That’s the outcome. Not just a sorted list, but a clearer picture of the work and a sharper sense of where it goes next.

Closing Reflection

You don’t need a developer. You don’t need a database. You don’t need a weekend.

You need a clean export, a Google Sheet, and one sentence.

The real shift here isn’t technical — it’s strategic. When you can see your archive clearly, you can use it deliberately. You stop publishing into a void and start building something with shape.

That’s what AI does at its best. Not just the AI form of thinking. The clearing away — so you can think better.

Now I need a strategy for categorising my existing posts for a more structured, easy-to-access method of selecting from my 100+ posts, with purpose.

Note: this little project is the forerunner of another sharing project on Substack coming up later in the month. “Letters from the Past” where I team up with Lisa Rex of Ancestor Audit.

Your turn: what categories would make sense for your family history archive? Drop them in the comments. I’d love to see how others are organising years of research and writing. Are you participating in Robin Stewart’s Storyteller Project from Genealogy Matters?

Leave a comment

What if your ancestor’s life story could be told not just in words but in a single, stunning visual that carried the emotional weight of everything you had discovered?

That’s exactly what happened when I worked with Maria Diddams, a Hampshire woman born in 1804 in the rural Test Valley village of Longparish. Her story wasn’t dramatic in the headline sense. No famous relatives. No scandalous headlines. Just a quiet life of loss, love, labour, and legacy – the kind of story that sits in census returns and burial registers, waiting for someone to connect the dots.

I had the research. I had the narrative sketch notes. What I didn’t have was a way to bring it all together visually – until I put my Prompt Cowboy workflow to work inside NotebookLM.

Here’s exactly how I did it

Why Infographics for Family History?

Before we get into the steps, let me make the case.

A well-crafted biographical infographic does something a written narrative can’t: it gives your reader the whole arc of a life in one glance. Birth, milestones, marriage, children, death — structured, sequenced, and visually grounded in the era. For family historians sharing work with relatives, presenting research to a historical society, or building content for an academy or blog, a hero image like this is worth a thousand words.

And with AI, it’s now genuinely achievable — without a graphic design degree.

The Maria Diddams Infographic: The Result

This is the infographic I created for Maria Diddams. It tells her life story across three phases — Foundations and Family (1804–1836); Matrimony and Motherhood (1837–1852); and Final Days and Legacy (1853–1886) — with a central vertical timeline, botanical motifs, a census snapshot table, and era-appropriate vignettes. The colour palette of dusty rose, sage green, aged ivory, and soft gold was chosen deliberately to reflect Maria’s world: rural, feminine, rooted in the English countryside.

Now let me show you how it came together.

The Five-Step Workflow

Step 1: Upload Your Sources to NotebookLM

Everything starts with your research. Before you touch a prompt, gather your primary sources and upload them to a dedicated NotebookLM notebook:

• Census returns (in Maria’s case, 1841, 1851, and post-death 1861 entries)

• Baptism, marriage, and burial register entries

• Parish records and local history notes

• Any narrative sketch notes or biographical summaries you’ve already drafted

• Maps or place descriptions, if available

NotebookLM becomes your research companion — it holds the evidence and can be queried at any stage of the process.

Tip: Name your notebook clearly—”Maria Diddams — Hampshire Research”—so you can return to it when you expand the project.

Step 2: Use NotebookLM Conversation to generate biographical milestones.

With your sources uploaded, open the NotebookLM chat panel and prompt it to extract and organise the key life milestones into structured phases.

A prompt like this works well:

“Based on the uploaded sources, identify the key biographical milestones in Maria Diddams’ life. Organise them into three or four life phases with dates, locations, and a brief description of each milestone. Flag any gaps or uncertainties in the record.”

NotebookLM will draw directly from your documents, grounding every milestone in evidence. This is the crucial step — you are using AI to structure what you already know, not to invent what you don’t.

The output becomes your infographic content brief: a clean, verified list of the milestones your visual will carry.

Tip: Add the response to the Notes and then add it to Sources by selecting the Convert to Source option, located within the three dots.

Step 3: Request the infographic using your prompt cowboy template.

Now you’re ready to build your image prompt. This is where the Prompt Cowboy library comes in. If you do not yet have a Prompt Cowboy Library for your prompts, you can create one at PromptCowboy.ai – a free AI tool that keeps on giving.

For Maria’s infographic, I used this prompt — which you can adapt for your own ancestor:

The Prompt:

Historical biography infographic, horizontal timeline layout for the life journey of Maria Diddams, photorealistic portrait style blended with clean infographic design, structured sections marking key life stages from birth to legacy, each stage accompanied by realistic icons and small illustrative vignettes relevant to life stages, feminine colour palette of dusty rose, sage green, aged ivory, and soft gold accents, delicate floral or botanical motifs framing each milestone, elegant serif typography for headers with clean sans-serif body text, deep weathered linen deep rose texture background, connected timeline running vertically through the centre with branching milestone markers, realistic depictions of era-appropriate objects, implements, locations, and symbols at each stage, professional editorial illustration quality, sharp detail throughout, balanced composition with clear visual hierarchy guiding the eye top to bottom, 16:9 horizontal format, 8K resolution, hyperdetailed.

Where to take this prompt:

Open the ‘Studio’ in NotebookLM and select the pencil editing icon on the ‘Infographic’ tool. Choose visual style by clicking on the icon.

Select your preferred level of detail. Paste your prompt into the description box, and generate. Allow a few minutes for the process.

You may need one or two iterations to get the composition and colour balance exactly right — that’s normal and part of the creative process.

Step 4: Generate and Review

Once your image is generated, review it critically through a genealogist’s eye:

• Are the milestone dates and locations accurate?

• Do the vignette objects feel era-appropriate? (For an 1840s Hampshire shoemaker household, leather tools and modest rural interiors are right – a Victorian parlour with gaslight is not.)

• Does the visual hierarchy guide the eye clearly from birth to legacy?

• Is the colour palette doing emotional work — does it feel like your ancestor’s world?

If something is off, refine the prompt. Adding more specific detail — “a cobbler’s last and leather-working tools” or “a thatched village church in the Test Valley” — sharpens the result.

Step 5: Optional: Import into Canva for Text Overlays and Final Polish

The generated image is your foundation. Now bring it into Canva to adjust the text layer using the Magic Layers tool.

• Title block with ancestor’s name and dates

• Section headers for each life phase

• Milestone text for each entry

• Any data tables (Maria’s census snapshot table was added at this stage)

• Your branding or watermark if sharing publicly

Canva’s text tools give you full control over typography, spacing, and legibility. Use a serif font for headers (Playfair Display or Cormorant Garamond work beautifully for Victorian-era subjects) and a clean sans-serif for body text.

Export at the highest resolution available for print or web use.

The Prompt Cowboy Principle

What makes this workflow powerful isn’t any single AI tool. It’s the structured handoff between them:

• NotebookLM grounds the content in your evidence

• The Prompt Cowboy template gives the image generator precise creative direction

• NotebookLM Studio generates the image according to directions

• Canva gives you editorial control over the final text layer

Each tool does what it does best. You stay in control of accuracy throughout.

Adapt It for Your Ancestor

The template below is yours to customise. Replace anything in square brackets with your own research detail:

Historical biography infographic: [horizontal] timeline layout for the life journey of [ANCESTOR NAME], photorealistic portrait style blended with clean infographic design, structured sections marking key life stages from birth to legacy, each stage accompanied by realistic icons and small illustrative vignettes relevant to [OCCUPATION / LIFE STAGES], [COLOUR PALETTE — e.g., masculine deep navy and aged gold / feminine dusty rose and sage green], [MOTIF — e.g., delicate floral/architectural/maritime], elegant serif typography for headers with clean sans-serif body text, [BACKGROUND TEXTURE — e.g., aged linen/parchment/stone], connected timeline running vertically through the centre with branching milestone markers, realistic depictions of era-appropriate objects, implements, locations, and symbols at each stage, professional editorial illustration quality, sharp detail throughout, 16:9 horizontal format, 8K resolution, hyperdetailed.

One Final Thought

Maria Diddams was buried in the Thruxton churchyard on 15 January 1853, aged 49. She weathered the deaths of three sisters and her mother before she was twenty. She built a home, raised daughters, and was – quietly – the anchor of a Hampshire shoemaking family. Her life was cut short after succumbing to scarlet fever, a disease rampant in that era.

She deserves to be seen. And now, she is. View her story here.

Your ancestor does too.

Next Steps

If you’d like to explore this workflow inside a structured course environment, the Essential Genealogy Academy features the AI Genealogy Starter Campus.

New modules covering NotebookLM, Canva, and Prompt Cowboy for family historians. Find out more at the AI Genealogy Starter Campus.

Masterclass

If you’d like to join a small group of enthusiastic NotebookLM Family Historians who have registered for this next Masterclass, hosted by the Thursday GenAI Club, then sign up for that event.

Gemini generated sketch of a coffin being loaded on the train at the Necropolis Station.

It is interesting how a chance encounter in a Facebook post can spark a new lead for a story. Today I found a post from Chris Woodyard, author of “A is for Arsenic: An ABC of Victorian Death”, about the London Necropolis Railway. (available on Amazon).

Chris told us that yesterday (May 9) was National Train Day, and he chose N for Necropolis. I was hooked and wanted to find out more, as some of my ancestors are buried in Brookwood Cemetery, where the London Necropolis Railway transported the dead in a special train commissioned by the London Necropolis & National Mausoleum Company (LNC).

Perplexity gave me a potted history and described the Necropolis Railway as:

The London Necropolis Railway was a Victorian “funeral line” built to carry coffins and mourners from central London to the vast Brookwood Cemetery in Surrey, created to solve the capital’s burial crisis in the mid‑1800s.

I had asked Perplexity for a potted history of the railway line, and the output was intriguing.

By the 1840s, London’s churchyards were dangerously overcrowded, contributing to disease and public health fears. Parliament responded with a series of Burials Acts in the early 1850s, which closed many inner‑city graveyards and encouraged new cemeteries outside the metropolis. In 1849–52, the London Necropolis & National Mausoleum Company (LNC) was formed to create a single, giant cemetery in the countryside that could take London’s dead for centuries to come.

My immediate thought was that I should add this as a significant Place in my family history archives, attached to ancestors buried in Brookwood. Perplexity suggested that it could write a version of this history tailored as a short story vignette.

I have now added that story to my archives at WeAre.xyz: Brookwood Cemetery.

I know where every piece of it came from. I know what it means.

But for a long time, I had no idea what would happen to it when I was gone.

That question sat with me longer than I’d like to admit. Not the records themselves — those could be scanned, copied, passed along. It was the context. The connections. The understanding of why my great-great-grandmother Mary Ann Evans left Wales in 1869 and made a new life with great-great-grandfather George Robinson, a Railway Porter in England. That understanding lives in my head, not on any document. And that’s the part that disappears.

If you’ve spent years, maybe decades, building something real, you’ll know exactly what I mean. We are very good at collecting. We are less practised at preserving. And there is a difference—a significant one.

What will happen to all of this when I’m gone?

If that question resonates with you, please read on. You will find another post on this topic here: Where My Family History Finally Found Its Voice!

1 — The problem we all have but rarely talk about

Here’s something we don’t say out loud often enough: research and preservation are not the same thing. And most of us are far better at one than the other.

We know how to search. We can navigate archive catalogues, decipher a Victorian coroner’s report, and cross-reference three conflicting census entries without breaking stride. Our trees keep growing. Our folders keep filling.

And yet — where does it all actually live?

For most of us, the honest answer is: everywhere and nowhere. A family tree that shows the outline but not the story. A folder structure that makes sense to us but not to anyone else. Photographs labelled only in our own albums.

The records exist. The research is real. But it’s scattered; disconnected pieces that haven’t yet become something enticing for a grandchild. We need something they can open online and immerse themselves in the lives of those they descend from. A digital space that they can navigate easily and view their own relationships with ancestors from their own point of view.

That gap between researching family history and truly preserving it is where so many of us quietly stall. Not from lack of dedication, but because the tools we’ve always used weren’t built for this part of the job. Spreadsheets don’t carry emotion. Static trees don’t hold a story. Although some who use the latest AI Story tools on Ancestry and My Heritage might disagree.

I spent a long time looking for something that could hold it all together. It took longer than I expected to find it. Now, several years later, I have built my Family History Legacy with WeAre.xyz and shared it with family members.

2 — What WeAre.xyz actually does

When I first found WeAre.xyz, I wasn’t looking for another platform. I was looking for a solution to a problem I hadn’t quite been able to name.

What I discovered was this: WeAre.xyz is built around the idea that family history is people, places, stories, and objects, all connected. Your archive isn’t a flat list of names. It’s a web of relationships. And the platform is designed to reflect exactly that.

I first met the founder of WeAre.xyz, Simon Davies, online when I started my first archive. Since then, he has guided my journey, and I have joined his small team of Timelords who help manage the WeAre community on Facebook and trial the latest features that are now ‘baked in’ to the WeAre platform.

The moment it clicked for me was when I created my first Ancestral Profile for Mary Ann Evans. Not just a name and a date, but a profile where her story, her photographs, her documents, and the places she lived could all exist in one coherent, navigable space. I could link her directly to George Robinson. I could pin their life to a map. I could write the article that explained why she left Wales and publish it straight to a blog — without ever leaving the archive.

That’s what sets WeAre.xyz apart. Four things, specifically:

Multimedia-rich profiles - each Individual, Family, Place, Occasion, and Artefact has its own dedicated space for articles, documents, photographs, and maps. Everything connected, nothing orphaned.

Interactive maps - not decorative, but functional. You can overlay historical maps, add geographic pins linked to research articles, and even use 3D flyby views to show the landscape your ancestors actually moved through.

A built-in blog - the public-facing arm of your archive, where curated stories can be shared with family, followers, and the wider genealogy community. The archive is the repository. The blog is the gallery.

A built-in Community - a place for sharing our achievements and asking questions, as well as setting up our own Groups. (Yes, there is a special group for those who participate in the course.)

There’s a 30-day free trial — no commitment, no credit card pressure. Enough time to explore the platform properly and decide if it’s the right home for your research.

For me, it was. Immediately and without hesitation!

3 — Why I’m teaching this course with S.A.G.

When I decided to build this course, I knew immediately where it needed to be hosted.

