The world demands productivity and perfection. The post-englightenment age requires us to be striving for more, better, and faster. We’ve achieved so much through this directive: surpluses of everything, modern medicine, needs and demands that can be attended to in hours or less. With AI, creation is cheaper than ever before, and my biggest fear is we forget the struggle and work that goes into creating beauty. I have written about my usage of AI in the past. There are many systems and workflows I have built to accelerate my effectiveness. But one thing I cannot do is pretend that AI can create art.

I know I am not alone in my fears. In the context of my day-to-day work, I know that Brand is a critical asset for B2B businesses. Brands require human connection, which is impossible to create with generative AI, technology which has never taken a breath, has never loved, had never experienced pain. It’s telling that brand teams are now hiring or promoting leaders into role titled “Head of Storytelling” (Rippling), and “Head of Narratives” (Langchain).

They’re no longer looking for folks to run “Content”. Content signals Context which signals AI, and performance-driven results. Storytelling is a two-way experience, a practice that relies on ritual.

My take: Marketing is going to evolve and the best marketers are going to focus on maintaining the ritual of artistic practice.

On the Evolution of Artistic Practice

With the advent of photography, painting had to evolve to become more conceptual. Fine art was focused on the ideas and feeling it could emote rather than the technique. Materials, their sources, the craft of finding and manipulating them, became a story worth telling.

One of my favorite pieces that is representative of this change is Tallow, a set of sculptures by Joseph Beuys, a German sculptor from the 70’s, and it’s literally a set of blocks made out of beef tallow. Tallow melts in certain conditions, so this piece transforms in different environments.

Beuys told a fable along with this piece: when his plane crashed during World War II, his rescuers saved him by rolling him in Tallow.

This story is not true, but it doesn’t really matter, does it? That’s what art is. It’s a series of rituals and stories that push us to relate to the world.

As technology changed, we found a new way to represent authenticity. We found a new way to tell others how and why we felt.

I think we are in the early phases of this with AI and our art. Many of us have felt the anger and frustration of reading words written by a machine. Like nails on a chalkboard, we cringe as we scroll past LinkedIn and Substack posts written with the support of an LLM. I think we’re in the painful, congestive part of the cycle where many people have been given access to this wildly popular technology and are trying to figure out its edge.

I think the part that makes us cringe is using it to be someone they’ve always wanted to be! But that’s the problem. By trying to fulfill their purpose, they’ve alienated themselves from their authenticity.

They’ve started chipping away at their aura.

Subscribe now

What is Aura?

Walter Benjamin’s seminal work “The Work of Art in the Age of Mechanical Reproduction” — which is not referenced nearly enough in discussions on AI and creativity — discusses the nature and purpose of art once it can be created or assisted by technology. He was living in Germany during the rise of the Third Reich, where film was being used to support a Fascist agenda. His concern was that the ability to reproduce art at mass scale can slowly remove the authenticity of a piece.

I believe these observations are crucial for our time, because this is the reason so many of us dislike reading AI-authored content. It sounds like nothing because it has no human experience!

By introducing technology that helps us mass produce content, we’re taking the art away from its original source, the aura. This essay stuck with me when I first read it in my — cringe — my undergraduate seminar, “Marx and Marxism”. This essay builds on Marx’s idea that alienating a person from their means of production is a core tenant of the human experience under captialism, and in the case of art, alienation of creation through technology slowly moves a person away from their authentic self. As humans move away from the creation and consumption of art with aura, they become less capable of communicating their core, authentic feelings.

History has shown us that there is art in photography. There is an art to capturing a scene and a feeling, and a certain genius has emerged from it. But it is not the same as the human experience of appreciating beauty without a lens.

If, while resting on a summer afternoon, you follow with your eyes a mountain range on the horizon or a branch which casts its shadow over you, you experience the aura of those mountains, of that branch.…the desire of contemporary masses to bring things ‘closer’ spatially and humanly, which is just as ardent as their bent toward overcoming the uniqueness of every reality by accepting its reproduction. Every day the urge grows stronger to get hold of an object at very close range by way of its likeness, its reproduction.

Walter Benjamin

And some compelling AI art has emerged as well — one example is Sheila Heti’s pieces in the New Yorker where she converses with a bot of her creation, Alice. There’s likely dozens more examples like this. But one thing that we can all agree on, AI writing sounds odd. It doesn’t feel good to read it. Substack even started tracking whether essays were written with AI — because 1, we hate reading them and 2, we don’t want to pay for AI-generated garbage.

What I love reading is your voice. By you I mean the Royal you. I appreciate sloppy and imperfect writing. I love writing that has personality. That leaps. It’s one of the reasons I love reading Substack. I am discovering so many talented writers who have never had a career as a writer, but who introduce me to their lives, thoughts and ideas. It’s one of the reasons I started writing on Substack myself. I wanted to flex my creative muscles and connect with other people.

