Every single day, we see a shiny new tool claiming it will automate your entire workflow. But what’s the reality? Most of us are paying monthly subscriptions for tools that either perform repetitive, basic tasks or completely fail to deliver on their hype.

If you want to know which AI tools are genuinely worth your monthly subscription, this article is for you. It’s built on months of hands-on testing and evaluating different AI platforms in real-world scenarios.

1. The Real Problem: You’re Paying for Redundant Tools

The biggest mistake most businesses make is “tool overlap”—paying for multiple subscriptions simultaneously when 90% of their features are identical.

The market is flooded with products that are essentially just ChatGPT wrapped in a fancier interface, charging you an extra monthly fee. Many of these tools simply use ChatGPT’s API behind the scenes to generate your answers. To put it clearly: it’s like paying a middleman to give you answers, while they secretly turn around, reach under the table, and ask ChatGPT themselves.

Before buying any AI platform, you need to look closely at its underlying infrastructure. Here are the three main causes of wasted time and money:

Relying on AI for things it shouldn’t be used for: The biggest offender here is content writing—something I’ve emphasized repeatedly.

Using general models for specialized work: You shouldn’t expect professional coding or web architecture from a general chatbot without a proper specialized environment.

Paying for overlapping tools: Holding multiple subscriptions that all draw power from the same base model underneath.

2. The Decision Framework: A Simple Rule

Before doing anything else, accept this reality: no single tool does it all. There is no “best” universal tool—only the right tool for the right job.

Rule 1: Separate your research tools from your creation tools. Searching and synthesizing information is fundamentally different from programming or brainstorming. Mixing the two only degrades the quality of your output.

Rule 2: AI is a supplement to skill, not a replacement for it. AI speeds up foundational work, but your SEO strategy, internal linking logic, ad campaign architecture, and email marketing funnels still require human reasoning and analytical thinking.

3. Hands-On Tool Breakdown: Lessons from the Field

Here is what months of testing and practical experience revealed about the major players:

ChatGPT Plus — Great for Quick Answers, Not for Copywriting

ChatGPT Plus provides fast responses, and its image generation prompts are exceptional. However, it falls flat when it comes to long-form explanations. In my testing across various scenarios, its output leans heavily on bullet points and brief summaries. On the flip side, it feels witty and remarkably human.

It handles everyday edits and straightforward translations well, but if you ask it to translate emotional text—like poetry—it fails to convey the feeling. I gave it Persian poetry and asked for a translation with the weight of Shakespearean verse; the output read like a 7-year-old rapping after finishing first grade with zero understanding of literature. It completely missed metaphors, idioms, and literary devices, choosing literal word-for-word translation instead.

For copywriting, ad copy, or SEO articles, it’s a poor choice—the tone is stiff and fails to engage real readers. Overall, it’s not built for core business execution, and I don’t believe the paid tier is worth it.

Perplexity Pro — The Research Powerhouse

Perplexity Pro is unbeatable for deep research, extracting data alongside clear, cited sources. If your business or role requires extensive research, comprehensive sourcing, and fact-checking, Perplexity Pro is outstanding. It saves hours of tedious web surfing and reference hunting.

However, if your day-to-day work doesn’t rely heavily on continuous research, the free tier will serve you just fine.

Claude Code — The Top Choice for Web & Development

Combining Claude with its dedicated coding environment yields incredible power for development, bug fixing, and understanding project architecture. Claude is significantly more precise than its competitors at comprehending complex logic and multi-layered prompts. I’ve proven this in my daily workflow—it’s a specialized coding power tool.

A quick disclaimer: This applies if coding is actually your expertise. If you run a general business and are using this purely to avoid paying a developer, please don’t quote me as your recommendation. AI outputs always require expert human oversight. Use AI where your professional domain knowledge lies.

On a lighter note, I saw a funny story on Reddit recently about this. A client was hiring a freelance web designer, and a candidate submitted a resume with portfolio sites hosted directly on Manus AI servers—all generated on the fly via prompts. The client handed over cPanel login credentials and said, “Migrate these sites over using JS, and you’re hired.” The candidate instantly vanished into thin air.

Manus AI — Impressive, but Replaceable

Manus generates autonomous agents. It’s highly technical and genuinely useful, though its design outputs feel dry and rigid.

Manus handles multi-step processes well, but the reality is that you can recreate most of these workflows using Claude Code combined with proper prompt scenario setups. If you already subscribe to Claude, there’s no need to spend extra on Manus.

Dear Substack, please let us draw tables💖😘. For now, suppose | as cols.

Quick Comparison Matrix

Tool | Best Used For |Worth the Subscription?

Claude Code | Coding, Web Architecture, Agent Construction | ✅ Essential for Developers

Perplexity Pro | Deep Research, Source-Backed Data Extraction | ✅ Great (If your role requires research)

ChatGPT Plus | Fast General Q&A, Image Generation |❌ Not Cost-Effective

Manus AI | Mixed Autonomous Workflow Agents | 🟡 Replaceable

Why Aren’t You Getting the Results You Want Yet?

Now you know where the money leaks are and how to pick your stack. But can you execute a winning SEO campaign or build a high-converting website just by knowing which tools to use?

Definitely not.

Knowing your tools is like owning a high-end toolbox—building a digital business still requires expert, specialized execution:

Standard Web Design: Clean codebase, rapid load speeds, and intuitive User Experience (UX).

Deep SEO Strategy: Sound internal linking architecture, adherence to E-E-A-T principles, and content that actually ranks.

Email Marketing & Sales Funnels: Re-engagement sequences and reliable automation logic.

Paid Ads (Google Ads): Budget optimization, continuous conversion A/B testing, and eliminating wasted ad spend.

Ready to Build a Digital Infrastructure That Works?

If you want to skip the trial-and-error cycle and build a tailored, data-driven execution strategy for your business, I’m here to help.

I provide specialized services in web design, sustainable SEO, high-converting email marketing campaigns, and performance Google Ads management—all customized around your specific data and goals.

If you’re ready to gain a clear edge over your competitors online:

Book a 15-minute project discovery call today. Submit your digital infrastructure review request, and let’s map out a custom growth trajectory for your business.

I’m Nasrin Bastani, and every week I share articles that make earning with AI a little easier. If you’re interested in marketing, follow my marketing newsletter: The Marketing Notes

Leave a comment

In the old days, Mullah Nasruddin and his wife decided one day to travel from their village to visit relatives in the neighboring village. Mullah Nasruddin and his wife prepared their donkey to set off, thinking to themselves that if they left in the morning, they would reach the neighboring village by nightfall. In short, they prepped and saddled the donkey, and both started walking alongside it. After going a little ways ahead, they ran into their neighbor. The neighbor greeted them warmly and chatted for a bit. Then he looked at the two of them, looked at the donkey, and said, “Goodness! If you were going to walk all the way to the next village, why did you bring the donkey along? Walking like this, you won’t get there even in a week!” Mullah Nasruddin and his wife looked at each other and nodded in agreement. Then both mounted the donkey and set off.