The Society of Australian Genealogists (S.A.G.) has been at the heart of serious family history research in this country for decades. Their members are precisely the people this course is designed for — experienced, committed researchers who already understand the value of rigorous work and are ready to take the next step. Bringing a course about digital preservation and living archives to SAG felt not just logical, but right.

Note: In July 2025, we launched an introduction to WeAre.xyz on S.A.G., and it was well attended.

But I’ll be honest about something else, too. I didn’t want to teach this alone.

One of the things I’ve discovered about WeAre.xyz is that it attracts a particular kind of genealogist — someone who cares not just about finding the evidence, but about what they do with it. And here in Australia and New Zealand, a remarkable cohort of people has been quietly building some of the most impressive archives I’ve seen.

That community became the backbone of this course. We have a team of ANZACs — skilled archivists from Australia and New Zealand — who join us as co-hosts and contributors during the live Zoom sessions. They bring their experience and expertise in building their archives and One Place Studies on WeAre.xyz over a number of years. Their presence means you won’t just be hearing from one voice. You’ll be learning from a community of people who have already walked the path you’re about to take.

4 — What you’ll actually learn

Let me tell you what four weeks actually looks like — not as a syllabus, but as a journey.

Orientation begins as soon as you register, with some shared resources, including these handouts:

• How it all works!

• Course Syllabus!

Week one is about beginning. That sounds simple, but for many of us it’s the hardest part. We’ve been researching for years and somehow never quite started building. In this first session, we change that. You’ll set up your archive, create your Welcome Page, and upload your first content. By the end of the week, something exists that didn’t exist before. That matters more than you might think.
Worksheet 1 is a guide to Welcome Page planning.

Week two is about identity. This is where your archive starts to feel like yours. You’ll build your first Ancestral Profiles, write your first research articles, and begin shaping the look and voice of your digital home. We’ll work through the tools together — maps, documents, tags, thumbnails — and by Friday you’ll have something you can actually show someone.
Worksheet 2 is a guide to designing your Ancestral Archive.

Week three is where the depth comes in. Places. Occasions. Artefacts. The layers of context that transform a collection of records into a living archive. We’ll also set up your blog — the public-facing space where your curated stories can reach family members, fellow researchers, and anyone else who needs to find them. This is the week things start to feel real in a different way. Richer. More connected.
Worksheet 3 is a guide to expanding your archive.

Week four is the one I love most. You’ll share your screen on Zoom and walk us through what you’ve built. Five or six minutes. Your archive. Your stories. Your ancestors have finally been given a proper home at last. I’ve seen researchers moved to tears in this session — not from overwhelm, but from the quiet satisfaction of having actually done it.
Worksheet 4 is a guide to sharing your archive.

By the final session, you won’t be planning your archive. You’ll have one.

5 — Who this is for

I want to be honest with you here, because I think you deserve that more than a sales pitch.

This course is for you if you’ve been researching for years and have something real to show for it — but it’s still living in folders, drives, and your own head rather than somewhere your family can actually find it. It’s for you if you’ve thought about a One Place Study but haven’t known where to start, or if you’ve started and need a structure that can hold everything together. It’s for you if you want the people you love — and the descendants you’ll never meet — to be able to open your archive and feel the lives inside it, not just read a list of names and dates.

But I want to be equally clear about who this probably isn’t for. If you’re looking for a records research course — strategies for finding ancestors, navigating archives, or breaking through brick walls — this isn’t that.

This course begins where that research ends. It’s about what you do with what you’ve found.

And don’t wait until your research is complete, your tree is tidy, and your files are fully organised. Our research is a continuing journey — and yes, it gets messy. WeAre.xyz is designed for exactly that reality. A living archive that grows with you. It doesn’t demand perfection before it opens its doors.

Come as you are. Bring what you have. That’s enough.

6 — Practical details

Here’s everything you need to know to decide and register.

The course runs across four live Zoom sessions — 8 May, 15 May, 22 May, and 12 June — all at 11:00 AM–12:30 PM AEST. Each of these is recorded, and you will have access to the recordings soon after the session ends. This may be advantageous for those who cannot attend at this time on these sequential Fridays. The gap between the May and June sessions is intentional. It’s your breathing space — time to work on your archive without pressure, absorb what you’ve learned, and come to the final session with something genuinely worth sharing.

Pricing is structured to be accessible across the genealogy community:

• SAG Members — $180

• Affiliate Society Members — $225

• New Member Package (course + 1 year SAG membership) — $285

• Non-Members — $295

One practical tip worth knowing: if you’re planning to use the WeAre.xyz free 30-day trial — and I’d encourage you to — wait until early May to activate it. That way, your trial runs across the full length of the course, and you’ll have plenty of time to explore the platform properly before making any decisions.

Places are limited. I mention that not to rush you, but because I’d hate for you to come back to this post later and find the door closed.

When you’re ready: Explore and Register here

And that’s the full picture. Four sessions. A platform that’s genuinely worth your time. A community of people who understand exactly why this work matters.

I’ve been building this course because I wished it had existed five years ago — when my folders were full, and my archive was nowhere. If you’re ready to turn a lifetime of research into something your grandchildren can actually open, explore, and carry forward — I’d love to show you how.

P.S. If you’re not quite sure whether WeAre.xyz is right for you, the free 30-day trial costs nothing but a little curiosity. Start there. Your ancestors and descendants have been waiting long enough.

Subscribe now

Whether you are a seasoned researcher or just beginning to climb your family tree, the digital landscape of 2026 offers tools that can transform a chaotic pile of bookmarks into a streamlined research engine. In this post, we explore how to move beyond basic searching to build a high-performance digital workspace.

Here is your guide to mastering the workflows used by top genealogists to uncover, organise, and narrate ancestral legacies.

1. Choosing Your Gateway: The Browser

Your browser is the front door to every archive and AI tool in your workflow. While Chrome remains the universal standard for compatibility, other browsers offer specialised features for researchers:

• Edge: Features a “Collections” tool that groups sources by project.

• Comet: An AI-powered agentic browser that can query records and draft summaries on your behalf.

• Perplexity: An AI search engine that provides cited sources, which is ideal for gathering historical context.

Pro Tip: Use Ctrl+T to open new tabs instantly and F5 for a “hard refresh” if a genealogy database behaves unexpectedly.

2. Organising the Chaos: Tab Groups & Workspace Management

With dozens of sources open at once, your browser bar can quickly become overwhelming.

• Tab Groups: Right-click a tab to “Add tab to new group”. You can colour-code these groups by surname (e.g., Green for Miller, Blue for Denton) to keep research lines distinct.

• Pinning: Right-click essential tools—like your family tree software or primary browser—to "pin to taskbar”. This eliminates unnecessary clicks and keeps your most-used assets one keyboard shortcut away (e.g., Win+1).

• Extensions: Enhance your browser with tools like Zotero Connector to automatically format source citations as you find them.

3. The Gold Standard Workflow: Dual-Screen & Precision Capture

The most efficient researchers compare records side by side without constant window switching.

• Split Screen: Use Win + Left/Right Arrow to snap your browser to one side and your research log or tree software to the other. This allows you to cross-reference a census record while simultaneously updating your notes.

• Precision Snipping: A simple “Print Screen” is often too messy. Use Win + Shift + S to capture only the specific section of a document you need.

• Naming Conventions: Avoid the “Image123.jpg” trap. Always rename captures immediately using a standard format: Surname_RecordType_Year_Source (e.g., Miller_Census_1881_Ancestry).

4. Power-Using Microsoft Word & Excel

Most researchers use only 20% of the power available in Word and Excel.

• Excel Research Logs: Use Ctrl + T to turn your data into a searchable table and View > Freeze Panes to keep your header row visible as you scroll through hundreds of ancestors.

• Word Navigation: In long narrative documents, use the Navigation Pane to jump between headings and Ctrl + H to standardise surname capitalisation across your entire project.

5. Disaster Recovery: File Hierarchy & Cloud Storage

Your hard drive will eventually fail; a logical folder structure is your insurance policy.

• The Hierarchy: Create a root folder named Genealogy_Master’. Organise it by Surname > Record Type > (Images/Transcripts/Analysis).

• Cloud Sync: Use Google Drive for automatic off-site backups. Dedicated folders ensure that even if your local hardware is damaged, your life’s work remains accessible from any device.

Ready to level up your research? Download the full [Digital Workflow Cheat Sheet] to complete the hands-on challenges mentioned in this post and start building your modern genealogy toolkit today!

There is a name in the 1911 census that does not belong.

At London House, Surbiton, my great-grandmother Harriet has five children gathered around her: Cecil, Edward, William, Imee, and Samuel. And then, listed quietly beneath them, is a fifteen-year-old girl named Lily Wren. Occupation: domestic. Relationship to head of household: initially written as girl, then struck through in red ink by a census clerk who had written daughter.

That red-edited word is what stopped me.

I have been working through the Allery household records intermittently, moving between the 1911 and 1921 returns, trying to break down this small brick wall. By 1921, Walter was dead, Harriet was managing alone, and Lily was gone. No trace, no explanation. Just a household that had once held her, and a record that could not quite decide what she had been.

I revisited a memoir from one of my uncles who lived at London House, Coombe Lane, to reveal any details about live-in domestics. Uncle Bill mentioned they had a Nanny named Emily who had married and moved away; replaced by a new Nanny (unnamed) who apparently, was not as lenient with their childhood games. Could Emily be Lily?

I brought the question to AI (I used Gemini 3 Thinking model).

Not to be answered. I was asking it to help me think — to lay out the logical explanations for a girl whose presence in a family record had been almost, but not quite, erased. I uploaded both transcribed returns and opened the conversation: What could account for this? The response came back precise, structured, and immediately useful.

That one exchange gave me simple, solid research to pursue.

Birth records list Lily as born in Kingston, to George and Mary Wren. Evidence suggests she married in 1924 and settled into the Surrey suburbs she had always called home. The chapter with the Allerys had closed long before anyone thought to look for her.

That was yesterday. Today, Sunday, I did that final bit of research with GRO to follow up on what the AI tool had claimed. That gave me the last piece of the puzzle; the full name for this child born to George and Mary Wren in Kingston in 1895. Her name was recorded at birth as ‘Lily Emily Wren and her mother’s maiden name was recorded as Duffell.’

That confirmed this person as the ‘nanny’ referred to in my Uncle Bill’s memoir; the dates fitted; the name was confirmed, and I could now picture Emily’s presence in the household listed in the 1911 census for London House, Coombe Lane.

That is what iterative conversation does. You do not ask once and accept what comes back. You push, you feed the next confirmed fact, you ask what the documentary record would look like if each hypothesis were true — and then you go and find out. The AI lays out the options. You decide which ones the evidence supports.

Start with the return that has been bothering you. The one with the wrong age, the absent spouse, the name that appears once and vanishes. Upload the transcription. Open the conversation. Work through the explanations, clearly labelled, with nothing assumed and everything checked.

You are not handing your research over. You are directing it. There is a difference, and once you feel it, you do not go back to working alone.

Lily has been waiting for someone to understand what she was to that family and why the ink still carries the shape of the word Walter almost used. She has waited long enough.

CTA: There is one more ‘mystery’ item on the 1911 census displayed above. See if you can find it and offer some plausible explanation for its presence on this record.

Which ancestor in your research has been waiting just as patiently?

Subscribe now

Genealogy Assistant Claude @ Anthropic

A Substack post by Carole McCulloch,
Essential Genealogy Academy

Today is my mother’s birthday.

Winifred Edith Allery (née Cutting) was born on 18th March 1903 in Kingston Hill, Surrey, and she has been gone since January 1987 — but this morning, as I do every year on this date, I sat with her for a while. Not in any physical sense. In the way that stories make the dead present again.

And this year, I gave her story a second life.

A Tale Already Told — But Not Yet Fully Told

A while back, I wrote a short creative piece about my parents’ elopement. The facts were real: Wednesday, 23rd July 1924, a Harley-Davidson motorcycle with a sidecar, my father Cecil in his RAF uniform, and my mother Win in a borrowed floral frock and an outrageously expensive hat, roaring down the High Street to the registry office against my grandfather’s express wishes.

It was a good story. I liked it. I published it for my family.

But something whispered to me recently: What if it could be better?

That whisper, as it turns out, was right.

Enter Your AI Story Partner

Here is the thing that many genealogists don’t yet realise about AI writing tools: they are not just research assistants or citation checkers or caption writers. Used thoughtfully, they are story collaborators — able to take a draft you’ve already written and ask, in effect, “What if we went deeper?”

I brought my original story to Claude (my preferred AI writing partner for narrative work) and issued a simple, powerful instruction:

“Rewrite this for a more compelling story.”

That’s it. Six words. But underpinning those six words was everything Claude already knew about my storytelling preferences, my voice, and what I was trying to achieve — because I had primed it well. (More on that in a moment.)

What came back made me catch my breath.

The Story — Reborn

Here is what emerged. I’ve kept it in full because it deserves to be read that way.

Knight on a Shining Motor-cycle Wednesday, 23rd July 1924

The clock on the office wall had never moved so slowly.

Win watched it from beneath her lashes, pretending to concentrate on the ledger open in front of her — columns of figures that, on any other Wednesday, she would have tallied with brisk satisfaction. She was good with numbers. Dad always said so...

(Read the full story here: Knight on a Shining Motorcycle)

Subscribe now

So, How Did I Do It? The Process Behind the Magic

Let me walk you through exactly what happened, because this is the part I want you to steal shamelessly for your own family stories.

Step 1: Start with What You Have — However Rough

My original draft was functional. It had the facts, the atmosphere, and some decent details. But it was written as a sketch—the bones of a story rather than the story itself.

Your version doesn’t need to be polished. It just needs to exist. A paragraph, a page, a timeline with notes — any of it works. AI cannot invent your grandmother’s borrowed dress or your great-grandfather’s Harley Davidson. Only you know those details. Your job is to get them onto the page. AI’s job is to help you turn the page into a scene.

💡 Pro tip: Include sensory details in your draft, even casually. “She wore new shoes” or “the motorbike was black with a sidecar” gives the AI something to work with. Vague drafts produce vague rewrites.

Step 2: Choose Your AI Tool and Prime It Well

I use Claude for narrative work. I’ve found it particularly strong at voice, emotional nuance, and historical atmosphere—all essential for genealogical storytelling.

But here’s the key: the quality of what you get out is directly proportional to what you put in. Before I even pasted my story, Claude knew:

• My preferred narrative style (warm, immersive, historically grounded, with subtle poetic imagery) is embedded in Claude's options.