How to Maintain Your Aura

I thought about my aura a lot when I started using AI more frequently at work. Or specifically, how I could continue to build on my cognitive strengths while automating so much of my work. The way I use AI is for operations and orchestration, but I also use AI to take and summarize meeting notes, write Slack messages or construct processes. I am by no means an expert, but ere’s some practices I’ve found helpful to keeping myself creating my own work, and refining my own voice.

Create

In my opinion, the most radical thing we can do these days is create. This can mean knitting – which I’ve learned is Art x Math – or writing a poem. I’ve begun The Artist’s Way, a popular (and effective) 12-step program for re-igniting creativity, multiple times and never finished. But one of the practices find their way into my rituals: Morning Pages. In the custom of “get the frog out of your throat”, Morning Pages devotees will, upon waking in the morning, and before they look at their phones, write three pages of whatever comes to mind in their journal.

Subscribe now

These are private thoughts, and private words. You’re not meant to show this to anyone, making it very non-judgemental practice. Either you power through or you don’t!

I like Oliver Burkeman’s perspective on Morning Pages:

Since the only fixed demand is that you fill three pages, Morning Pages demonstrates the principle that quality is often a function of quantity: there’s no time or incentive to judge each idea, to get it right before getting it down, and the result is that ideas flow more freely, unimpeded by the clenched grip of perfectionism. There are far more of them, and sometimes a few of them are good.

This is a very difficult practice for me to stick to as a mom of a young child and two dogs, but I do it when I can. When I can’t, I just wait until after daycare drop off, or whenever I can get a break. At first it’s hard to write three full pages, but eventually it comes out very easily. It’s very similar to an endurance sport. Writing three pages a morning requires and builds creative endurance.

Attribute When Necessary

If AI does any content-creation on my behalf, I attribute it to AI. I do this for fear that I might get caught, but also because I don’t want anyone to think I wrote that! AI is great at some things and not good at others. I recently had to map out processes that were the result of many conversations, interviews and feedback loops. The final document was authored by me, but AI was much better at summarizing and generating steps based on the hours of content I had recorded. I gave Claude the credit.

How do you keep up a ritual of creation in the age of AI?

I’m not always the sharpest crayon in the box, clearly.

A friend was looking for concrete benchmarks on paid subscriber retention.They asked what percentage of paying self-serve customers typically make it from Month 1 to Month 2.

The context: The current setup they described:

• They specifically wanted to know if something around 50% retention was normal (they mentioned one company they used to work for started at 50% and considered it pretty bad).

• They also asked whether the benchmark should be viewed on a monthly or yearly basis.

• They clarified that this question was for a non-contracted, self-serve platform, and shared that their self-serve customers spend around $250/month on average. ($3k ACV)

My response:

This question seemed pointed heavily at logo retention, aka the sheer count of customers that stick around after a period of time. For a PLG product with a usage-based lever (this is an observability product), there will be a very dramatic mix in the logos that come onboard and their spend. In many of these scenarios, 20% of the customers drive 80% of the revenue. A better measure of self-service effectiveness is net dollar retention (NDR) . Here’s how I measure NDR.

I dug in a little more with my friend. What did the actual usage-based revenue coming in the door look like after month 1? A year? Per my developer gtm playbook she can consider anything over 110% NDR pretty good, with 140% best-in-class.

This friend was spiraling because in the short term she had some logo churn—users were signing up and kicking the tires but not making it to the first or second month of using the product. But, when she pulled back and looked at the bigger picture, those $3k ARR accounts were expanding well, quite well in fact, some folks were ramping up spend so that NDR was as high as 120% at month 3–well on her way to best-in-class. We then went on to discuss how to lock those accounts into contracts, which is a subject for another newsletter entirely.

That’s it for this week’s batch. If you’ve got a gnarly growth question you’ve been texting your friends about, send it my way — I might feature it next time.

Exceptional product marketers, according to Sam, are future C-suite execs (like two of my friends, Leo Zheng, CMO at Fireworks.ai and Graham Neray, founder and CEO of Oso). They identify opportunities, drive cross functional projects, define the GTM strategy according to the market, and are powerful allies to the leadership team. In this post, we’re going to define the rubric for identifying this caliber of product marketers.

As my guest today shares, the most talented product marketers are pillars of the organization. Clarke Patterson has led Product Marketing teams at dbt Labs and Snowflake where he’s worked on product strategy, messaging, brand, campaigns, sales enablement and pricing. I worked with him at dbt Labs and can attest he attracts incredible talent, recruiting some of the best technical leaders over to marketing and bringing in great external hires.