A bit further down the road, they saw the village headman. After greetings, the village headman took a look at Mullah Nasruddin, a look at Mullah’s wife, and a look at the donkey. He frowned and said, “Nasruddin! Have you and your wife both mounted this poor animal? This donkey suffers. Do you know how many kilos you weigh and how many kilos your wife weighs? How much endurance does this poor beast have? I bet you work it just as hard in the field! It won’t last two weeks before it dies.”; Mullah Nasruddin looked at his wife, looked at his own belly, and the realization struck both of their minds: “Yes, the headman is right, this poor donkey suffers.” Mullah Nasruddin dismounted and said, “I will walk beside the donkey; you stay mounted, my dear.” And they continued on their journey.

Before they had gone a few blocks further, the village greengrocer ran into them and greeted them warmly. He looked a bit at Mullah Nasruddin, looked a bit at his wife, frowned, and said, “Mullah! Don’t your feet hurt?” Mullah Nasruddin replied, “Why yes, my feet hurt; in fact, my knees hurt quite a bit.” The greengrocer replied, “Then why have you put your wife—who doesn’t have foot pain at all and is fine, thank God—on the donkey while you walk?” Mullah Nasruddin’s wife felt ashamed. Right then she said, “He is right, Mullah. Let me get down; I’ll walk, and you ride the donkey.” Both agreed and swapped places.

They hadn’t even left the village yet when Mullah Nasruddin’s cousin caught up with them and greeted them warmly. After chatting for a bit, the cousin looked at Mullah Nasruddin, looked at the donkey, and said to Mullah, “Goodness! Do you know how much times have changed? You make your wife walk behind you while you ride the donkey? Look at the poor beast, for God’s sake! Your weight is heavy; you shouldn’t do this. If this donkey breaks a leg, it’s a goner, and your own investment is gone.” Now Nasruddin felt embarrassed. He got down from the donkey, and he and his wife talked it over. Then they looked at the poor donkey and said, “He is right, cousin is right. This old donkey suffers; it has served us, now it’s time for us to serve it.” So both stood side-by-side, picked up the donkey on their shoulders, and walked on toward the neighboring village.

A Tool in Human Hands, or a Human Serving the Tool?

Well, the theme of this parable is about how we shouldn’t listen to what people say, because no matter what we do, others will ultimately judge us. But I want to draw another takeaway from it. In the year 2026, we have a donkey called “Artificial Intelligence” that we are all using. However, some people have started letting this donkey ride them instead of riding the donkey!

When we ask AI to perform a task, we are riding the donkey. But when we allow AI to make decisions for us, it means we have allowed it to ride us. Artificial intelligence should not make decisions for us. AI is not in society; it has no feelings, no human interaction, doesn’t see anyone’s face, and doesn’t observe human beings—it decides purely based on events and data.

Let me give you a good example from an interaction I had with a client. I was doing email marketing work for a client and prepared a format and structure for the campaign. When I presented the proposal, he objected and said, “This structure isn’t good at all, and this is my proposed structure.” When I opened the attached file and saw his proposed structure, it was an out-of-this-world structure better suited for self-branding than business marketing! I asked him, “Did a specialist provide this structure?” He said, “No, I asked ChatGPT, and it gave me a project plan.” He spoke with such pomp and pride that I didn’t dare tell him that ChatGPT lacks human interaction and doesn’t understand marketing psychology! I said nothing and agreed to run the email marketing with his proposed structure.

To explain the structure a bit: the first thing it did was place a huge banner of himself at the top of the email, followed by a quote from himself, and then the rest of the email content structured in a way that didn’t make sense. But the most prominent part was where his photo and a quote of his were placed. At the time, I thought to myself: how can you put a quote from yourself? And the funny thing was that ChatGPT had generated that quote for him too! Even the tone in the email was bizarre.

Let’s assume this friend’s name is Mark. Mark placed his own photo at the top of the email; then, assuming he had an article about computers, it read something like: “Computers will change the world. — Mark”😐😐😐!

Well, honestly, the last time I tried searching for someone’s quotes was related to Nietzsche—and that was a really interesting quote on sociology that I wanted to post on my personal Instagram story. But for an unknown person recognized as a local business owner to put a quote from himself—as if he were a 12th-century philosopher or an Indian ascetic who meditated for 100 years to reach enlightenment, with God revealing all cosmic secrets to him—is a bit bizarre, isn’t it?

I sent the email using his exact structure. The first thing that happened was that because the subject line was very good, the open rate was roughly around 40%, which is extraordinary in its own right and shows his email list trusted his business. But what do you think happened after they opened it? The click-through rate on the purchase link was around 1%, and the unsubscribe rate was 50% of the open rate! Meaning half of the people who opened the email unsubscribed without making any purchase, just by seeing that header and that “precious” quote from him. He lost clients over one small mistake, and this was the result of trusting an AI’s decision.

So I beg of you: ride the AI, and do not let the AI ride you.

I’m Nasrin Bastani, and every week I share articles that make earning with AI a little easier. If you’re interested in marketing, follow my marketing newsletter: The Marketing Notes

Leave a comment

She wanted to learn how to write a book with AI — and told me: “Introduce me to an AI that will deliver a complete book from start to finish.” So far, pretty normal. We have to accept that AI is gradually replacing human work in many areas. But what came next was a whole different world. I asked her why she wanted AI to write it for her. She said: “Look Nasrin, I read in a book that you absolutely must have a book in your field of business. That way your credibility goes up and you get more clients.”

You can probably picture the look on my face — those two seconds of judgmental staring before I tried to pretend I was looking at somewhere else. It is a habit of mine, and by revealing it here I have basically just dug my own grave.

Anyway. I paused for a moment and answered with full conviction: Never. You are too dear to me for me to teach you that.

How Much Should We Use AI?

Look — I love AI. I write about it every week. I use its tools. I read about it every day and keep learning. I tell my clients they absolutely should use AI in their work. I love AI so much that sometimes I joke around with it. But there are places where AI not only does not help — it actively hurts you.

Writing a book for credibility is one of those places.

The Problem With AI and Book Writing

AI writes well. I will not deny that. But “writing well” and “being an expert” are two completely different things.