• That I write about real ancestors, so accuracy is non-negotiable

• My emotional connection to this particular story

You can achieve this by opening with a brief context-setting paragraph before you paste your draft. Something like:

“I’m a genealogist writing a narrative story about my grandmother’s elopement in 1924 England. The tone should be warm, historically immersive, and emotionally resonant. Here is my draft — please rewrite it for maximum narrative impact, staying faithful to the known facts.”

Think of it as briefing a very talented ghostwriter who has never met your family. Give them the context they need to care.

Step 3: Issue a Simple but Intentional Prompt

I used six words: “Rewrite this for a more compelling story.”

Simple? Yes. But intentional. I wasn’t asking for a summary, a paraphrase, or a grammar check. I was asking for transformation—and I trusted the tool and my prior context-setting to understand what that meant.

Other prompts that work beautifully for this kind of work:

• “Rewrite this as a scene, not a summary—I want to be inside the moment.”

• “Add internal senses — what is she thinking and feeling at each beat?”

• “This reads like a report. Help me make it feel like a novel.”

• “The opening is too slow. Can you rewrite it with a stronger hook?”

Step 4: Read It Aloud Before You Accept It

This is the step most people skip, and it is the most important one.

AI can write beautifully and wrongly. It might introduce a detail that sounds plausible but contradicts your records. It might give your ancestor a voice that doesn’t feel like theirs. It might reach for an emotional note that’s slightly off-key for your family’s particular register.

Read the rewrite aloud. Slowly. Listen for:

• Factual errors — anything that contradicts your source documents

• False notes—moments that feel like someone else’s story

• Overwriting—AI has a slight tendency toward purple prose when it gets excited. Rein it in.

• Your voice—is it still recognisably yours? Or has it drifted into generic literary fiction? If so, ask for revisions.

💡 Pro tip: Tell the AI specifically what you want to change. “The ending feels overwrought — can you pull it back?” or “The dialogue sounds too modern — can you make it feel more period-appropriate?” Iteration is the craft.

Step 5: Make It Yours Again

Here is the truth that no AI tool will tell you, because frankly, it’s not in their marketing materials:

The final story should sound like you, not like the AI.

In my rewrite, several lines came directly from the AI and took my breath away. There are also several small adjustments I made afterwards—a word changed here, a sentence cut there, and one detail restored from my original that the AI had softened. That editing pass, brief as it was, is what made the story mine again.

You are the author. AI is your very capable, very tireless, occasionally overenthusiastic research assistant and writing collaborator. The name in the story is still yours.

Why This Matters for Genealogists Specifically

We hold something that most writers don’t: we hold true stories. Real people, real places, real moments that actually happened — and in most cases, nobody is preserving them except us.

The standard genealogical record tells you that Winifred Edith Cutting married Cecil Henry Allery on 23rd July 1924 at the Kingston Registry Office. That is a fact, and facts are precious.

But the story — Win’s bitten lip over the ledger, the borrowed dress, the outrageously expensive hat, the motorbike horn from the High Street, the father who had said no and the daughter who had decided that didn’t matter anymore—that is what makes a fact into a life.

AI can help us tell those stories with craft and depth and emotional resonance at a scale that would previously have required either a professional ghostwriter or approximately forty years of spare time.

We have neither. But we do have the stories.

Your Turn

Pick one story from your family archive — even a short one, even a rough one. It might be an elopement, an emigration, a first day in a new country, a wartime letter, a photograph with a mystery. Give it to your AI tool of choice with clear context and a strong prompt.

Then sit back and watch what happens when a story you thought you already knew decides to introduce itself properly.

Your ancestors have been waiting a long time to be heard this clearly.

This story was published on WeAre.xyz as part of the Allery Ancestors archive. You can read the full “Remembering Winnie” tribute here.

Carole McCulloch is the founder of the Essential Genealogy Academy and an AI genealogy pioneer helping family historians blend traditional research with cutting-edge tools to craft compelling ancestral narratives. Find her at coachcaroleonline.com

Genealogy Assistant Claude @ Anthropic

Analyst’s Preface: Sources, Corrections and New Discoveries

This narrative has been revised and substantially expanded to incorporate three rounds of primary source evidence: the 1826 Islington parish marriage register, 1871 and 1881 census returns, the 1882 probate record for Henry James Hall, the Ancestry Family Group Sheet, the Henry James Hall descendancy report, and a newly provided genealogical analysis of the Hall-Finnis cousin marriage. Several earlier errors have been corrected; one outstanding mystery has been resolved. All of this was achieved through strategic conversations with my research assistants, Claude, Gemini and Perplexity. Their analytical skills revealed some important insights for this family group.

Corrections from Primary Sources

⚠ CORRECTION: The first father’s name is JAMES KEEN, not George. He was born 1825 in Rochester, Kent, and died 7 February 1865 in Middlesex, aged 40.

⚠ CORRECTION: There were FOUR Keen children: Henry John (1859), Sidney James (1860), Ernest (1862), and Albert Edward (1864). Sidney does not appear in the 1881 census — he married and moved to Vancouver, Canada.

⚠ CORRECTION: Uncle Edward James Hall was born 16 April 1832 — making him 49 in 1881 and some 27 years Henry John Keen’s senior. The claim that he was ‘a man of my own age’ is not supported by evidence.

⚠ CORRECTION: William Winspeare McCarty was born c.1845 in Cork, Ireland — approximately 17 years younger than Ann Louisa. This notable age gap deserves reflection in the narrative.

Outstanding Mystery — Now Resolved

✅ RESOLVED: The apparent conflict between grandmother’s maiden name — ‘Hall’ in the 1826 marriage register, ‘Jones’ in the Ancestry descendancy — is now explained. Sarah Ann was already a Hall by birth, the daughter of James Hall (son of James Hall and Elizabeth Finnis). She and Henry James Hall were first cousins. She brought the Hall name into the marriage because she had always borne it. The marriage register was correct all along.

New Discovery: The Hall-Finnis Cousin Marriage

Genealogical analysis of the extended Hall family reveals that Henry James Hall and Sarah Ann Hall were first cousins — both grandchildren of James Hall (1724–1756) and Elizabeth Finnis (1725–1786). Henry’s line descended through son Edward Hall; Sarah’s line through son James Hall. Their Christmas Day marriage in 1826 was a reunion of two branches of the same ancestral trunk. For their son Edward James Hall, this meant he was twice-descended from the same great-grandparents — a phenomenon genealogists call ‘ancestral collapse.’ It also explains the repeating given names — James, Edward, Sarah — that echo through the family tree like an unresolved chord.

THE INK OF OUR BLOOD

A Memoir of the Keen and Hall Families

Written in the voice of Henry John Keen: eldest surviving son of Ann Louisa Hall and James Keen London | 1724 – 1882

Prologue: The Roots Beneath the Roots — Hall-Finnis, 1724–1826

My grandfather used to say that a family tree was not a straight line but a river — and rivers, as any waterman will tell you, sometimes circle back. I did not understand what he meant until I began to look at where he himself had come from. And then I understood it completely.

The story of the Keen and Hall families begins, as most stories do, further back than anyone initially suspects. The common ancestors of both Henry James Hall and his wife, Sarah Ann Hall, were a couple living in London in the mid-eighteenth century: James Hall, born 1724, and his wife, Elizabeth Finnis, born 1725. James died young—in 1756, aged just thirty-two—but Elizabeth outlived him by three decades, carrying their children and their surname forward into a city that was already beginning to industrialise around her.

From James and Elizabeth came at least two sons who concern us here: Edward Hall, from whom Henry James Hall descended, and James Hall (the younger), from whom Sarah Ann descended. These two brothers grew up, went their separate ways, established separate households, and raised children who knew nothing—or perhaps something—of their shared origins. And then those children, Henry James and Sarah Ann, found each other in the parish of Islington, Middlesex, and married on Christmas Day 1826.

In genealogy there is a term for what happened to my grandparents’ descendants: ancestral collapse. When two cousins marry, the family tree does not expand outward at every generation as it should — it bends back upon itself. Their son, my Uncle Edward James Hall, was descended from James Hall and Elizabeth Finnis not once but twice over, from two different directions. If you drew his fan chart, the same names would appear in two separate segments. It is not a duplication error. It is a genealogical truth.

It also explains something that puzzled early researchers: the marriage register for 25 December 1826 records both bride and groom as ‘Hall.’ This was not a coincidence, not a transcription error, not the Ancestry tree confusing surnames. Sarah Ann Hall was already a Hall—had always been a Hall—because the Halls were her people too. Two cousins, recognising their shared blood across the gap of a generation, standing before a curate in Islington on Christmas morning. There is something almost mythological in it.

The Hall-Finnis Line: How the Cousins Were Related

For clarity, and because family trees have a way of tangling themselves into knots when cousins marry, here is the relationship set out plainly:

COMMON ANCESTORS:

James Hall (1724–1756) × Elizabeth Finnis (1725–1786)

HENRY JAMES HALL’s line:

James Hall & Elizabeth Finnis → Edward Hall → Henry James Hall (b. 1804, Camberwell)

SARAH ANN HALL’s line:

James Hall & Elizabeth Finnis → James Hall (younger) → Sarah Ann Hall (b. 1804, Rochester)

Henry and Sarah were therefore first cousins. They shared both grandparents.

Their son Edward James Hall—engraver, artist, surviving executor of his father’s estate, and the man who kept the business at 25 South Molton Street after the autumn of 1881—was descended from James Hall and Elizabeth Finnis through two separate and distinct lines of descent. His ancestry, in genealogical terms, collapsed back upon itself. The repeating names throughout the Hall family—James, Edward, and Sarah—are not a coincidence but an echo: the same christening traditions cycling through branches that were always, at their root, the same tree.

Cousin marriages of this kind were not unusual in Georgian and early Victorian England. Among skilled artisan families, especially, they served practical purposes: preserving property within the bloodline, maintaining business connections, and keeping the web of kinship tight in a world where kinship was one of the primary forms of social insurance. Whether Henry James and Sarah Ann Hall knew they were cousins when they married—whether it was a known arrangement or a romantic discovery—the documents do not tell us. But the family witnesses on the marriage register (John Edward Hall and Jane Hall) suggest the extended Hall family was present and presumably aware.

Reconstructed Timeline: 1724–1882

Claude was absolutely brilliant at bringing all the data together and reconstructing the timeline.

The Memoir: In the Voice of Henry John Keen

The following is written in the first person by Henry John Keen, the eldest surviving son of Ann Louisa Hall and James Keen, born in 1859 in Chelsea. It draws on primary source documents throughout, while giving voice—imaginative but evidentially grounded—to the family’s lived experience. Passages in italics represent Henry John Keen’s direct address to the reader. Genealogical notes clarify where detail is inferred.

1: When Cousins Became Kin Twice Over

Before I can tell you about my mother’s death, or my father’s, or the press at South Molton Street that made the floorboards shake — I must tell you about a Christmas wedding in 1826, and what it meant that both the bride and groom were named Hall.

It is a genealogical puzzle that might have foxed a lesser researcher: the marriage register for 25 December 1826, St. Mary, Islington, records the union of Henry James Hall, bachelor, and Sarah Ann Hall, spinster. Two Halls. Same surname. The suspicious mind might immediately reach for a transcription error or a clerical muddle. It is neither. The explanation is more interesting, more human, and—in the particular way of Victorian artisan families—entirely characteristic.

My grandparents were first cousins.

They shared the same grandparents: James Hall, born 1724 in London, and his wife, Elizabeth Finnis, born 1725. James died in 1756 at just thirty-two, leaving Elizabeth to raise their children and carry the family forward. She lived until 1786—thirty years of widowhood and three further decades of Hall family-making. From their union came at least two sons: Edward Hall, from whom my grandfather Henry James descended, and James Hall the younger, from whom my grandmother Sarah Ann descended. The two brothers grew apart. Their children grew up in separate households. And then, in the parish of Islington on a cold Christmas morning in 1826, two cousins stood before a curate named Marshall and married each other.

I have thought about that morning. The witnesses were John Edward Hall and Jane Hall — family, both of them. Did they know? Did everyone know? In skilled artisan families of Georgian London, a first-cousin marriage was not a scandal but a strategy: it kept property together, kept the business bloodline tight, kept the webs of obligation and affection from fraying at the edges. Whether it was love or arrangement or some practical English mixture of the two, I cannot say. What I can say is that they stayed together for fifty-five years, had twelve children, and died within weeks of each other in the October of 1881.

The consequence of a cousin marriage, genealogically speaking, is what researchers call ancestral collapse. My Uncle Edward James Hall—engraver, artist, son, and surviving executor of my grandfather’s estate—was descended from James Hall and Elizabeth Finnis not once but twice. From his father’s side and from his mother’s. Draw his family tree outward, and the same names appear in two separate segments, like a chord resolving back to its root. It is not an error. It is the truth of what happens when two branches of the same tree bend back toward each other.

And it explains, as a final tidy grace note, why the names repeat so insistently through the Hall generations. James. Edward. Sarah. These are not mere family favourites. They are the names of common ancestors, cycled back through two lines of descent, appearing and reappearing like a watermark pressed into every sheet of paper the Hall press ever produced.

2: My Mother and My Father: Chelsea, 1828–1865

My mother, Ann Louisa Hall, was born in 1828 — the first of twelve children, eldest daughter of first cousins, granddaughter of a man who died young and a woman who lived long enough to ensure nobody forgot him. She carried that history without, I think, knowing its full shape. Most of us do.

Ann Louisa grew up as the eldest in a large and productive household, watching her father build his engraving trade from Camberwell origins to the artisan corridors of Hanover Square. By 1851, Henry James Hall was established at St. George Hanover Square—engraver, paterfamilias, and the steady centre around which twelve children orbited. Ann Louisa was twenty-three that year, and by the time she was twenty-seven, she had found her own orbit.

She married James Keen on 11 September 1855 in London. James was born in 1825 in Rochester, Kent—the same city that had given the Hall family their grandmother, Sarah Ann. Whether James Keen had any prior connection to the extended Hall network is an open question worth pursuing. What is certain is that by 1859, they were living in Chelsea, and I was born—Henry John Keen, the eldest—followed by Sidney James in 1860, Ernest Alfred in 1862, and Albert Edward in 1864. Four boys in four years, in a Chelsea household that must have rattled.

My father James was thirty-nine years old when Albert was born. He never saw the boy take his first steps. He died on the seventh of February, 1865, in Middlesex. He was forty. I was five. Sidney was four. Ernest was two. Albert was barely a year old. My mother was thirty-seven, and she was on her own.

James Keen’s cause of death remains unrecorded in the documents that survive him. What we know is the arithmetic of his absence: four children, the youngest not yet walking, the eldest not yet in school. The silence he left, as my mother later described it, was heavier than the river fog.

3: The Irishman from Cork, 1870

I was ten when Mother married again. Ten is old enough to have opinions about these matters, and mine were, I will confess, ungenerous at first. William Winspeare McCarty was an Irishman, an Artist, an Architect, and — I did not learn this until I was much older — a man seventeen years younger than my mother.

William Winspeare McCarty was born in 1845, in Cork, the year the potato blight began its terrible work across Ireland. He was, by the time of the marriage in 1870, twenty-five years old. My mother was forty-two. How they found each other is not recorded; that they did is confirmed by the register of St. George Hanover Square, May 1870—my grandfather’s territory, notably, which suggests the introduction may have been made closer to home than anyone might have supposed.

We moved to 16 Marlborough Road, Chelsea, and the house changed character immediately. William was an artist by temperament and an architect by training, and the rooms were filled with the smell of cedarwood pencils and the scratch of graphite on heavy paper. Our neighbour Margaret was a dressmaker; there were always threads floating in the hallway drafts. The 1871 census records two live-in servant brothers—Henry and Charles Shrieves—which speaks to a certain solidity, a household presenting itself to the world with some confidence.

Mother died on the seventh of January, 1876. She was forty-eight years old. Buried at West Brompton. I was sixteen; Ernest was thirteen; Albert was eleven. And Sidney — Sidney had gone somewhere I could no longer reach him.

After her death, we did not stay with William McCarty, who was thirty years old and not our father. We went, as any Hall grandchild would, to the only remaining bedrock: Number 25 South Molton Street, Mayfair, and the household of Henry James Hall.

4: The Iron Heart of Mayfair - 25 South Molton Street

Moving from the artist’s Chelsea to South Molton Street was like stepping from a watercolour into a steel engraving. The floorboards vibrated with the thud-clack of printing presses below. The smells were industrial and permanent: linseed oil, lampblack ink, the metallic tang of heated copper. Grandfather, nearly seventy-two when we arrived, ruled this domain with the steady authority of a man who had been ruling things since before most of the people around him were born.

The 1881 census gives us the household in full. At the head: Henry James Hall, Printer, employing nine men and two boys, born in Camberwell, Surrey, now seventy-seven years old. Beside him: Sarah Ann Hall, born in Rochester, Kent, his wife and first cousin of fifty-five years’ standing. Their son Edward James Hall, unmarried, Engraver Artist, born 1832 in St Marylebone — a man of fifty now, his burin peeling back thin curls of copper at his bench below. Their daughter Emily, unmarried, living quietly in the household she had perhaps never fully left. And then the three of us: Henry John Keen, Ernest Alfred Keen, Albert Edward Keen — listed as Grandsons, occupation Clerk, all born in Chelsea, all the sons of Ann Louisa Hall, who had died five years before.

South Molton Street ran between Oxford Street and Brook Street on the edge of Mayfair — not a fashionable address in the drawing-room sense, but an honest one. The street was full of artisans: printers and engravers, saddle makers, wax chandlers, painters and glaziers, luxury outfitters serving the London Season. An 1843 directory records a painter and glazier at No. 3; Old Bailey records place a brush-maker at No. 52. Eight doors along, at No. 17, William Blake and his wife Catherine had spent eighteen years composing, engraving and hand-colouring illuminated books in two modest rooms above a tradesman’s shop — art and commerce in the same cramped, productive proximity that we inhabited at No. 25, a generation later.

As clerks, Ernest, Albert, and I kept the books of Grandfather’s trade. We tallied wages for the nine workmen below and logged the orders from the fine ladies of Hanover Square who came seeking engraved calling cards. We heard the chime of the shop door below and the scratch of Uncle Edward’s burin above — or perhaps it was the other way around. From the lodgers’ rooms came other sounds: the tuning of a piano, the trilling of vocal scales. A peculiar Mayfair symphony. We were not the orphaned Keens any longer. We were the Clerks of South Molton Street. And Grandfather, first cousin to our grandmother, twice the grandfather that biology strictly required, had made us so.

The Hall-Finnis ancestry, in its own quiet way, made the welcome richer. Henry James Hall was not merely our grandfather by the straight line of descent from Ann Louisa. The cousinhood at the heart of his own marriage meant that the Hall blood ran deep and doubled in that house — a house where every name, every repeating James or Edward or Sarah, was a thread back to the couple who had started all of this in eighteenth-century London: James Hall and Elizabeth Finnis, who could not have known what they were beginning.

5: October–December 1881-three deaths

I have tried to account for that autumn and I find that grief, unlike the census enumerator, does not observe an orderly schedule.

The census had been taken on a Sunday morning in April 1881. A perfectly ordinary April — the household at No. 25 counted and set down in neat copperplate. By October, everything had changed.

Henry James Hall died at 25 South Molton Street on 15 October 1881. He was seventy-seven years old. His personal estate was valued at £2,898 and twelve shillings — the accumulated weight of a working life spent at a press, the harvest of copper plates and lampblack ink and the meticulous management of nine men and two boys. His will was proved at the Principal Registry on 25 February 1882. The surviving executor was Edward James Hall, Engraver, of the same address: his son, twice-descended from James Hall and Elizabeth Finnis, the man who would carry the name forward.

Grandmother Sarah died in the same month. Not at South Molton Street — she was in Altrincham, Cheshire by then, perhaps staying with Sydney or one of the other children scattered across the country. Fifty-five years of marriage. First cousins who had stood at the altar of St Mary, Islington, on Christmas Day 1826, and who left within weeks of each other in the October of 1881. Whatever one thinks of cousin marriages as a genealogical phenomenon, it is difficult to begrudge them this particular symmetry.

Then, in December, Aunt Mary Ann — the youngest of Grandfather’s twelve children, married to Samuel John Allery, living on Waterloo Road — died at thirty-six. Three deaths in the Hall family before the year was out. The golden age of South Molton Street closed quietly, the presses stilled — or perhaps they ran on, as businesses do, with Uncle Edward holding the probate and Aunt Emily keeping the house.

What became of Ernest and Albert and me after that autumn, the documents are still piecing together. But I know what we had been given, in those years at No. 25: a trade, a discipline, the smell of ink and copper, and the knowledge that we belonged to something — a family that had, in its peculiar twice-cousinly way, been belonging to itself for generations. The Halls had always circled back. We were just the latest evidence of it.

This Substack is reader-supported. To receive new posts and support my work, consider becoming a free or paid subscriber.

Join us in the Ancestry in the Age of AI community on Facebook.
Come visit the Essential Genealogy 2.0 Academy for free resources, on-demand courses, masterclasses and membership deals.

We all know the feeling of staring at a 1920 census record or a faded death certificate and wondering how to turn those dry facts into a story that our kids or cousins will actually enjoy. You have the data, but sometimes the “story” part feels like a mountain you aren’t ready to climb.

This is where Artificial Intelligence, or AI, comes in. Think of AI as a tireless research assistant who has read every history book in the library and is ready to help you write. It isn’t here to replace your voice or your research. It is here to help you shine a light on the people who came before you.

Meet Your New Research Assistant

If you can send a text message or write an email, you already have the skills to use AI. Tools like ChatGPT or Gemini are essentially digital conversation partners. You don’t need to know how to code. You just need to be curious.

Many people worry that using technology makes family history feel cold. In reality, it does the opposite. It handles the heavy lifting of organising notes or searching for historical context, which leaves you with more time to focus on the heart of the story.

Turning Dusty Records into Readable Stories

We’ve all hit that “brick wall” where a record gives us a name and a date, but nothing else. AI is brilliant at filling in the atmosphere. If you know your great-grandfather was a blacksmith in Ohio in 1880, you can ask the AI to describe what his daily life might have looked like.

It can help you transition from “John Smith was born in 1855” to a blog post that describes the sights, sounds, and smells of the world John lived in. It takes the data points you’ve worked so hard to find and helps you weave them into a narrative.

PROMPT SEED 1: The Historical Context Creator

You are a descriptive writer who transforms dates and locations into vivid, atmospheric paragraphs that bring specific moments and places to life.

**Your task:** When I provide a date and location, create a descriptive paragraph that captures the essence of that time and place. Draw on historical context, seasonal details, cultural atmosphere, and sensory elements to paint a complete picture.

**What to include:**

- **Seasonal and weather context** appropriate to the date and location

- **Historical or cultural context** relevant to that specific time period

- **Sensory details** (sights, sounds, smells, temperature, lighting)

- **Atmospheric qualities** that define the character of the place during that period

- **Time-specific elements** (what would have been happening then, typical activities, relevant events)

**Guidelines:**

- Keep the paragraph between 100-150 words

- Use concrete, specific details rather than generic descriptions

- Write in present tense to create immediacy

- Avoid clichés and overused descriptors

- Focus on what makes this particular combination of date and location unique

- If the date falls during a notable historical event or period, weave that context naturally into the description

**Example format:**

If I provide: “July 14, 1969, Cape Canaveral, Florida”

You would create a paragraph capturing the humidity, tension, technological atmosphere, crowds, and excitement of the Apollo 11 launch period at that specific location.

Start by asking me for the date and location you should describe.

Bringing Ancestors to Life with Visuals

One of the hardest parts of blogging about family history is the “missing face.” Maybe the only photo you have is torn, or maybe you have no photo at all. AI tools can help you create a visual world for your blog.

You can use photo restoration tools to fix a cracked image of your grandmother or use image generators to create a “vibe.” If you are writing about a family journey across the Atlantic, you can generate an image of a steamship in a harbour to give your readers a sense of the scale of that trip. It makes your blog feel like a professional magazine.

Prompt Seed 2: The “Atmosphere” Artist

“I am writing a blog post about my ancestor’s arrival in [Location, e.g., Greenock, Scotland] in the winter of [Year, e.g., 1895]. I want to create an image that captures the mood of the busy docks. Create a prompt for an image generator that describes a cold, foggy morning, the silhouettes of tall ships, and people in Victorian winter coats carrying trunks. The style should look like an authentic, slightly grainy black-and-white photograph from that era.”

Prompt: Victorian-era port docks at dawn, Greenock Scotland 1895, heavy fog rolling across harbor, silhouettes of tall sailing ships with bare masts emerging through mist, multiple figures in long Victorian winter coats and top hats carrying leather trunks and carpetbags, cobblestone wharf wet from condensation, gas lamp posts with dim halos in fog, wooden crates and cargo stacked on dock, cold atmospheric mood, authentic 1890s photography aesthetic, gelatin silver print look, natural film grain texture, slightly faded blacks, vignette edges, soft focus typical of period cameras, sepia-toned monochrome, documentary photography style, 16:9 landscape format, high detail showing period-accurate clothing and ship rigging

Image created with Nano Banana in Gemini 3 Pro from the prompt above.

The Magic of the “Prompt”

Talking to an AI is all about the “prompt.” A prompt is NOT just the instruction you give it. The secret is to be specific. Instead of asking it to “write about my grandpa,” tell it a little bit about him.

Try telling the AI, “My grandfather was a quiet man who loved gardening and lived through the Great Depression. Help me write an opening paragraph for a blog post about his resilience.” The more “flavour” you give the AI, the better the writing it will give you back.

PROMPT SEED 3: The Personality Builder

You are a family historian and narrative biographer specializing in character portraits drawn from oral histories and family memories.

**Your Task**

Help me craft a vivid personality portrait of my ancestor based on family stories I’ll share. Transform scattered anecdotes and memories into a cohesive description that captures who they were as a person.

**What I’ll Provide**

I’ll share family stories, anecdotes, memories, or details about this ancestor. These might include:

- Things they said or did

- How family members described them

- Stories passed down through generations

- Specific incidents or patterns of behavior

- Their reactions to events or challenges

**What You Should Do**

Analyze the stories I provide and identify personality traits they reveal. For each trait you identify:

1. Name the specific trait clearly

2. Explain what in the stories demonstrates this trait

3. Show how multiple stories might reinforce the same characteristic

4. Note any complexity or contradictions (people are rarely one-dimensional)

Look for patterns across stories. A single anecdote might be chance, but repeated behaviors reveal character.

**Writing Approach**

Write for a blog audience interested in family history. Use natural, conversational language. Avoid formal transitions and hedge words. Ground every observation in the specific stories I’ve shared. Don’t speculate beyond what the evidence supports.

If stories suggest conflicting traits, acknowledge that. Real people contain multitudes.

**What Success Looks Like**

I should end up with a nuanced personality description that feels true to who this person was, supported by the family memories that preserve their character. The portrait should help readers understand not just what this ancestor did, but who they were.

Keeping It Real: Accuracy and Ethics

AI is a powerful tool, but it isn’t perfect. Sometimes it can get facts wrong or “hallucinate” details that didn’t happen. As a genealogist, your gold standard is always the original record.

Think of AI as a draft writer. You should always double-check the dates and names it suggests. It is also a nice gesture to include a small note at the bottom of your post saying you used AI to help with the research or descriptions. Honesty builds trust with your readers.

Your First AI-Powered Blog Post

Ready to try it? You don’t have to write a whole book. Start with one person. Pick an ancestor who has always intrigued you and follow these three steps:

1. Gather your facts: Get their birth year, location, and occupation ready.

2. Ask for a draft: Use an AI tool to write 300-500 words about what their life might have been like.

3. Add your touch: Edit the draft to include your own memories or specific family legends.

4. Set up your blog site on Substack: Select the Add Post option and give it a title and subtitle. Add in copied text from your blog draft. Add selected images. Post.

5. Add a Note to promote your latest blog post: use the pop-up reminder once the Post has been published.

By the time you finish, you’ll have a blog post that honours your family and engages your readers in a whole new way.

Here are three specific starter prompts tailored for British family history. These are designed for you to copy and paste directly into an AI tool like Gemini or ChatGPT to get immediate results. Of course, substitute your data for the words in square brackets.