1- AI’s tone is a whole other world

AI writing has a distinct feel to it. The sentences often sound very similar, following a clear, logical path, but they lack life. It’s like there’s no real experience behind the words. AI can use data, but it can’t truly think. If a book is written by AI, any expert in that subject, or even someone who’s just read a lot in that area, will probably know within the first few pages. Actually, most people will probably tell. You’ll see phrases like “it is worth noting that,” “thus,” “furthermore,” and “in this manner.” (And this paragraph is were that Substack recognizes my text as AI because I’m using those words!)

I’ve said many times about how bad these common expressions are used by AI such as “why is this good?”

Also, why do all the videos have to open with “If you have this issue with x or y?” (Someone save me from the wave of Instagram videos with those AI scripts starting with this sentence😭)

Reading it feels a bit like watching a cheesy commercial for a glow-in-the-dark plastic broom. It’s like an all-night infomercial trying to convince you it’s not just any old mop, complete with twenty-minute interviews from ten random people.

2- AI makes mistakes

AI has lots of mistakes that you can’t imagine. An expert never make these mistakes. I’m talking about wrong dates, invented statistics, fake quotes. If someone with knowledge sees a blinding error in your book, they question your credentials.

3- Overexplaining is the real problem.

AI loves to overexplain everything. It doesn’t matter which platform you use; it repeats every point three times using different words. When you write a book with AI, you end up with a 300-page book that has maybe 80 pages of actual content.

I’m not that strict when it comes to lead magnets because they’re given to the audience for free, and in exchange for their email, you get some valuable information about them. But when someone actually pays for something, they genuinely expect more than a piece of text written by AI.

A Practical Guide to Destroying Your Credibility

If you really want to destroy your credibility, and you insist on getting your name onto Amazon’s list of authors at any cost, I’m going to teach you exactly how to do it. Please follow these steps without wasting any time.

Step One.

Write a prompt: “Write a 200-page book about [your area of expertise].”

You can throw in a few impressive-sounding words to make sure it’s guaranteed to become a bestseller.

Step Two.

Publish the output without any serious editing. You don’t have time — you need to move fast.

Step Three.

Announce on LinkedIn: *“My book is published!”* and ask everyone to buy it. Print business cards and put a QR code linking to the book on them. Promote your book at conferences. Tell everyone you know about it. And if they don’t know you, introduce yourself through your book:

*“Hi, I’m the author of X.”*

In short, put your entire credibility on the line.

Step Four.

Wait for the first expert in your field to spot a mistake and leave a comment under your post. Or, in the best-case scenario, wait for your buyers to leave reviews saying your book was absolutely terrible.

Step Five.

Well, now you have no choice but to explain that *“AI wrote it”* to prove that you’re not actually as unqualified as everyone now thinks.

All done! Now everyone knows that you weren’t willing to spend even an hour a night writing two pages of a book in a field where you claim to be an expert.

Maybe you don’t even have the knowledge in the first place.

Congratulations. Mission accomplished. You officially have zero credibility left. 😁

Where Does a Real Book Come From?

A real book comes from experience. It comes from the nights when you sit down and write two pages, even when you’re exhausted. From the mistakes you’ve made and learned from. From disagreeing with something everyone says is good, while you can give ten pages of reasons why it’s wrong.

This is not something AI can create for you.

Because AI has no experience. No perspective. No sleepless nights. AI doesn’t have real interactions.

A book that builds credibility is a book that only you can write.

Even your competitor can’t write your book, because they don’t have your perspectives or your experiences either.

Where Can AI Help?

This is where I can honestly say that AI is genuinely useful:

• Initial editing: Give your own writing to AI and ask it to find grammatical and structural issues.

• Chapter structure: Use AI to help organize the structure of your book. Give it your idea and ask it to help you come up with chapter titles. Edit them, and then start writing.

• Research: AI can help you find initial sources for you to review yourself — books you should consult before writing, and articles you should read.

• Summarizing: AI can shorten your long paragraphs or provide a summary at the end of each chapter.

• Titles: This pretty little technology can suggest different options for chapter titles.

But remember: these are tools. The pen is still in your hand.

One Last Thing

The writer who spoke with me is a skilled writer. She has years of experience in her field. She has great ideas. She has stories worth telling. Most importantly, she is a good friend of mine. 💖

I told her: “Write your book. Use AI for editing and structure. But the voice has to be yours.”

Credibility is about quality. Quantity doesn’t matter nearly as much.

I’m Nasrin Bastani, and every week I share articles that make earning with AI a little easier. If you’re interested in marketing, follow my marketing newsletter: The Marketing Notes

Imagine this: on Monday you speak with a potential client. The conversation goes well. You say “I’ll send the proposal by Wednesday.” Wednesday arrives. But you now have 400 new emails to answer, two emergency meetings, and a technical problem that appeared out of nowhere. You forget the whole thing completely — and a week later you remember you never sent the proposal.

When you follow up a week later, they have already signed with your competitor.

This is not a story. According to a Salesforce report cited by Convergehub, businesses without a structured follow-up system lose up to 79% of their marketing leads — meaning 8 out of every 10 people who showed interest receive no response.

You are not disorganized. You are not lacking willpower. You have a scattered, chaotic system that needs to be fixed.

Where the Real Problem Is

When we talk about “forgetting to follow up,” we usually think it is a personal organization problem. But it is actually entirely structural.

Most small and medium businesses store client information across multiple places:

• Unsorted emails scattered across an Inbox

• Different email addresses for different platforms

• Phone notes saved in the Notes app

• Multiple Excel files each with a different version

• WhatsApp messages buried in chat history

• Mental reminders that disappear at the first sign of stress

When leads are stored across these scattered systems, follow-ups get delayed, client details get lost, and sales opportunities slip away.

And most importantly: when a client reaches out to your competitor after a few days of silence from you, they do not come back.

I recently worked with a client whose scattered system caused them to lose one of their highest-performing social media accounts — along with all the potential clients attached to it — because they could not remember the email address or password. Just like that. Honestly, I used to have the same problem with my own passwords. But now I have a single unified system, so even if a password goes missing, I know I can recover it easily through the account recovery process.

Three main reasons follow-ups get forgotten:

• Volume: When you receive 50 emails a day, that one important message gets buried on page two

• No smart reminders: Manual notes depend on memory, and memory does not scale

• No prioritization: Everything feels important, so nothing actually gets priority

The Logic of a Follow-Up System That Actually Works

Before we get to tools, we need to understand the principle:

Successful follow-up does not depend on willpower — it depends on structure.

A good follow-up system has three qualities:

1. All information in one place

Every contact, every email, every commitment, every piece of business information needs to be recorded in one central location — not spread across different platforms. The challenge is that integrated platforms are either very expensive or simply do not exist for every use case. For example, you might have a separate email marketing system from your CRM, on top of your regular email inbox. So what do you do?