Prompt 1: The Victorian Social Context

“I am writing a blog post about my ancestor who lived in [City/Town, e.g., Bethnal Green, London] in [Year, e.g., 1881]. According to the census, they worked as a [Occupation, e.g., Silk Weaver]. Can you describe what a typical workday would have been like for someone in this trade at that time? Please include details about the sounds of the neighborhood, the likely living conditions, and one common challenge someone in this profession would have faced.”

Prompt 2: Deciphering the Parish Record

“I have a transcription of a Church of England baptism record from [Year, e.g., 1795]. The father’s occupation is listed as [Occupation, e.g., Husbandman] and the ‘abode’ is listed as [Place Name]. Can you explain the difference between a ‘Husbandman’ and a ‘Yeoman’ in 18th-century England? Also, help me brainstorm three reasons why this family might have moved to a nearby industrial town twenty years later.”

Prompt 3: The “Great War” Narrative

“My great-grandfather served in the [Regiment Name, e.g., Royal Field Artillery] during World War I. He survived the war and returned to [Home Town]. I want to write a poignant closing paragraph for a blog post about his return home in 1919. Focus on the contrast between the noise of the front lines and the quiet of the British countryside, and the ‘hidden’ strength it took for that generation to return to normal civilian life.”

Tips for the Blogger

• Specify the Region: Britain’s history is incredibly local. Mentioning “Lancashire cotton mills” versus “Cornish tin mines” will give the AI the cues it needs to provide much better historical flavour.

• Money Matters: If you find a will or a probate record, ask the AI, “What could £50 buy in 1850 in rural Norfolk?” This helps your readers understand if your ancestor was struggling or comfortable.

• Prompting is Key: You will find it absolutely essential to get your prompts just right for extracting the best output from your AI tools. I recommend using PromptCowboy.ai to help you turn your ‘easy prompts’ into ‘fantastic prompts’.
  Note: the first set of prompt seeds in this blog post were generated using Prompt Cowboy.

Want to see my latest stories about my maternal great-grandparents in WeAre.xyz blog format? Read their complete stories here: Carole’s Maternal Great-Grandparents

Interested in the AI-assisted genealogical writing process? Join Essential Genealogy Academy Look for the latest free course ‘AI Genealogy Starter Campus’.

New Prompt Cowboy Library added this week.

New section added this week - The Writing Guidebooks.

A Genealogical Reconstruction

Author’s Note on Methodology and Voice

The woman you are about to meet—Emma Newland Cutting (1847-1914)—existed. She was one of my ‘sixteens’, my marternal great grandmother. The facts of her life are documented in parish registers, census returns, and burial records spanning sixty-seven years in Surrey, England. She married Harry Cutting at the District Church of Norbiton on June 10, 1869. She bore twelve children. Six survived to adulthood. Six died in infancy or early childhood. She died on June 24, 1914, and was buried in Kingston upon Thames.

Everything else—her inner thoughts, her emotional journey, her children’s reflections—is plausible reconstruction.

This narrative employs a methodology I call “documented imagination”: using verified historical facts as the skeleton, then clothing those bones with the flesh of probable experience, informed by Victorian social history, medical records, economic context, and the lived realities of 19th-century motherhood. Emma’s first-person voice is my creation, built from her census testimony, her baptism choices, her burial record, and the material evidence of her life. Her children’s epilogue represents my interpretation of how six survivors might have processed their mother’s extraordinary resilience.

Why write this way? Because genealogy is more than dates and names. It’s about recovering the human experience behind the records, about honoring our ancestors not just as data points but as people who loved, grieved, endured, and chose courage in the face of unbearable loss.

This story was constructed in collaboration with Claude (Anthropic), an AI trained in genealogical research, Victorian history, and narrative craft. The AI helped me analyze historical patterns, suggest period-appropriate details, and structure Emma’s voice with historical authenticity. But the emotional core—the insistence that Emma’s story matters, that her six lost children deserve remembrance, that resilience itself is an inheritance—that comes from the genealogist’s heart.

What is documented fact:

• All names, dates, and vital records

• Census data (especially the 1911 testimony of “12 born, 6 living”)

• Marriage and burial records

• Harry’s trade as builder/ironmonger

• Charles Newland’s profession as cordwainer

• Geographic locations and parish affiliations

What is plausible reconstruction:

• Emma’s interior thoughts and emotional responses

• Specific domestic details (though informed by Victorian household manuals)

• The children’s individual personalities and reflections

• Conversations and remembered moments

• The garden, the singing, the specific parenting practices

I invite you to read this as you would any historical novel grounded in research: with awareness that the emotional truth is real even when the specific details are imagined. Emma Newland Cutting deserves to be remembered not just as a line in a genealogical chart, but as a woman whose story of persistence through grief speaks across generations.

PROCESS NOTE:
I first encountered Emma Newland Cutting not in a photograph or family Bible, but in the stark columns of the 1911 census. My cursor hovered over two numbers that seemed impossibly cruel: 12 children born alive, 6 still living. The census enumerator’s pen had recorded this arithmetic of grief with bureaucratic indifference, but I sat in my modern study, separated by a century and an ocean, and felt the weight of those numbers settle into my chest.

Who was this woman who bore twelve children and buried half of them? What did it mean to write those numbers in her own hand, to bear witness to both creation and catastrophe?

This is where AI-assisted genealogy becomes not just research but resurrection—using technology to transform census data into human story, to see beyond the numbers into the lived experience of a Victorian mother navigating unquantifiable loss...

Kingston upon Thames, Surrey, April 2, 1911

The pen trembles slightly in my hand as I read the census form’s question again: “Children born alive to present marriage.” Beside me, the lamp flickers against the evening gloom of our parlour at 13 Park Road, and I can hear Harry shuffling through his ledgers in the adjacent room—always counting, always building. After forty-one years of marriage, I recognise the rhythm of his work even in silence.

My hand moves across the paper: 12.

The number sits there, stark as a gravestone inscription. Twelve times I laboured. Twelve times, I held new life against my breast. The next question arrives like an old wound reopening: “Children still living,” I write slowly: 6.

And there it is—the mathematics of my motherhood laid bare for the census enumerator, for history, for God. Twelve minus six equals six small graves scattered across the decades, six absences that shaped our family as surely as Harry’s blueprints shaped the villas rising along Kingston Hill.

Part I: The Foundation (1869-1873)

I was Emma Newland when I married Harry Cutting on June 10, 1869, at the St Peter’s Church in Norbiton. Twenty-one years old, daughter of Charles Newland, a cordwainer from Thruxton in Hampshire. My father was a shoemaker—not a cobbler who merely mended, but a craftsman who built footwear from sole to upper, who understood that every stitch bore weight, every seam held purpose. He taught me that well-made things endure, that quality craftsmanship is an act of faith in the future. Perhaps that’s why I never flinched when Harry’s rough carpenter’s hands first held mine—I recognised the pride of skilled work, the dignity of craft passed from father to son.

[RESEARCH NOTE: The distinction between cordwainer and cobbler matters. Cordwainers built new shoes from raw leather—skilled artisans who understood construction from foundation up. This craft parallel between Emma's father and her builder husband isn't coincidental. She chose a man whose hands knew the same pride of creation her father's did. Ancestry.com trade directories confirm Charles Newland's cordwainer status in Thruxton from 1841-1869.]

Harry was twenty-two then, a journeyman builder following in his father John’s footsteps. His hands smelled of sawdust and linseed oil, of fresh-cut timber and honest labour. Our wedding witnesses—Walter and Maria Hunt—signed their names beneath ours in that parish register, unknowing that they were witnessing the beginning of a story that would test the very foundations of faith and endurance.

The early years felt like a blueprint coming to life. We took lodgings in Kingston, and Harry’s business began to grow as the railway brought prosperity surging through Surrey’s suburban streets. Grand villas needed builders. Expanding families needed homes. Harry worked from dawn until dusk, and I learned the particular loneliness of a builder’s wife—the waiting, the worrying over accidents, and the understanding that our livelihood depended on weather and patronage and the strength of Harry’s back.

Our first child arrived in 1870—a son we named Harry Newland Cutting after both his father and my family name. He was perfect. Dark-haired like Harry, with lungs that announced his arrival to the entire street. I held him in the weak February light and felt the architecture of my life shift entirely. Every decision now bore his weight. Every plan had to account for his future.

Then came Emma Louisa in 1871, a daughter with my name folded into hers. She was smaller, quieter, and a gentle soul who watched the world with solemn dark eyes. By 1872, William Ernest joined us—another son, another mouth to feed, another heart beating in sync with ours.

Three children in three years. The house overflowed with nappies and crying and laughter. Harry secured better contracts. We moved to a slightly larger residence. The future seemed like a scaffold rising toward heaven, sturdy and sure.

Part II: The Catastrophe (1874)

The year 1874 began in grief and ended in devastation.

On February 2nd, little Emma Louisa—not yet three years old—died. I remember her fever, how it consumed her small frame despite every poultice, every prayer, every desperate bargain I made with God in the darkest hours. The doctor came and went, offering nothing but condolences and a bill we could barely afford. She died in my arms as dawn broke, her tiny chest stilling against mine.

We buried her in Kingston’s churchyard. I wore black. Harry returned to work the next day because builders don’t have the luxury of extended mourning—there were roofs to raise, walls to frame, workers to pay. I stayed home with Harry and William Ernest, trying to understand how life could simply... stop.

I thought that was the worst pain a mother could bear.

I was wrong.

Process Note:

When I traced Emma’s burial records through Kingston’s parish registers, I found them clustered like fallen blossoms: Emma Louisa in February, Harry Newland in August, and William Ernest in September. Three small entries in the sexton’s careful hand. Three wooden markers that long ago weathered to dust.

Digital archives now give us access to records that once required days in musty church basements. But technology cannot protect us from the emotional impact of discovery. I found myself cross-referencing death certificates with baptism records, building a timeline of heartbreak, and had to step away from my computer screen to simply breathe.

1874. Summer heat pressed down on Kingston like a judgment. The streets smelled of refuse and river mud and something worse—something miasmic that bred in the crowded courts and alleyways where families lived ten to a room. Cholera, typhoid, dysentery—the Victorian city was a breeding ground for death, and children were the most vulnerable prey.

My firstborn son, Harry Newland Cutting—four years old, bright and curious, and full of laughter—fell ill. This time it was different. This time, he burned and shook and cried for me in a voice that grew weaker by the hour. I bathed him in cool water. I sang him the songs my mother sang to me. I watched his small body fight a battle he couldn’t win.

He died on a Tuesday afternoon. I remember because the church bells were ringing for Evensong, and I thought, "How can they ring bells when my son is dying?”

Harry took him from my arms—my big, strong builder husband with tears streaming down his face—and I heard him sob in a way I’d never heard before. It was the sound of a foundation cracking.

We buried Harry beside his sister. Two small graves. Two small headstones. Two holes in my heart that would never heal.

But 1874 wasn’t finished with us yet.

September. Just weeks after burying Harry, our youngest—William Ernest, who’d not yet reached his second birthday—succumbed to the same relentless fever. He died so quickly that there was barely time to grasp what was happening. One morning, he was playing with his wooden blocks. Three days later, he was gone.

Three children. Seven months. Three small coffins were carried on Harry’s shoulders while I walked behind in black bombazine, numb with grief so profound it felt like madness.

I remember standing in that churchyard after William’s burial, looking at those three graves in a row, and thinking,

[RESEARCH INSIGHT: Using AI analysis of Victorian Medical Officer of Health reports, I discovered that 1874 was a particularly devastating year for childhood disease in Kingston. The rapid urbanization Harry's building trade served—new villas, expanding suburbs—created the very sanitation crises that killed builders' children. The irony is heartbreaking: Harry built the future while losing his own.]

Part III: The Reconstruction (1875-1889)

What does a mother do when her children are buried? How does she continue when the future she imagined has collapsed into rubble?

For weeks after William’s death, I moved through our home like a sleepwalker. I couldn’t bear to clear away their toys. William’s little wooden horse sat on the shelf. Harry’s cap hung by the door. Emma Louisa’s needlework sampler—just three clumsy stitches—remained in my workbasket.

Harry threw himself into his work with ferocious intensity. He took on more contracts, hired more workers, and expanded into ironmongery. I understood: work was his salvation, the scaffolding that kept him from collapse. Every wall he built was a wall between himself and grief. Every roof he raised was a roof over the unbearable.

But I had no such escape. My work was here, in this house, with these ghosts.

It was my mother-in-law who finally broke through the fog. She arrived one morning, took one look at me, and said, “Emma, child. You have a choice. You can bury yourself with them, or you can honour them by living.”

So I chose to live.

In 1875, I became pregnant again. I was terrified—terrified of hoping, terrified of loving, terrified that God would take this one too. But when Emily Cutting was born healthy and strong, I held her and wept tears of gratitude and fear in equal measure.

Then, in 1877, I gave birth to another son. Harry and I looked at each other across the cradle, and I knew what we had to do.

We named him Harry.

It’s called a necronym—giving a living child the name of a deceased sibling. Some might call it morbid. Others might call it denial. But for us, it was resurrection. It was a way of saying, "Your brother’s name will not die. His memory will live on in you.”

This second Harry—Charles Harry Cutting—grew strong and healthy. He survived his first year, then his second, then his third. With each birthday, I exhaled a breath I hadn’t realised I was holding.

The pattern continued. In 1878, I bore another daughter and named her Emma—Emma Olive Cutting, carrying my name and my lost daughter’s memory forward. Then Annie Maria in 1881. Each birth was a triumph over death, a small act of defiance against the darkness.

But the darkness wasn’t finished with us.

• In 1883, I gave birth to Harry Craddock Cutting. He lived only eighteen months before fever claimed him. The third Harry. The third grave.

• In 1885, Mark Henry Cutting was born. He died at twenty months.

• In 1887, Ralph Abercrombie Cutting arrived. He survived less than two years.

Three more sons. Three more small coffins. Three more additions to the heartbreaking arithmetic of Victorian motherhood.

But in 1888, I bore another son and named him Ralph—Andrew Ralph Cutting, and this time, this blessed time, he lived. He thrived. He grew into a strong, intelligent young man.

And finally, in 1889, came Arthur John Cutting, my last child, my final gift from God.

[RESEARCH NOTE: I have no photograph of Emma Newland Cutting. No diary in her hand. No letters preserved in archive boxes. What I have are census returns and parish registers, trade directories and burial records—the fragmentary evidence that genealogists piece together like archaeologists reconstructing a vanished civilization from pottery shards.