My answer is this: set up a system that is aware of your workflow and all the tools you use, so you can manage reminders inside it. Think of it like a single box that holds all the scattered pieces — instead of having them spread across the room.

2. Automatic reminders, not manual ones

The system should tell you “it is time to respond to client X” — not require you to scan a list every morning and try to remember.

3. Response at the right time

Harvard Business Review research cited by Convergehub shows that companies responding to leads within one hour are 7 times more likely to have a meaningful conversation with a decision-maker. Timing matters.

This Is Where AI Comes In

In 2026, AI tools can automate a large part of this work — not to remove you from the equation, but to make sure what should not be forgotten, is not forgotten.

How AI helps with follow-up:

According to Instantly, AI tools in 2026 can read your sent and received emails, automatically extract commitments and deadlines, and surface pending follow-ups in a daily briefing — so nothing falls through the cracks.

In plain terms:

• You send an email saying “I’ll send the proposal by Friday” → AI notes it, and if Friday comes without a send, it reminds you

• A client has not responded → after 3 days, AI prepares a follow-up email draft that you simply review and approve

• You had a meeting → a meeting summary and next steps are automatically organized

Tools you can use:

Gmail + Gemini

If your work email runs on Google servers, setting up automated reminders is straightforward. Gemini can connect directly to Gmail and other Google tools like Google Tasks, Google Calendar, and Google Keep. With access to your Google services, it can:

1. Manage and categorize emails: Review your incoming messages, identify important business emails or client requests, and give you a summary

2. Set reminders and tasks: Based on email content, create reminders or tasks — like “respond to client” — directly in Google Tasks or Google Calendar

3. Draft replies: Prepare appropriate responses to client emails so you only need to review and send

To use this, simply ask Gemini to review your recent emails or create a reminder in Google Tasks — for example by writing “review my emails from today” or “set a reminder to reply to this person’s email.”

Ready-made tools:

• Shortwave / Superhuman: AI-powered email assistants for Inbox management, cutting through noise, and setting reminders

• SaneBox: Smart email categorization

• Notion AI: Completely practical for building a simple client follow-up dashboard

Claude Code — for a more custom solution:

Claude Code is one of the most powerful tools that can connect to your system and build reminders. For a more customized solution, Claude Code can build a dedicated automation that connects your inbox, analyzes incoming messages for urgency, and creates follow-up reminders automatically. This requires some technical setup — if you need help building it for your specific workflow, feel free to reach out.

There are three approaches:

Option 1 — Automation tools (simplest, no coding required)
Connect your email service (Gmail / Outlook) through Make.com or Zapier to the Claude API and a task management tool like Google Tasks, Notion, or Todoist. A new email enters Make/Zapier → the text is sent to Claude to determine whether a response is needed and assess urgency → based on the priority rating, a reminder is created in Notion or Google Tasks with the date and a summary of the client’s request.

Option 2 — Custom development with Claude Code / Python
Build a dedicated bot or script that can read emails through Gmail API or IMAP/SMTP, send email content to the Claude API for analysis, and implement a reminder and notification system using services like Google Tasks API.

Option 3 — Ready-made AI email assistants
If you do not want to build from scratch, ready-made tools exist that do exactly this on top of AI models — Shortwave and SaneBox being two strong examples.

Important note: Either buy a system or use it properly. I have seen too many businesses purchase complex tools and never use them. A good system is one that actually gets used. A simple Notion page with 5 columns — client name, last contact, next step, follow-up date, status — beats a complex CRM that never gets opened.

But This Is Not Where It Ends

Let me be honest.

AI can categorize emails, send reminders, and draft messages. These are useful things.

But what AI cannot replace:

• The strategy behind follow-up: When do you follow up? What message do you send? Which lead deserves more of your time?

• Fine-tuning the system around your real business data: Not a generic template

• Ongoing testing and optimization: Which email gets opened more? Which message gets more replies?

80% of deals require at least 5 follow-ups, but 44% of people give up after the first attempt. Knowing this statistic is not enough — you need to know what those 5 follow-ups look like in your business, with your clients.

This is where a professional system — from email marketing to full automation — goes beyond tools and requires expertise.

Take Your First Step This Week

If you want to start right now:

• Today: Make a list of every client you have spoken with in the past 30 days who has not been followed up with

• Tomorrow: Write a short email for each one — use ChatGPT to help it sound like you

• This week: Build a simple follow-up system in Notion or Google Sheets

If you have questions, ask them in the comments below and I will do my best to answer every one.

If you need help with automation or want to talk through what a follow-up system could look like for your business specifically, send me a message. I'm happy to help.

I’m Nasrin Bastani, and every week I share articles that make earning with AI a little easier. If you’re interested in marketing, follow my marketing newsletter: The Marketing Notes

Exactly “.

I understand the reason for this message. Beyond this newsletter, there are hundreds of them that make you believe you will make millionaire incomes with the AI, in addition to billions of posts that tell you to start, start, but, then, what? There are neither of these hundred nor of the billions of posts that explain to you what you could perhaps do on a Tuesday with forty-five free minutes and no idea about it.

What do I reply to them?

“Dear [friend / acquaintance], The Monday issues provide you with the idea.

The professional prompts that actually pay on the market arrive on Wednesday in your inbox. Selling goes to completion on Friday. Your mistake is the fact that you want to take flight with all ideas at once - don’t have 240 hours per day have 24, like the rest of us, I hope -.

Pick one, one only of those listed in front of you, go blindly in the process (system) that makes the sale step-by-step and do not modify any single step of the whole system until you have made your first sale. Only after the first sale you will be allowed to modify, only slightly, then in a very light but real way, continue in this system again, again, always. ”.

She picks the ElevenLabs book to audiobookidea to convert the public domain book into an audio recording using AI and then sell it on Amazon.

We’ve said we will keep her on her toes.

What She Actually Did

It’s not a redesign of strategy. It’s not a “more advanced” version of it. What she did was follow the exact steps we shared: She listened to the market update… before the first tool made it onto the page – she knew why the audiobook market had a supply shortage and that public domain was how we’d get in.

She used the identical voice settings and prompts we’dshared in Wednesday’s training to immediately produce clean, natural sounding audio on the first actual attempt.

She chose a public domain title and processed it through the Friday production workflow piece by piece. She took care of the ACX legal disclosur requirement. Her audiobook was published. That’s it.

Not shortcuts.

No “I’ll figure this one out on my own. That whole process took ten days from when she reached out to the day after her first sale was recorded – a small number at best, but a sale of a book she’d only started working on two weeks ago with tools she’d never seen or heard of. I don’t share it because a tiny commission transformed her.

I’m sharing it to show you the entire system does in fact work - and has been for a long time, whenever you’rewillingto actualy applyit, instead of skimmingsavingitthenjumping to the next new idea.

What This Actually Looks Like Over Time

Now, let me make this abundantly clear from the beginning: I’m NOT going to make me a million. You’ll likely end up with supplemental income. Supplemental income which might take years or months to reach even, say, $30/month on average.

We might get up to where you will not have trouble leaving your nine-to-five on year three if everything goes perfectly, or the side income may rise, slowly, year on year to $400 a month, but still that is extra money you will buy things with.

I don’t have plans for the net income to replace me taking a salary from me current job. Yet still, the extra funds shall pay for the luxury of actually going on trips and acquiring those pricey novels you insist you must take with you and all that stuff.

Let’s just summarize things. We have been doing all this since April 2, 2008 but for her this is just beginning; it may seem very unlikely to someone on the outside that this takes work, Elena will surely share her observations on building a site and everything later on!

Why I’m Telling You This

Every Monday I give you an idea. Every Wednesday, the direct prompts. Every Friday, the selling system. That’s the whole machine, and the machine doesn’t crank anything out until someone actually uses it.

Elena wasn’t more genius than you, nor more technically skilled. She simply realized that one idea pursued diligently was worth far more than ten pursued half-heartedly.

If you’re just consuming and bookmarking and not doing the thing, this is for you: Go dig through the archive, find one issue, and follow the plan exactly. Don’t deviate one iota until you have ten sales.

And if you’re looking to hold all of the pieces of this in your hands-prompts, workflow, sales system, everything-and know that I’ll show you exactly what to do when you inevitably get stuck-well, that’s what the paid plan does. It separates the folks who are browsing from the folks who are actually building. Paid subscribers stick with it because they’ve got skin in the game and they want far more than the issues: they want the community to share with each other, and they get direct access to me.

Subscribe now

I’ll keep sharing these stories every Sunday — real people, real numbers, real proof that this works when you work it. If you made your first sale by following something from this newsletter, reply and tell me. You might be the next story I share.

— Nasrin

This is the issue that makes everything else actionable.

Subscribe now

Read more

Every prompt in this issue is ready to copy. Every step is in order. By the end of this issue you will have everything you need to produce your first sellable invitation design today.

Subscribe now

Read more

Wedding invitations are not an afterthought. For most couples, they are a decision that takes weeks. They browse Etsy for hours. They save hundreds of pins on Pinterest. They want something that feels exactly right — exactly them — and they are willing to pay for it.

The wedding stationery market is worth over $8 billion globally. It does not slow down. People get married every single day, in every country, in every season. Birthdays happen every day. Engagements happen every day. Baby showers, anniversaries, quinceañeras, retirement parties — every one of them needs an invitation, and every one of them represents a buyer who wants something beautiful.

Until recently, creating that something beautiful required design skills, expensive software, and years of practice. AI changed that. Completely.

Subscribe now

What’s Actually Being Sold

Digital invitation sets. Not printed cards — digital files that the buyer downloads, customizes with their details, and either prints at home or sends directly to guests via email or messaging apps.

The digital format matters for three reasons.

First, there is no production cost. You design once, upload once, and the file sells indefinitely. The hundredth buyer costs you exactly as much as the first — which is nothing.

Second, the market has fully embraced digital. Couples who would never have considered a digital invitation five years ago are now actively seeking them — partly for convenience, partly for sustainability, partly because digital invitations can do things printed ones cannot. Animation. Music. RSVP links built directly into the design. A QR code that opens a wedding website.

A well-designed digital wedding invitation set on Etsy sells for between $8 and $35 for the template. That sounds modest until you understand that the same file sells to hundreds of buyers. A listing with 500 sales at $15 has generated $7,500 from a single design created once.

Premium fully customized sets — where the buyer pays for personalization, not just the template — sell for $50 to $200 per order. These take more time but attract buyers with higher budgets and lower price sensitivity. Wedding season buyers are not shopping for the cheapest option. They are shopping for the right option.

The Occasions — It’s Not Just Weddings

Weddings are the anchor of this market, but they are not the whole market. Every occasion that requires an invitation is an opportunity.

Wedding suites — invitation, RSVP card, details card, envelope liner, thank you card. The full set commands the highest prices and the most repeat purchases as couples buy additional pieces.

Engagement party invitations — often the first purchase a newly engaged couple makes. High emotion, high intent, high willingness to spend.

Birthday invitations — children’s birthdays in particular. Parents spend more on their child’s birthday party invitation than most people realize. Themed designs — dinosaurs, princesses, superheroes, florals — sell in consistent volume year-round.

Baby showers — strong seasonal spikes, highly giftable aesthetic, repeat buyers as family members and friends host multiple showers over the years.

Bridal showers and bachelorette parties — often coordinated with the wedding aesthetic. A buyer who purchases a wedding invitation template frequently comes back for matching shower invitations.

Corporate events and milestones — retirement parties, company anniversaries, product launches. Less emotionally driven but higher volume and often higher budgets.

Each of these occasions has its own aesthetic language, its own buyer, and its own price point. A single Etsy shop covering three or four of them is covering a market that generates purchases every single day of the year.

Why AI Makes This Accessible Now

Designing a wedding invitation from scratch used to require Illustrator, InDesign, or Photoshop — professional software with steep learning curves and steeper price tags. The barrier kept this market tightly held by trained graphic designers.

AI tools have dismantled that barrier almost entirely.

With the right prompts, AI image generators produce the decorative elements — florals, watercolor washes, geometric patterns, calligraphic flourishes — that form the visual foundation of most invitation designs. Canva, which is free and requires no design training, handles the typography, layout, and text placement. The combination of AI-generated visuals and Canva’s drag-and-drop interface produces results that compete directly with professionally designed templates — at a fraction of the time and cost.

The skill that matters now is not technical. It is taste. The ability to recognize what looks beautiful, what fits a particular aesthetic, and what a specific buyer is looking for when they type “boho floral wedding invitation” or “modern minimalist black and white save the date” into the Etsy search bar.

Taste is something most people have more of than they realize. And it gets sharper with practice.

Share

The Seasonal Opportunity Nobody Talks About

Wedding season creates predictable spikes. Spring and early summer — May through July — account for the largest volume of weddings in North America and Europe. The months before those dates — January through April — are when couples are actively planning, which means invitation shopping peaks in winter and early spring.

This is not a reason to wait until January to start. Etsy’s algorithm rewards shops that have existing sales history and reviews when peak season arrives. A shop that launches in October with ten well-designed listings and accumulates its first fifty sales by December is positioned to capture January’s surge. A shop that launches in January is starting from zero when everyone else has a head start.

The time to build is before the demand arrives.