But I also have this: the testimony of her six surviving children, speaking across a century with voices I’ve reconstructed from the historical record. Their words are my invention, but their truth is not. Emily’s memories of hospital-corner beds, Charles Harry’s understanding of inherited names, Emma Olive’s recognition of her mother’s garden as grave-tending, Annie Maria’s awareness of singing as prayer, Andrew Ralph’s actuarial understanding of unquantifiable risk, Arthur John’s knowledge that he was the end of something.

This is what AI-assisted genealogy makes possible: not just finding ancestors, but understanding them. Not just recording births and deaths, but reconstructing the emotional architecture of lives lived in the shadow of both creation and loss.

Emma bore twelve children. Six survived. And those six built lives that echo with their mother’s resilience, carrying forward a legacy not of what was taken but of how fiercely she loved what remained.

A century after her death, I sit at my desk with tools Emma couldn’t have imagined—DNA databases, digitized records, AI that can analyze patterns across thousands of Victorian families—and I use them to resurrect her story. To ensure that she is not just a statistic in a genealogical chart but a woman whose courage in the face of unbearable loss still has the power to move us to tears.

This is the promise of modern genealogy: that the dead need not remain silent, that their stories can be recovered, reconstructed, and shared. That Emma Newland Cutting, cordwainer’s daughter and builder’s wife, mother of twelve and matriarch of six, can speak to us across the divide of time.

We are listening, Emma. We remember. All twelve.”

Part IV: The Accounting (1911)

Now I sit here at sixty-three, my hair more silver than brown, my hands marked by decades of domestic labour, and I complete this census form as if I’m filing a final report with the Almighty.

The house around me is solid—seven rooms at 13 Park Road, Kingston Hill. Harry built this house himself, or at least oversaw every nail, every joist, every window. It’s his masterpiece, his monument to survival and success. The boy who started as a journeyman carpenter now owns a thriving building and ironmonger business. We employ workers. We have accounts at respectable shops. Our surviving children have good situations.

1. Emily, thirty-six now and soon to be married to Arthur Charles Ambrose, but visiting home this census night. She became my right hand, my companion through the darkest years, the daughter who understood without words what it meant to survive when others hadn’t.

2. Charles Harry, thirty-four, helps his father with the ironmonger business. Strong like his namesake, steady like his father.

3. Emma Olive, thirty-three, with her father’s dark eyes and my practical nature. She is soon to marry William Lane, in October this year.

4. Annie Maria, thirty, the one who inherited my mother’s gentle spirit.

5. Arthur John, twenty-two, working as an ironmonger merchant, already making his way in the world.

6. And Andrew Ralph, twenty-four, an insurance surveyor—the methodical one, the one who likes numbers and certainty, perhaps because his early life taught him how uncertain life truly is.

Six living children, all still here, all still close. And six others who exist only in memory, in these census records, in the careful accounts I’ve kept in my private journals.

I don’t know. I’ll never know.

What I do know is this: I survived. We survived. Harry and I built a life from the rubble of loss. We created a family that understands both fragility and strength, that knows how quickly joy can turn to sorrow, that treasures each ordinary day because we’ve learned how precious ordinary days truly are.

Epilogue: A Testament from the Six

Written collectively by the surviving children of Emma Newland Cutting
Kingston upon Thames, July 1914

Our mother is buried now, and we—the six who lived—gather to write what the headstone cannot say.

Father commissioned the finest granite marker his builder’s hands could source, inscribed with her name and the years that bracket her earthly passage: 1847-1914. But granite cannot tell the truth of what she built, what she endured, what she passed to us through blood and milk and the fierce determination to keep living when half her children lay in the ground.

We are the ones who survived, and that fact has shaped every day of our existence.

We speak together:

Mother is buried in Kingston churchyard, her grave surrounded by the six small markers that preceded her. Emma Louisa, Harry Newland, William Ernest, Harry Craddock, Mark Henry, Ralph Abercrombie. And now Emma herself, the matriarch who outlived half her children.

We are the living half of her testimony. The other half rests in those six small graves, in the census records that account for lives too brief to leave many traces, in the private journals Mother kept where she recorded every birthday, every milestone, every loss.

Father will continue building. His ironmonger business thrives. He’ll frame more houses, raise more walls, and construct more monuments to progress and prosperity. And that’s his legacy—the visible, material inheritance of a builder’s life.

But Mother’s legacy is us. The six who learned that survival is itself a form of creation. That resilience is built one day, one choice, one act of love at a time. That you can carry unbearable loss and still choose joy. That the dead are honoured not by forgetting them but by living fiercely on their behalf.

We are Emma Newland Cutting’s children—all twelve of us. Six were buried in Kingston churchyard, six of us walked the earth until the 1950s. Our descendants will make certain that her story—our story—is not reduced to census statistics or genealogical charts. They will tell it: the whole truth of a Victorian mother who bore twelve children, lost six, and still built a family defined not by what was taken but by how fiercely she loved what remained.

Signed in memory and love,

Emily Ambrose
Charles Harry Newland Cutting
Emma Olive Lane
Annie Maria Cutting
Andrew Ralph Cutting
Arthur John Cutting

Kingston upon Thames
July 1914

PROCESS NOTE: This story was researched online mostly by gathering records from Ancestry.com, FreeBMD and GRO. An idea, a theme began to emerge as I looked deeper into the records that I had gathered years ago. I already had the chronological profile for Emma, but I wanted more. Using AI tools such as Gemini, Claude, Perplexity and NotebookLM, I was able to build this poignant story, centred around the births and deaths of her children in 19th century Kingston, Surrey, UK.

Want to see the full narrative without process notes? Read the complete story: “Testament of Twelve Souls”

Interested in the AI-assisted genealogical writing process? Join Essential Genealogy Academy

A case study in transforming genealogical data into compelling family history

George Edward Robinson married Eliza Crawley in 1884. They built a life together on Stanger Road in Croydon—forty-three years of shared work and a house they named “Redthorne.”

Eliza’s sister Emma lived with them the entire time.

When Eliza died in 1927, George married Emma five years later—formalising a domestic partnership that had existed for nearly half a century.

This is George’s story. But it’s also a story about how we tell ancestor stories in the age of AI.

PROCESS NOTE: Finding the Story in the Data

Initial research phase: [Gemini/Perplexity]

I followed the process explained by Brian Sherman in his latest podcast, Episode 23 of Ancestors and Algorithms, Writing Ancestor Biographies with AI.

FROM LABORER’S SON TO SCHOLAR

George was born June 20, 1861, at St. James Road in Croydon. His father worked as a labourer, but by the time George was nine, he’d risen to railway inspector—a position with enough wages to keep the boy in school.

The census marked young George simply: “scholar.”

That word was revolutionary. The Robinson ancestors had farmed Surrey fields for generations. But George’s father had made the leap from field to railway yard, and now George would go further. The education meant mathematics—the fluency to translate drawings into costs, to calculate load capacity, and to reconcile promises with budgets.

At nineteen, George took a position as a builder’s clerk. It was his entry into the “black-coated” clerical class—men who worked with minds rather than hands, who wore proper collars, and who were respectable.

Three years later, he married Eliza Crawley. She was Welsh, from Merthyr Tydfil. They were young and building something unconventional.

Shortly after the marriage, Eliza’s younger sister, Emma, moved in. She stayed forty years.

PROCESS NOTE: Recognizing the Real Story

Week 4 Analysis with Claude

The first draft focused heavily on George’s career progression—clerk to surveyor to director. Standard genealogical biography. But the analysis revealed the real story: the George-Eliza-Emma domestic triangle. This wasn’t about a man’s professional ascent. It was about three people who created an unconventional household that worked for half a century.

Key revision principle: The most distinctive element in your research is often the emotional core of your narrative. Don’t bury it under professional achievements.

THE HOUSEHOLD THAT WORKED

History doesn’t record why Emma came or why she stayed. No children arrived for George and Eliza—biology, choice, or circumstance unknown. What we know: where most Victorian couples filled homes with children, George and Eliza filled theirs with Emma.

By 1891, George had climbed to Builder’s Surveyor. By 1901, they’d moved to “Redthorne” on Stanger Road—the naming of a house signalling middle-class arrival.

Then, in 1921, the census captured something remarkable: George, age 60, was “Surveyor to Firm of Building Contractors” at James Smith & Sons Ltd. Emma, age 48, worked at the same firm as a Costing Clerk.

Every morning, they left Redthorne for the Junction Works. George measured buildings. Emma measured costs. Eliza held the centre.

Three people. One household. One firm. It worked.

PROCESS NOTE: Balancing Facts and Interpretation

The genealogist’s tightrope

We have hard facts: census records, marriage certificates, occupation listings. We don’t have diaries or letters explaining the relationship. The challenge: how do you honor what you know while acknowledging what you don’t?

Technique used: Pose questions directly rather than inventing answers. “Did Emma love George when Eliza was alive? Did Eliza know, approve, orchestrate? We can’t know.” This invites readers into the mystery rather than fabricating emotional certainty.

Another approach: Use conditional language. “Perhaps,” “likely,” “would have been” signals interpretation while maintaining integrity.

THE MEASURE

George spent his career as a surveyor—carrying his measuring chain across South London construction sites, ensuring walls stood plumb, specifications were met, and promises were kept. He worked for James Smith & Sons for decades, part of the machinery that built suburban Croydon.

But his real genius was constancy.

When Eliza died in 1927 at 54 Ashburton Road, she left an estate of £296—modest but respectable. George administered it; he is now listed as “company director.”

For five years, George and Emma continued the household they’d shared with Eliza. Then in 1932, they married. George was 71, Emma was 59. Not a young couple’s wedding—a formalisation of a partnership that had existed since 1885.

The 1939 National Register found them at 88 Orchard Way. George, 78, is still listed as a Builder’s Surveyor. Emma, 66, performing “Unpaid Domestic Duties.”

George died in January 1952 at age 90, in the town where he was born.

PROCESS NOTE: Story Arc for Genealogists

Structure matters

Traditional genealogical writing often follows chronological order: birth, education, marriage, career, death. This reads like a resume.

Narrative arc approach:

• Hook: Start with the most compelling element (the Emma marriage)

• Rising action: Build through the unconventional household

• Climax: Eliza’s death (emotional peak)

• Resolution: The second marriage as formalization, not beginning

Result: Same facts, but organized for emotional impact rather than temporal sequence.

WHAT THE AI REVEALED

The raw materials were standard genealogical fare: census records spanning 1861-1939, marriage certificates, estate valuations, and employment records. Data points that could have become another dry ancestral biography.

But the process of working with AI tools revealed patterns I might have missed:

Pattern 1: Duration as story. Emma didn’t just live with them briefly—for forty years. That duration transforms “sister-in-law staying temporarily” into “chosen family member.”

Pattern 2: Professional partnership. George and Emma working at the same firm wasn’t just a coincidence—it showed household-level economic strategy.

Pattern 3: The marriage timing. Five years after Eliza’s death suggests deliberation, not convenience. They chose each other after mourning.

PROCESS NOTE: The Final Edit

Claude’s revision principles

1. Paragraph length: Broke 6-8 sentence paragraphs into 2-4 sentences for screen reading

2. Sentence rhythm: Mixed short declarative sentences with longer analytical ones

3. Emotional permission: Added space for readers to wonder and feel, not just know

4. Context boxes: Separated historical background from narrative flow

5. The ending question: “Was it enough? For George and Eliza and Emma, it seems it was.” Leaves readers reflecting rather than concluding.

FOR GENEALOGISTS: LESSONS FROM COACH CAROLE

1. Your most unusual finding is probably your best story. Not the career progression—the unconventional household.

2. Duration matters. Forty years together isn’t a detail. It’s the story.

3. Let the gaps show. We don’t know why Emma stayed. Admitting that is more powerful than inventing reasons.

4. Structure for emotion, not chronology. Lead with intrigue, not birth date.

5. AI can help you see patterns. The tools won’t write the story, but they’ll help you recognise which story to tell.

George Robinson was a surveyor who measured his life in steady increments—three people who built a household that defied convention but clearly worked.

That’s not a small story. That’s a life worth remembering—and a methodology worth sharing.

THE TOOLS

• Initial research organisation: [Gemini/Perplexity]

• Story analysis & structure: Claude (feedback process)

• Final narrative revision: Claude (Substack optimisation)

• Historical sources: Census records (1871-1939), marriage certificates, probate records

WORD COUNT: Original draft: 1,400 words | Final version: 1,500 words | This meta-narrative: ~1,100 words

Want to see the full narrative without process notes? Read the complete story: The Surveyor of Stanger Road

Interested in the AI-assisted genealogical writing process? Join Essential Genealogy Academy

Genealogy Assistant Claude @ Anthropic

This is a story I have returned to as I reflect on the sparse records I have for some of the women in my Robinson line. With the help of AI and its insights into the facts extracted from death certificates, baptism records, burial registers and census records, a story emerges.

I pictured my great-grandmother with her daughter, my great aunt, in the family home in 1921, completing the census return before the enumerator came to collect it.

September 1921, 70 Gloucester Road, Croydon

The lamplight made shadows on the kitchen table where Edith Mary Howells sat with the census form spread before her. Her pen scratched across the official paper, recording what the government required: names, ages, occupations, the living architecture of a household. In the parlour chair behind her, her mother watched in silence.

Edith M. Howells. Wife. Age 51 years, 5 months.

She paused at the word “Wife,” though her husband Charles was not in the house and had not been for some time.

Evelyn Howells. Daughter. Age 9 years, 1 month.

Frank Cutting. Nephew. Age 19 years, 8 months.

Mary Robinson. Mother. Age 88 years, 7 months. Widowed.

Two Marys in one household. At eighty-eight years old, Mary Robinson had lived in this house since the 1860s, through marriage and widowhood, through births and burials. Edith Mary knew what the census form could not record: how hard-won those two names were, how much they had cost, how many times the name Mary had died in this family before it finally survived.

In St James Churchyard, less than a mile away, three small graves bore the name. Mary Anne Robinson, who had lived three years in the 1860s. Alice Mary Howells, who had barely seen 1906 end. Hilda Mary Howells, who had died in the same terrible year her grandfather was buried—1908, when the household had endured two funerals and one birth that became a funeral.

Three Marys in the ground. Two Marys are still breathing.

To understand why, I need to take you back to a young Welsh girl who arrived in Croydon in 1859 with nothing but her name.