What’s Coming This Week

Today you understand the market, the opportunity, and why AI makes this possible for someone without a design background.

This Wednesday — Paid Issue:
The exact AI prompts for generating wedding and occasion invitation designs across every major aesthetic — boho, modern minimalist, vintage, tropical, rustic, luxury — plus the Canva workflow that takes your AI-generated elements and turns them into a finished, sellable product. Every prompt, every step, in order.

This Friday — Paid Issue:
The complete selling system — how to set up your Etsy shop for this specific product category, how to write listings that rank for the searches buyers are actually making, pricing strategy for templates versus custom orders, and the action plan for your first ten sales.

Subscribe now and get both issues this week — plus the full archive of every prompt, workflow, and income strategy we’ve published.

[Subscribe here →]

Subscribe now

Monday we looked at why the market exists and who’s already buying.

Wednesday we went inside the tools — the exact prompts that produce professional results across every major musical style.

Friday we covered the complete system for turning this skill into income — where to sell, how to find customers, and how to deliver professionally.

Today: you get to hear it.

The Sample

What you’re about to listen to started as a poem. Four lines. Written by hand. Nothing special about the process — just words that carried feeling.

What came out the other side is this.

No studio. No composer. No singer in a recording booth. Just a prompt, a tool, and about ten minutes.

What This Means for You

If you have a poem — something you wrote, something a loved one wrote, something tucked away in a notes app that never went anywhere — this is what it could sound like.

And if you want to build a service around doing this for other people, this week’s three issues cover everything you need to know to start this week.

The tools are free to try. The market is real. The first order is closer than it feels.

Not subscribed yet? Free subscribers get Monday’s ideas. Paid subscribers get Wednesday’s prompts and Friday’s complete systems — everything you need to go from idea to income.

[Subscribe here and don’t miss next week’s series →]

Everything in this issue is in order. Follow it from top to bottom and you will have a functioning business by the end of the week.

Part 1 — Where to Sell

Not all platforms suit this service equally. Here is the honest breakdown.

Read more

When you’re done reading this week’s newsletter, you’ll know precisely what you should be typing into each tool, where you should be typing it, and what to expect when you do.

Read more

Now picture that same poem — with melody, instruments, a real singing voice — turned into an actual song. Something that can be played, gifted, sold.

That’s exactly what we’re talking about today: turning poems into songs with AI — one of the least crowded, most emotionally resonant digital markets taking shape right now.

Why Now

Until just a couple of years ago, turning a poem into a song meant a studio, a composer, a singer, mixing, mastering — costs completely out of reach for a regular person. Today, free AI music tools let you drop in the lyrics, pick a style, and have a complete song with vocals and instrumentation in minutes.

A door that used to open only for professional composers is now open to anyone — and almost nobody is using it as a business yet.

Who’s the Market

Unlike wallpapers or printables, this idea has one key difference: the final product is completely personalized. Instead of selling one ready-made file to a thousand people, you’re selling a service built fresh for one person each time. That comes with real advantages:

• Higher price point — because the product is unique, not a repeated file

• Less competition — most AI sellers chase bulk products, not custom work

• Strong emotional attachment — customers ordering a gift or a memorial piece care far less about price

Core audiences:

• People wanting a unique gift for a partner (anniversaries, Valentine’s Day, proposals)

• Families who want to turn a poem or memory from a lost loved one into a song

• Amateur poets who have verses tucked away and want to actually hear them

• Content creators and video makers looking for original, royalty-free music

Turning It Into a Business

The formula is simple: take the customer’s poem, feed it in with the right prompt, generate a few versions in different styles (pop, folk, classical, lullaby, epic), pick and polish the best one, then deliver it as an audio file with album art.

You can sell this as a custom service on Etsy or Fiverr, or build a “gift song” package for special occasions (anniversaries, birthdays, proposals) priced well above typical digital products.

Why We’re Covering This Now

Wednesday we get into the practical side — several free tools you can use, and the exact prompts that get the best results out of each one, whatever style you’re going for.

Friday we cover the full sales process: where to find customers, how to price it, how to take and deliver orders, and exactly who this service is for.

This week you’ll learn how to turn an old art form — poetry — into a modern digital product that carries both meaning and money.

Wednesday: Free tools compared, plus the exact prompts to get the best results from each

Friday: The complete sales process — from taking the order to final delivery

[Subscribe to catch every part of the series →]

Subscribe now

This is the issue that turns everything you’ve built into a business.

Read more

Every prompt in this issue is ready to copy. Every step is in order. By the end of this issue you will have everything you need to generate your first sellable wallpaper pack today.

Read more

Someone designed that. Someone chose those colors, that composition, that mood. And if you’ve ever gone looking for a wallpaper that felt exactly right — the right aesthetic, the right vibe, the right fit for your screen — you know how long that search can take.

That search is happening millions of times a day. And the people selling the results are making real, consistent, passive income from digital files that cost nothing to ship, nothing to store, and nothing to restock.

This is one of the quietest opportunities in the AI income space right now. And it’s wide open.

What’s Actually Being Sold

Phone wallpapers. That’s it.

Not physical prints. Not framed art. Not products that need to be manufactured or shipped. A digital image file — downloaded instantly, used immediately, zero overhead on your end after the initial upload.

Buyers search for wallpapers the same way they search for anything else — by aesthetic, by subject, by mood. Anime wallpapers. Minimalist wallpapers. Cyberpunk city wallpapers. AMOLED black wallpapers for Samsung. Botanical wallpapers for iPhone. Dark academia. Cottagecore. Y2K. Retro gaming.

Every one of those search terms has buyers behind it. Active, repeat buyers who update their wallpapers regularly and come back for more when their mood changes.

Why AI Changes Everything About This Market

Before AI image generation, creating wallpapers required design skills, software licenses, and significant time investment. The barrier was real enough that the market stayed relatively small and relatively niche.

AI removes that barrier completely.

Tools like Midjourney, Flux, Stable Diffusion, and Ideogram produce stunning, original images in seconds. The quality — when prompted correctly and upscaled properly — is indistinguishable from professionally designed digital art. And the variety is limitless. Any aesthetic, any subject, any color palette, any mood — generated on demand.

What used to take a designer hours now takes minutes. What used to require technical skills now requires a well-written prompt and an understanding of what sells.

That shift creates an opportunity. The market for wallpapers is not going away — if anything it’s growing as phone screens get larger and higher resolution and people become more intentional about how their devices look and feel. But the cost of production has dropped to almost nothing.

What the Numbers Look Like

A single wallpaper pack — say, 20 wallpapers in a cohesive aesthetic — sells for between $3 and $8. That sounds modest until you understand that digital products have no marginal cost. The hundredth sale costs you exactly as much as the first sale — which is nothing.

A listing that sells five packs a day generates between $15 and $40 daily. From a file you created once and uploaded once.

Sellers with larger catalogs — 50, 100, 200 listings across different aesthetics and device types — report monthly incomes ranging from a few hundred dollars to several thousand. The range is wide because the variables are real: niche selection, listing quality, thumbnail design, and how well the product matches active search demand.

But the floor is accessible and the ceiling is genuinely interesting.

The Aesthetics That Are Selling Right Now

Not all wallpaper niches perform equally. The ones with consistent, proven demand include:

AMOLED and dark mode wallpapers — Samsung users in particular seek these out because true black pixels on AMOLED screens save battery life. This is a functional purchase as much as an aesthetic one, which means buyer intent is high.

Anime and gaming aesthetics — passionate communities with strong visual identities and buyers who update their wallpapers constantly.

Minimalist and clean designs — broad appeal, works across iPhone and Android, consistent search volume year-round.

Cyberpunk and sci-fi — visually distinctive, AI image generators produce exceptional results in this style, and the aesthetic has a dedicated and growing fanbase.

Nature and botanical — calm, broadly appealing, works well across demographics and for buyers who want something beautiful rather than something trendy.

Seasonal and holiday packs — search volume spikes predictably. A Christmas wallpaper pack uploaded in October generates significant income for two months before becoming passive income year after year.

Why This Is Different From Selling Art

Selling art requires taste, originality, and often a personal brand behind the work. Buyers of fine art prints are buying the artist as much as the image.

Wallpaper buyers are buying the aesthetic. They want something that looks good on their screen. They don’t particularly care who made it or how. This is not a criticism — it’s a commercial reality that works in your favor. You don’t need to be known. You don’t need a following. You don’t need a brand story.

You need images that look good on a phone screen and a listing that shows up when people search for them.

That’s the whole model.

What Paid Subscribers Are Getting This Week

Today you know the opportunity is real and the market is active.

This is where the free issue ends — and where the real work begins.

This Wednesday — Paid Issue:
The exact prompts for generating wallpapers that sell — broken down by aesthetic and device type. The upscaling workflow that takes your images from good to 4K and 6K professional quality. How to build a cohesive pack, format your files correctly, and design the thumbnail that stops someone mid-scroll and makes them click. This is the production system — everything you need to go from zero to a finished, ready-to-sell product.

This Friday — Paid Issue:
The complete selling strategy. Which platforms welcome AI-generated wallpapers and which ones will reject your submission. How to write listings that rank in search and convert browsers into buyers. Pricing strategy for single wallpapers versus packs versus bundles. The exact workflow from your first generated image to your first sale — every step, every decision, every number.

Two issues. The complete picture. Everything you need to turn this into actual income this week.

Subscribe now and get both issues — plus the full archive of every prompt, workflow, and income strategy we’ve published.

Subscribe now

There are more ways to monetize AI-generated audio content than most people realize — and the smartest creators aren’t picking just one. Here is every realistic option, what each one requires, what each one pays, and how to decide which combination makes sense for where you are right now.

Read more

There’s a gap between what most people produce with ElevenLabs on their first attempt and what the tool is actually capable of. That gap is not talent. It’s knowledge. Here’s what you need to know.

Read more

That barrier just disappeared.

ElevenLabs is an AI voice generation platform that produces audio so natural, so expressive, and so human-sounding that most listeners cannot tell the difference. You paste in text. You choose a voice. You hit generate. What comes out the other side sounds like a professional narrator who has been doing this for twenty years.

And people are already selling what it produces — on Amazon, on Google Play, on podcast platforms — and making real money doing it.

What ElevenLabs Actually Does

At its core, ElevenLabs converts text to speech. But that description drastically undersells what makes it different from every text-to-speech tool that came before it.

Previous generations of text-to-speech were recognizable instantly. The flat delivery. The mispronounced words. The robotic rhythm that made everything sound like a GPS giving directions. Nobody was going to pay to listen to that.

ElevenLabs produces something categorically different. Its voices have natural pacing. They breathe. They emphasize the right words. They handle dialogue, emotion, and complex sentence structures in a way that sounds considered rather than mechanical. There are voices that sound authoritative, voices that sound warm, voices that sound young or old, male or female, accented or neutral.

You choose the voice that fits your content. You paste in your text. The audio is ready in seconds.

The Tools Worth Knowing About

ElevenLabs is the market leader — but it’s not the only option, and depending on your budget and needs, alternatives are worth considering.

ElevenLabs starts at $5 per month for the Starter plan, which gives you 30,000 characters per month — enough for a short book or several podcast episodes. Higher plans unlock more characters, more voices, and commercial licensing. The quality at every tier is exceptional.

Revoicer is a strong alternative with a one-time payment option rather than a subscription — appealing if you want to avoid recurring costs. The voice quality is solid, particularly for straightforward narration, and it handles long-form content well.

Murf AI offers a free tier with limited characters and a selection of voices. Good for testing and short-form content. The paid plans are competitive and the interface is clean and easy to use.

Speechify started as a tool for people with reading difficulties and has evolved into a full voice generation platform. The free version is functional. The premium voices are genuinely impressive for the price.

TTSMaker is completely free with no account required for basic use. The voice quality doesn’t match ElevenLabs, but for someone just starting out who wants to experiment without spending anything, it’s a legitimate starting point.

The honest recommendation: start with TTSMaker or Murf’s free tier to understand what AI narration sounds like and whether it fits what you’re trying to build. If the answer is yes, ElevenLabs is worth the investment.

Why Audiobooks and Podcasts Are the Opportunity

The audiobook market has a supply problem. There are millions of books — particularly older titles — that have never been recorded. There are millions of potential listeners who prefer audio to reading. And there are very few people doing the work of connecting those two things at scale.

AI narration changes the economics of that equation completely.

Before ElevenLabs, recording an audiobook meant either narrating it yourself — hours in a quiet room, expensive equipment, significant editing time — or hiring a professional narrator, which runs anywhere from $200 to $400 per finished hour of audio. A standard novel is eight to twelve hours of finished audio. Do that math.

With ElevenLabs, the production cost drops to the price of a monthly subscription. The time cost drops to however long it takes to paste text and hit generate. The barrier that kept most people out of the audiobook market effectively disappears.

And then there’s podcasting. Subscription podcast platforms — where listeners pay monthly for exclusive audio content — are a growing and underserved market. AI-generated audio makes it possible to produce consistent, high-quality podcast content without the production overhead that kills most shows before they find their audience.

The Public Domain Angle

Here’s something most people don’t know: thousands of books are legally free to use, reproduce, and sell in any format you choose.