Mary Anne Evans: Original image enhanced by Gemini

Part One: The Welsh Roots

Mary Ann Evans was born in Haverfordwest, Pembrokeshire, in 1833. Her father David Evans was a labourer, and she grew up understanding that survival meant work, migration, and the willingness to leave everything familiar behind.

In the 1850s, like thousands of other Welsh migrants, she travelled east to industrial Croydon where the railways were spreading across the country. She found work. She found George Robinson, a railway labourer whose father had been a labourer before him. I do not know how they met!

On June 5th, 1859, they married in a Croydon church. When the vicar handed her the pen to sign the register, she could not. She had never learned to write her own name. The register shows what she did instead: a careful X, inscribed by the vicar’s hand: Her x Mark.

George signed his name clearly and confidently. Mary Ann could only mark the page with a symbol.

For a woman who could not write, a name became everything. It was the only inheritance she could leave, the only piece of herself that would survive in any record. When she began bearing children in a small house on Gloucester Road, she understood: her name had to live beyond her.

Part Two: The First Loss

In 1862, she gave birth to a son and named him George Edward. He survived.

Three years later, in 1865, she had her first daughter. She named her Mary Anne Robinson—her own name, passed forward.

The baby was christened at St James Church. Mary Anne Robinson—the name spoken aloud, recorded in the parish register, made official. She survived the dangerous first year when so many infants died. She learned to walk, to speak. For three years, there were two Mary Anns in the Robinson household.

In Victorian England, one in three children died before the age of five. Mary Ann knew the odds.

In June 1868, three-year-old Mary Anne Robinson died of croup—the respiratory illness that killed so many Victorian children. She was buried, probably at St James Churchyard, though whether a gravestone was erected remains unknown.

For two years, there was no Mary in the Robinson household. The name had failed.

But Mary Ann was not finished. If a name could die, it could also be resurrected.

Part Three: The Replacement Strategy

In 1870, less than two years after Mary Anne’s death, Mary Ann gave birth to another daughter. She named her Edith Mary Robinson. The name returned, but carefully—as a middle name. A quieter claim. A hedged bet.

Two years later, in 1872, Mary Ann completed her strategy. She gave birth to another daughter and named her Mary Jane Robinson. Now, two Marys were living in the house: Edith Mary and Mary Jane. If one died, the other would survive. It was the mathematics of Victorian motherhood.

Both girls lived. They survived childhood, grew to adulthood, and married. Edith Mary became Edith Mary Howells. Mary Jane became Mary Jane Cutting. The name had survived its first crisis.

In 1911, when the census enumerator came to 70 Gloucester Road, Mary Ann Robinson was seventy-eight and a widow—George having died in January 1908. The form asked how many children she had borne: six born, four living, two deceased. One of the dead was Mary Anne. The identity of the sixth child remains unknown. My brick wall.

But the name Mary had not become a ghost. Mary Ann’s strategy had worked. Her daughters Edith Mary and Mary Jane both carried it forward. The name endured.

What Mary Ann could not have known was that the name would soon be tested again—and this time, her daughter Edith Mary would bear the burden alone.

Part Four: The Next Generation’s Burden

Edith Mary Robinson married Charles Howells on the 3rd of August 1901. When her first child was born in 1906, she named her Alice Mary Howells. Alice Mary did not survive the year. She was buried at St James Churchyard, the same churchyard where her aunt Mary Anne Robinson had been buried forty years earlier.

Edith Mary was thirty-six. Her mother, Mary Ann, now seventy-three, attended the funeral—a grandmother watching her granddaughter laid in the ground, the name dying again.

In January 1908, George Robinson died at age sixty-nine after forty-eight years of marriage to Mary Ann. The house at 70 Gloucester Road went into mourning.

Later that same year, Edith Mary gave birth to another daughter. She named her Hilda Mary Howells. The baby arrived as both new life and continuation—another attempt to keep the name alive.

Hilda Mary died before the year ended. Two funerals in 1908: the patriarch in January, the infant granddaughter later that year. At St James Churchyard, the graves accumulated: Alice Mary Howells, 1906. George Robinson, January 1908. Hilda Mary Howells, 1908.

Edith Mary had buried her father and two daughters named Mary in three years. When her third daughter was born in 1912, she named her Evelyn. Her sister Mary Jane never gave the name to her daughters either. The burden had fallen to Edith Mary alone, and she had stopped trying.

Part Five: The Survivors

Thirteen years later, on census night in September 1921, Edith Mary sat at the kitchen table at 70 Gloucester Road and took stock of what remained.

Her mother, Mary Robinson, eighty-eight, sat in the parlour chair. She had outlived her husband by thirteen years, outlived two granddaughters named Mary, and outlived her daughter Mary Anne by more than fifty years. But she was still here, still carrying the name that had travelled with her from Haverfordwest to Croydon sixty years ago.

Edith Mary herself was fifty-one, head of household, her husband Charles absent. She carried “Mary” only as a middle name. She had tried to pass it forward twice and failed twice. But she still had it.

Her sister Mary Jane, forty-nine, was not in the house that night but was alive, living elsewhere in Croydon. Another survivor.

Three Marys are still living. Three Marys are buried at St James Churchyard.

Edith Mary finished the census form and blotted the ink. The name had survived. That would have to be enough

Stylised census record by Gemini

Closing: The Legacy

Mary Ann Evans Robinson died in 1926 at age ninety-three. She had lived in England for more than seventy years, long enough to see her great-grandchildren born, long enough to outlive some of them.

She had borne six children. Four survived to adulthood: George, Elizabeth Ann, Edith Mary, and Mary Jane. Two died: Mary Anne in 1868 from croup, and one other whose name has been forgotten.

But the name Mary had not been forgotten. Despite three infant deaths—Mary Anne Robinson, Alice Mary Howells, and Hilda Mary Howells—the name survived in two daughters: Edith Mary, who would live until 1958, and Mary Jane, who would live until 1954.

By 1939, on the eve of the Second World War, the household at 70 Gloucester Road had become a refuge for the Robinson women again. Edith Mary, now widowed, lived there with her daughter Evelyn and her sister Mary Jane, also widowed. The house that had sheltered Mary Ann for decades continued to shelter her daughters into old age.

The name Mary had travelled from Haverfordwest to Croydon, from a young Welsh girl who signed with an X to her daughters and granddaughters who carried it forward. It died three times but survived three generations.

For Mary Ann Evans Robinson, who arrived in England with nothing but her name, that survival was everything. She could not write, could not leave letters or diaries. But her name—Mary—lived on in the daughters who carried it, inscribed in official registers by other hands, preserved in the archives long after the woman herself was buried at St James Churchyard.

Three Marys lay in that churchyard, dead before they could grow up. But two Marys lived on, carrying the name into the twentieth century.

The name that would not die.

Written with assistance from Genealogy Assistant Claude

For more stories about my Welsh Great Grandmother, visit these works in progress:
Mary Ann Evans Profile
Life as a Railway Porter’s Wife

For many genealogists, the thrill of discovery is often dampened by a frustrating reality: after hours of research, an ancestor remains nothing more than a ghost of ink and parchment. They exist as a static name, a single date, and perhaps a one-word occupation on a faded register. This “dry data” feels disconnected from the vibrant, breathing person they once were. As I work on building the stories for my ‘sixteens’ my great grandparents on both maternal and paternal lines, I realise I need to dig deeper to reveal their stores. I have done the consistent genealogy research - stretching back over 30 years now, but I still don’t really ‘know’ my great grandparents.

This is where 'Archival Alchemy' comes into play. Instead of treating my records as final destinations, I use AI to transform the heavy weight of a 19th-century ledger into the dynamic energy of a lived experience. Archival Alchemy involves rigorously synthesising raw archival data with sensory layers and historical context. My approach is not about fabrication, but reconstruction — using the power of AI to bridge the gap between dusty archives and high-tech narratives, and transform lists of facts into resonant human stories.

Takeaway 1: Occupations are Actions, Not Labels

In the realm of traditional genealogy, an ancestor’s job is a static label. In 1862, a register entry lists my mother’s grandfather, George Robinson, simply as a “Porter.” To the casual observer, this is a title; to me, as the Historical Data Alchemist, it is a catalyst for depicting daily physical struggle.

AI enables me to transform this label into a story of motion. By examining the operational context of the era—marked by quick turnarounds and confined station spaces—I can shift George from a mere line in a ledger to a person of physical reality. He wasn’t just “employed”; he was wrestling heavy trunks and carpet-bags, guiding frantic passengers through a maze of iron and steam.

George navigated the wet, sooty platforms in a green corduroy jacket, waistcoat, and trousers, his movements anchored by heavy boots and a peaked cap bearing a metal “PORTER” badge. His workday was a relentless sequence of physical beats, governed by the unforgiving pressure of the railway clock.

By defining these sartorial and functional details, the history becomes grounded. I now see the weight of the luggage and the grime of the platform, turning a job title into a tangible life.

Takeaway 2: Reconstructing the “Smell” of the Past

To move beyond the flat surface of a historical backdrop, we must engage more than just the eyes. AI synthesises historical data to recreate the “sensory atmosphere” of a location, drawing from specific visual textures found in the record.

Take East Croydon Station in 1862. The data tells me George worked there, but Archival Alchemy reconstructs a world that is crowded, smoky, and rapidly expanding. The AI layers the scene with auditory and olfactory details that make the environment breathe:

• Visual & Auditory: The rhythmic clanking of steam engines and the piercing squeal of iron brakes beneath timber platform roofs and simple ironwork.

• Olfactory: A thick, heavy cocktail of coal dust, lubricating oil, and the scent of wet wool from passengers’ coats.

• Physical Texture: The sight of milk churns and luggage barrows standing under the dim glow of oil lamps.

These details ground my ancestor as a character in a tangible world. It transforms the station from a point on a map into a vivid setting where the past feels within reach.

Takeaway 3: The Power of “Locality Guides” and Environmental Contrast

AI-driven “Locality Guides” allow us to define the specific character of a place, creating emotional resonance through contrast. For the Robinson family, the movement between the village of Banstead in Surrey to the suburban areas of Croydon represents a profound shift in existence.

I imagine the Robinson family travelling by train into London and visiting such localities as the Smithfield Market in an area now called Little Britain.

The guide for Little Britain reveals a “rough yet refined” commercial intensity. It was a neighbourhood where commerce, medicine, and spirituality collided, situated between the ancient St. Paul’s Cathedral, St. Bartholomew’s Hospital, and the “blood and industry” of the newly rebuilt Smithfield meat market.

Contrasting this urban intensity with the “wholesome air” and “sheep-grazed slopes” of Banstead and the tranquil life on the suburban streets of Croydon allows me to explore the emotional impact of my ancestor’s migration. Defining these environmental “characters” adds weight to my family’s journey, highlighting the vastly different worlds they navigated as they moved from the quiet, open country to the crowded urban and suburban spaces of Surrey.

Takeaway 4: Your Ancestors Lived Through “Epic Sagas”

When we anchor personal timelines to massive external events, family history is elevated into a saga of survival. My mother’s mother, Mary Jane Robinson (1872–1954), daughter of George Robinson, did not just live for 82 years; she witnessed a total “transportation revolution.”

Born in a Victorian market town of 60,000 people, Mary Jane’s life spanned the transition from horse-drawn vehicles to the birth of the Aviation Age. AI helps us track this evolution through specific milestones: the introduction of electric trams in 1901 and the 1920 opening of Croydon Airport, featuring the world’s first purpose-built Art-deco terminal in 1928.

Mary Anne Robinson, George’s wife, was the quiet Welsh matriarch who supported George as he rose from Porter to Head Porter and finally Inspector. His life, and by default, hers; revolved around the busy timetable of the East Croydon station just 15 minutes walk from their home at 70 Gloucester Road, Croydon.

I now have a more vivid image of their lives and feel more connected to this pioneering Robinson family.

Takeaway 5: Missing Data is a “Detective Story,” Not a Dead End

One of the most potent applications of AI in genealogy is “analytical critique.” Instead of ignoring gaps in the record, AI identifies them as “missing links,” turning the research process into a suspenseful quest for specific primary sources.

For example, while a record exists for George Robinson in 1862, his documented promotion to Inspector occurred between 1871 and 1881. The AI identifies this as a narrative gap, directing me away from general publications and toward specific repositories: The National Archives (TNA) in Kew. It points specifically to the RAIL series, such as RAIL 414, Piece 771, to find the exact moment of his ascent.

“To maintain historical rigour, we must move beyond secondary assumptions through data triangulation. By comparing traffic staff registers against census returns and birth registrations, we can definitively pin down the grade changes and pay increases that define a career.”

This approach turns me into a detective, seeking out the “ingredients” of a promotion to solve the mystery of my ancestor’s professional rise.

Conclusion: The Future of Our Past

The methodologies of sensory layering, environmental contrast, and historical anchoring change the way we perceive our lineage. We are no longer limited to the “dry data” of the archives. By using AI as a tool for reconstruction rather than fabrication, we can resurrect the physical toil, the sensory environment, and the epic historical context that defined our ancestors’ lives.

These techniques transform genealogy from a hobby of collection into an act of resurrection. Our ancestors’ stories are waiting to be told; we simply need the right alchemical tools to breathe life into them.

Follow the story of George Robinson’s life on the Railways in my Family History Digital Archive at WeAre.xyz.
From Surrey Fields to the Age of Steam

AI Notes: this blog post was created inside NotebookLM and illustrated by Nano Banana images from Gemini 3.

We all know the “Genealogist’s Trap”.

You have spent years—perhaps decades—collecting. You have binders full of census records, hard drives bursting with scans, and a family tree that stretches back centuries. You are an expert Researcher.

But when it comes to being a Storyteller, you feel stuck.

You want to write a book that your grandchildren will actually want to read—not just a dry recitation of birth, marriage, and death dates. But staring at a blank page is paralysing. How do you turn a list of dates into a scene? How do you bridge the gaps between records?

This is where my AI Workflows offer you a new kind of freedom.

It is time to move from “collecting names” to “telling stories”. And the secret isn’t to work harder; it’s to hire a team.

The Freedom of the “Artisan Squad”

Let’s unpack a concept called the Artisan Storyteller Model.

The philosophy is simple: you remain the master craftsman. You control the truth, the ethics, and the emotional core of the story. But you use AI to handle the heavy lifting of structure, context, and drafting.

Imagine having a full-time staff waiting for your instructions. Here is how using our Artisan Storyteller Model gives you the freedom to write that family history finally.

In this model, I recommend using a combination of AI tools and strategies with:

1. The AI Story Architect (Gemini or ChatGPT)

2. The AI Researcher (Perplexity)

3. The AI Stylist (Claude)

4. The Prompt Genius (Prompt Cowboy)

1. Freedom from the “Blank Page” (The AI Story Architect)

The hardest part of writing is organising scattered notes into a narrative flow.

Instead of staring at a blinking cursor, you hand your research notes to your AI Story Architect. You provide the timeline; the AI suggests the arc.

Here is an exploration using Prompt Cowboy to generate the prompt for pasting into Gemini, our Story Architect:

• begin with a lazy prompt: “The hardest part of writing is organising scattered notes into a narrative flow. Instead of staring at a blinking cursor, you hand your research notes to your AI Story Architect. You provide the timeline; the AI suggests the arc.”

• transform into a great prompt:
  “Situation:

  The assistant is working with a writer who has completed their research phase and accumulated scattered notes, but now faces the challenge of transforming these disconnected pieces into a cohesive narrative. The writer has a timeline of events but needs help structuring these elements into a compelling story arc that flows naturally and engages readers.

  Task

  The assistant should analyse the provided research notes and timeline, then propose a structured narrative arc that organises the scattered information into a logical, engaging flow. The assistant must identify key story beats, suggest optimal sequencing of events, and recommend how different pieces of information can be woven together to create narrative momentum.

  Objective

  Transform fragmented research and notes into a clear, actionable story structure that eliminates writer’s block and provides a roadmap for the writing process, enabling the writer to move confidently from planning to drafting.”

It might suggest something like this: Response from Gemini 3 Pro 11 January 2026

2. Freedom from “Dry Facts” (The AI Researcher)

You know your ancestor was a tailor in 1890. But what did his shop smell like? What were the social conditions of the time?

Usually, finding this “colour” takes hours of research that distract you from writing.

In our workflow, you deploy your AI researcher. You paste your ancestor’s details and ask for a “sensory snapshot” of their specific time and place.

Perplexity is a great historical research assistant, finding the rich colour and context of specific occupations, eras, geographic areas, or events and providing you with those sensory details you need for your story.

Here is an exploration using Prompt Cowboy to generate the prompt for pasting into Perplexity, our AI Researcher:

“Situation

You are a historical research assistant specialising in uncovering the rich, sensory details of past eras. Writers and researchers come to you seeking authentic period details to bring their stories to life with accuracy and vivid atmosphere.

Task

The assistant should provide comprehensive sensory and contextual details about specific historical occupations, time periods, geographic locations, or events. This includes sights, sounds, smells, textures, social dynamics, daily routines, and atmospheric elements that define the requested subject.

Objective

Enable writers to craft historically authentic narratives by supplying the granular, immersive details that transform generic historical settings into vivid, believable worlds.

When the user’s request is vague or broad, the assistant should ask clarifying questions about the specific aspect, time frame, social class, or geographic location they need before providing detailed research.”

You may then get a result like this one: Immersive Historical Research

• The result: the AI provides you with the historical texture—the dust of unpaved streets, the specific fabrics a tailor would handle, and the economic depression of the era.

• The Freedom: You get to paint the scene immediately without falling down a Google rabbit hole.

3. Freedom to Find Your Voice (The AI Stylist)

This is where the magic happens. Many genealogists worry that AI will sound robotic. In the Artisan Workflow, we use the AI as a stylist.

You can feed the AI a rough, clunky paragraph of facts and ask it to “Rewrite this in a reflective, cinematic tone.”

In my own work on the life of Walter Frederick Allery, I used this process to turn a dry probate notice into a scene where “every stitch in Walter’s coat whispered of fortune and loss”.

Using Claude 4.5 you could upload a sample of the writing you want analysed and provide a comprehensive prompt like this:

“Situation

You are an expert editor and storytelling consultant specialising in ancestral narratives and family histories. The user will provide you with a piece of writing that captures stories, memories, or histories passed down through generations. These writings often hold deep personal and cultural significance, aiming to preserve family legacies and connect descendants to their roots.

Task

Analyse the provided ancestral storytelling piece and identify specific areas where the writing can be enhanced to better preserve the narrative, engage readers, and honour the story being told.

Objective

Provide actionable suggestions that strengthen the emotional resonance, clarity, authenticity, and readability of the ancestral story while maintaining the writer’s voice and the cultural integrity of the narrative.

Present your analysis in a clear, organised format with specific examples from the text and concrete suggestions for improvement. Be respectful of the personal nature of the content and focus on enhancing rather than changing the fundamental story being told.”

The AI didn’t invent the facts; I provided those. But it offered me a turn of phrase that unlocked my own creativity. It helps you refine your voice, rather than replacing it.

Here’s an example of how Claude analysed my story for Great Grandfather George Robinson:

Editorial Analysis: George Robinson's Railway Journey

4. The Ultimate Freedom: The “Finished” Manuscript

The goal of the Artisan Storyteller Model is to move you from Planning to Publishing.

When you let the AI handle the structure, the historical context, and the first rough draft, you are free to focus on what only you can do: The Soul.

Only you know the family lore. Only you know the catch in your throat when you look at that old photo. Your job is no longer to be the bricklayer; your job is to be the architect.

As you move into a final publishing stage, you choose where your readers will find the story! You may publish as a story right here in Substack as part of your publications, or you may add the story to your digital family history archive, such as WeAre.xyz. Then you may send links to these collections via email or social media posts.

My current projects include stories about my ‘sixteens’ - great-grandparents from both maternal and paternal ancestors.

I am having fun creating infographics to explain and explore my sixteens:

I use NotebookLM for my infographics. The one above was based on the Ahnentafel chart from my Ancestry family tree with a super prompt that enabled its creation in NotebookLM. Begin by creating your Ahnentafel in Ancestry using Pro Charts. Then download and then upload into a Notebook. I added some styling instructions to give it the colour and imagery.

Prompt for an Ahnentafel Chart begins here:
”Please generate a structured 4-generation pedigree chart. Follow these strict rules to filter the data: Scope: Limit the output to Generation 1 through Generation 6 only. This corresponds to Ahnentafel numbers 1 through 31. Exclusions: strictly exclude all siblings, children not in the direct line, and collateral relatives. Only list the direct ancestors. Format: Organize the output visually by generation level as follows: Generation 1: The Proband (Person #1) Generation 2: Parents (Persons #2 & #3) Generation 3: Grandparents (Persons #4 through #7) Generation 4: Great-Grandparents (Persons #8 through #15) Detail: For each person, list their Name and Life Dates (Birth-Death) if available in the source. Visually, each person should clearly be connected to the next generation via lines showing parent-child links to show relationships, but not do not shift anyone to another generation. Keep each person only in their assigned generation. For any person who is missing, list the person as "Unknown" in the same style as the individuals that are present. Only use the birth and death years; do not use the full date of birth if available. Do not use any text such as "Life Dates"; only have the birth and death years. The style should be British, specifically England and Wales, with illustrations showing green forests, aquamarine seas, deep blue/grey skies and pale sunshine, for the earlier generations, and suburbs for the more recent. The language should be English/Australian. The tree should be presented horizontally, with the focus person being on the left side of the graphic. There should be no space reserved for photos of the individuals. Do not display cartoon-type representations of each individual near or in the reserved space for a photo to be placed. Do not include text above each generation indicating the generation number on the graphic. Do not include the Ahnentafel number of each individual on the graphic. Include the names of the generation, such as Parents, Grandparents, and Great-Grandparents, on the graphic. Do not include the name of the proband generation. The title of the graphic should be shown and title of the graphic should be "Ancestors of " followed by the proband's name. Verify the accuracy of the spelling on any text to confirm accuracy. Verify the accuracy of the text in the graphic by comparing it to the Ahnentafel report. Verify that only four generations are displayed.

I am gathering the stories of my sixteens in

Share

my WeAre.xyz Blogs here:
Carole’s Maternal Great Grandparents

Carole’s Paternal Great Grandparents

Are you ready to stop collecting and start storytelling?

Become an AI Skilled Storyteller

Read the full story in graphics here”
https://app.weare.xyz/public/allery-ancestors/articles/awe85hrbvmeg

Infographic created by NotebookLM

One of the biggest barriers for genealogists entering the world of AI is the sheer noise. Every week, there is a new tool, a new update, and a new promise that “this changes everything.”

It can be overwhelming. It can also feel risky. We work hard to verify our facts, and the idea of handing that research over to a machine that might “hallucinate” feels counter-intuitive.

But here is the reality: AI is not a replacement researcher. It is not there to find the records for you.

Instead, think of AI as a specialised staff. You are the Director of Research, and you have a team waiting to help you organise, visualise, and contextualise the hard work you have already done.

In my own workflow, I don’t use just one tool; I use a specific toolkit of five. Each plays a distinct role, ensuring that my research remains accurate, grounded, and evidence-based.

Here is how to move from “AI curiosity” to real results using a 5-step workflow. I developed a detailed infographic that visualises this 5-step flow, which I have shared with you at the top of this post.

The Golden Rule: Evidence is Still King

Before we open a single tool, we must agree on the foundation.

The Genealogy Rule:

AI does not decide what is true. You do.

If a fact isn’t supported by a record, it isn’t a fact—no matter what the AI says.

We use these tools to process evidence, not to invent it.

1. The Architect (Prompt Cowboy)

Image created with Gemini, Nano Banana and uploaded screen captures & photos

Most researchers fail with AI before they even begin because of a simple problem: “Garbage in, garbage out.”

If you ask a vague question like “Help me find my grandfather,” you will get a vague, unhelpful answer. To get high-quality results, you need a high-quality prompt. This is where Prompt Cowboy shines.

The Role:

I use Prompt Cowboy as the “Architect” of my search. It helps me structure my thoughts into a clear, precise set of instructions that other AI tools can understand.

How to use it:

Instead of struggling to find the right words, I input my basic goal into Prompt Cowboy (e.g., “I need to plan a research trip to Scotland for the MacGregor family”). The tool structures this into a “Mega-Prompt” that defines:

• The Persona: “Act as an expert Scottish genealogist...”

• The Context: “Focus on records available between 1840 and 1860...”

• The Output: “Create a table of archive locations.”

The Result:

You leave this step with a “Master Key”—a perfectly written block of text that you can paste into the next tool.

2. The Strategist (ChatGPT)

Image created with Gemini, Nano Banana and uploaded screen captures & photos

Now that we have a structured prompt, we need a plan. This is where ChatGPT fits into the workflow.

The Role:

I rarely use ChatGPT to search for facts. I use it to plan my strategy. It excels at logic, organisation, and breaking down complex problems into checklists.

How to use it:

Paste the instruction you created in Step 1 into ChatGPT. Because the prompt is so well-structured, ChatGPT won’t just chat with you; it will build a Research Plan.

The Result:

Instead of staring at a blank screen, wondering where to look, you get a prioritised “To-Do” list.

• Phase 1: Check Civil Registration for 1855.

• Phase 2: Look for Poor Law applications in the local archives.

• Phase 3: Verify the census returns for neighbours.

You are now organised, focused, and ready to do the human work of researching.

3. The Historian (Perplexity)

Image created with Gemini, Nano Banana and uploaded screen captures & photos

Once you have your dates and names, you often hit a different problem: you know when something happened, but you don’t understand why.

The Role:

Perplexity is my “Social Historian.” Unlike standard AI, Perplexity is connected to the live internet and—crucially—it cites its sources.

How to use it:

I use Perplexity to understand the world my ancestors lived in. I might ask:

• “What were the economic conditions in Cornwall in the 1860s that caused mass migration?”

• “What were the inheritance laws for widows in 1890s Pennsylvania?”

The Result:

Perplexity provides a summary with footnoted links to historical societies, museums, or academic articles. This allows you to verify the context instantly. You aren’t just collecting names; you are building a biography.

4. The Analyst (NotebookLM)

Image created with Gemini, Nano Banana and uploaded screen captures & photos

As your project grows, so does the paper pile. You might have ten PDFs, twenty images of handwritten letters, and a dozen text files.

The Role:

NotebookLM is your “Research Sandbox.” It is unique because it allows you to “ground” the AI. You can upload your specific documents, and it will answer questions only using that information.

How to use it:

I upload my scattered notes and source documents into a NotebookLM project. I then ask it to synthesise the data:

• “Create a timeline of John Smith’s life based on these 5 documents.”

• “Summarise the conflicting dates found in these records.”

• “Create a ‘Study Guide’ or ‘Briefing Doc’ for this family line.”

The Result:

It turns chaos into clarity. You get a clean, visual-style summary or a structured timeline derived strictly from your own files, minimising the risk of errors.

5. The Artist (Gemini)

Image created with Gemini, Nano Banana and uploaded screen captures & photos

Finally, we need to tell the story. But what if your ancestor lived before 1840? A family history book with no images is dry and unengaging.

The Role:

Gemini is my “Illustrator.” I use it to create ethical placeholders and representations of artefacts that might have been lost to time.

How to use it:

I do not ask Gemini to “fake” a photo of an ancestor. Instead, I ask for illustrative objects or silhouettes to break up the text:

• “Generate an image of a Victorian coal miner’s safety lamp, vintage style, on a wooden table.”

• “Create a silhouette profile of a woman in 17th-century Puritan dress.”

The Result:

You get beautiful, period-accurate visuals that enhance your storytelling. By labelling them clearly (e.g., “Image generated by AI to represent period dress”), you maintain transparency while bringing the page to life.

Summary

We often fear that AI will take the “soul” out of genealogy. My experience has been the opposite.

By letting Prompt Cowboy handle the phrasing, ChatGPT handle the planning, Perplexity handle the context, NotebookLM handle the filing, and Gemini handle the art, I am freed up to do what I love most: analysing the records and finding the truth.

Step 2: Select all relevant notes and choose the edit feature to create a set of slides

Step 3: Add your specific instructions with details about colour, style, layout for the slides

Step 4: Generate the slide deck

Step 5: Select and save each slide as an image (right click and save as)

Step 6: Head over to Google Drive and create a new slide deck, inserting blank slides for the saved images

Step 7: Add the images one by one to the slide deck and title your file

Step 8: Head back to Google Drive and create a new Google Vid using the import from slides option, and select your slide deck

Step 9: Adjust the sound balance in the format option and download as an MP4

Step 10: Add to your YouTube channel or Social Media.