Books published before a certain date — the specific year varies by country, but in the United States it’s generally works published before 1928 — have entered the public domain. Their copyright has expired. Nobody owns them. You can take the text, narrate it with ElevenLabs, and sell the audio version without paying a single cent in licensing fees.

Sherlock Holmes. Jane Austen. Mark Twain. H.G. Wells. Edgar Allan Poe. F. Scott Fitzgerald. These are not obscure titles people have never heard of. These are books people actively search for, actively want to listen to, and actively purchase in audio format every single day.

The text is free. The production cost is minimal. The market is real and established.

That combination is rarer than it sounds.

One Important Thing to Know Before You Start

Platforms have different rules about AI-generated audio, and those rules are evolving.

Amazon’s ACX platform — the main route to Audible distribution — currently requires disclosure if AI voices are used in production. This is not a reason to avoid it. It’s a reason to be informed before you submit. Google Play Books has a different set of requirements. Spotify for podcasters has its own policies.

The rules exist. They’re navigable. But going in without knowing them wastes time and risks having your content rejected after you’ve already done the work.

Knowing the rules before you start is the difference between a smooth launch and a frustrating one.

What’s Coming This Week

Today you know what ElevenLabs is and why the opportunity around it is real.

This Wednesday: the exact settings, voice selection process, and prompt structure that produces professional-quality narration on the first attempt — not after hours of trial and error. The specific adjustments that make AI voices sound human in the moments most tools get wrong.

This Friday: the complete workflow for taking a public domain book from text to finished audiobook to live listing on a distribution platform — every step, every decision, every number you need to know before you start.

Subscribe now and get both issues this week — plus the full archive of every prompt, workflow, and tool breakdown we’ve published.

Subscribe now

Every episode you record contains thousands of words of valuable insight, stories, and expertise. But most of that value disappears the moment the episode ends. Your listeners hear it once. Nobody else ever finds it. And you move on to recording the next one.

A blog post changes everything. It makes your content searchable. It brings in readers who will never listen to a podcast. It builds your SEO over time. And with the right prompt, it takes about ten minutes to produce from any episode you’ve already recorded.

Here’s exactly how to do it.

Read more

The answer is yes. But here’s what nobody tells you: ChatGPT doesn’t make you money. You make money. ChatGPT just cuts the time it takes from hours to minutes. That’s the real opportunity.

Here are five realistic ways to make $500 this month.

1. Freelance Copywriting — The Fastest Path

Businesses need words constantly. Landing pages, email sequences, product descriptions, social media captions, About pages. Most small business owners either hate writing or don’t have time for it — and they will happily pay someone else to handle it.

Here’s what most people miss: you don’t need to be a great writer to sell copywriting services. You need to be a great editor. ChatGPT produces the first draft in seconds. Your job is to refine it, make it sound human, and deliver it on time.

Where to find clients: Upwork, Fiverr, or cold outreach to local businesses whose websites look like they haven’t been touched since 2015. Those businesses are everywhere, and they need help.

What to charge when starting out: $50 to $150 per piece. Five pieces this month gets you to $500. As your portfolio grows, so does your rate.

The key is speed. With ChatGPT, you can deliver a polished 400-word landing page in under an hour. That’s a strong hourly rate by any standard.

2. Resume and LinkedIn Rewriting — High Demand, Easy Delivery

The job market is competitive. People are desperate for resumes that actually get them interviews, and career coaches charge $200 to $500 for this exact service. You can offer the same result at $75 to $100 and still make excellent money for the time invested.

One client takes about 45 minutes with ChatGPT if you have a clear process. You take their existing resume, ask them a few questions about the role they’re targeting, and produce something dramatically better.

Where to find clients: local Facebook groups, LinkedIn, Reddit communities focused on job searching, or even Kijiji. Post a simple offer. You’ll get responses faster than you expect.

Five clients at $100 each = $500. This is genuinely one of the most accessible services you can offer right now.

3. Social Media Content Packages — Recurring Income

This is where things get interesting, because the income repeats.

Small businesses know they need to post consistently on social media. Most of them don’t. Not because they don’t want to — because they run out of ideas, run out of time, or both. You solve that problem for them.

Sell a monthly content package: 20 captions, hashtags included, delivered as a Google Doc at the start of each month. Charge $100 to $200 depending on the client. Land three clients and you’ve passed $500 — and next month, you do it again with the same clients.

The pitch is simple: “You focus on running your business. I’ll make sure you never miss a post.” That’s a message that resonates with every small business owner you’ll ever meet.

4. Etsy Digital Products — Passive After Setup

The previous three options require active work for each dollar you earn. This one is different.

Digital products on Etsy sell while you sleep. You create the product once using ChatGPT and Canva, list it on Etsy, and every sale after that is pure profit with zero additional effort.

The catch is that it takes longer to see results. You might sell nothing for the first two weeks. But a product that finds its audience keeps earning for months or years.

Price your products between $5 and $9. Focus on a specific niche — the more targeted, the better. A generic “budget planner” competes with thousands of listings. A “budget planner for single moms going back to work” speaks directly to someone and stands out immediately.

Sixty to one hundred sales gets you to $500. Sounds like a lot — until one of your listings starts getting organic Etsy traffic.

5. Sell Your Own Expertise as a Prompt Package

This one is underrated and almost nobody talks about it.

You have knowledge and experience in your field. A teacher knows what teachers need. A nurse knows what nurses deal with daily. A marketing professional knows the exact questions clients ask. That expertise, turned into a well-crafted set of prompts, is something other people in your field would pay for.

Package 15 to 20 prompts built around your own profession or area of expertise. Not generic prompts — your prompts, built from your real experience. Sell them on Gumroad, Etsy, or directly through your newsletter for $15 to $29 per package.

Twenty sales at $25 = $500. And unlike the other options on this list, nobody else can replicate what you know.

The Honest Truth

None of these will make you $500 if you spend this week reading about it instead of doing it.

ChatGPT doesn’t make you money. Action does. ChatGPT just makes the action faster, cheaper, and more accessible than it’s ever been in history. The barrier is gone. The only thing left is you deciding to start.

Pick one method from this list. The one that matches what you already know, what you’ve already done, or what you’re already curious about. Don’t pick the most exciting one. Pick the most realistic one for where you are right now.

Start this week. Not next Monday. This week.

Every week in Earn With AI, we go all the way on one method — step-by-step workflow, exact prompts, and a clear action plan so you know precisely what to do next.

This week in the paid issue: how to create and sell wall art printables on Etsy using AI image generators. The exact prompts for botanical prints, abstract art, and kids room illustrations — plus the technical requirements, pricing strategy, and your Etsy listing written for you.

The archive is waiting. Every method. Every prompt. Every week.

Subscribe here and start this week