During our conversation I walked away with one lesson: Clarke focuses on hiring leaders regardless of role or their span of control in the organization. He’s hiring people who innately take on a strategic role in their organizations. He also admits that this helps him build a strong bench of talent to elevate when there’s an opportunity. To quote him: “Great leaders aren’t afraid to hire themselves out of a job. They build teams where there’s an obvious succession plan.”⁠

From our conversation, I walked away with stronger rubric for hiring strategic PMMs.

• Scrappiness

• Passengers vs. Drivers

• Locus of Influence

Scrappiness

Naturally, it’s ideal to bring in candidates who are familiar with your domain. Those who understand the product and the go-to-market motion could integrate more easily into the organization and get ramped more quickly. Regardless of domain knowledge, you need to look for folks who are proactively educating themselves to become more of an expert. Exceptional product marketers will always be learning and bringing themselves closer to the customer’s perspective.

“More often than not, you’re not going to get the person who already has perfect domain knowledge. So the question becomes: how do you spot the people who are just scrappy learners?”

This is important not just for learning about the product, but also for figuring out those key opportunities for driving and leading. “The best hires don’t sit still. They’re constantly looking around corners for how they can expand, grow, and have more influence.”

What it looks like

As you progress the candidate through the interview process, notice if they ask more questions, push themselves to learn and do their homework to show up as more of an expert each time.

How to vet

Give your candidates a type of take-home exercise that is intentionally vague and see how they thrive in the response. These days take-home exercises have lost their lustre because of the prevalence of AI tools that can help you build content, identify patterns and generate insights at rapid speed. I still find them useful if you ask someone to reference their own experience, and especially if you pair it with a live presentation.

Passengers vs. Drivers

“You can’t hire someone who is just a passenger along the way. You need someone who takes the bull by the horns and becomes a driver.”⁠

⁠Are you taking us to our destination or are you along for the ride? I think this dynamic is currently en vogue under a different designation: High or Low agency. I personally prefer the visualization of the Driver metaphor, and I think it is easier to describe the environment where two different personality types can thrive.

Driver

Exceptional Product Marketers will proactively steer the organizational strategy. These folks have a point of view, make it known, and then go out to build the solution. This is often referred to by my friend Graham Neray as a “heat seeking missile”. Drivers look for the opportunities, get involved, demand a seat at the decision-making table and make an impact.

What this looks like

Identify a problem with the product messaging, talk to the right people about it, circulate your ideas for feedback, get qualitative and quantitative data to back it up, the answers you need and implement the solution.

Subscribe now

Passenger

Passengers need direction in order to make moves, but will offer feedback along the way. This leaves everyone else needing to be a driver.

What this looks like

Hear about a problem with the product messaging, talk to Product about it, bring it up in your team meeting and suggest someone fixes it.

Pay attention to how candidates show up in every interview. Are they waiting for you to ask the questions, or are they proactively showing you how they’re learning and thinking?

How to vet

Watch for “I” vs. “we” language. “We” often signals the manager or their entire team owned it, not the candidate. To interrogate a level deeper, ask what their specific role was in a particular project end-to-end.

Sphere of Influence

If you are consistently driving, you’ll likely get closer and closer to more strategic company problems to solve. “There are a lot of directors of product marketing in the world. But there are also a lot of very mediocre directors of product marketing in the world. The thing that separates the good ones is that they actually have a seat at the table from a corporate strategy standpoint.”⁠

Exceptional product marketers build trust and are brought into these conversations. It may start with getting pulled into a project where executives need inputs for a board meeting. They do a great job there and impress, and eventually, they are pulled into those conversations all the time and become a key to the company strategy.

What it looks like

Does the candidate put themselves in a position where they are shaping the strategy, or are they taking the strategy from someone else? “Some PMMs build the slides once the strategy is done. The stronger ones are sitting down with the founder or CEO and shaping what they’re going to say on the main stage.”

How to vet

Ask “What’s the most strategic, high-impact corporate-level thing you’ve personally delivered?” Evaluate whether it meets the bar. Also ask them about failures in the chain of this project, to see how they handle the challenges that they encounter with these higher-profile projects.

Humility

All three of those traits, Driver, Locus of Influence and Scrappiness, all build towards one another like a loop. If you are scrappy, you find opportunities to drive, and that gets you closer to the Locus of Influence.

All of this is underpinned by Humility, the ability to hold a strong point of view while remaining open to feedback. Product Marketers have to capture input from their accountability partners, often Sales and Customers, without a defensive posture

“Product marketing has this interesting dynamic where you kind of have to be the smartest person in the room, but you never really are,”⁠ says Clarke. “You can’t come into it with arrogance and say, ‘I built this message and everyone is going to line up behind it.’ You have to be humble enough to accept that maybe you got it wrong.”⁠

⁠​

The people you hire are a direct reflection of your leadership. My former CEO, Dev Ittycheria used to say at MongoDB, “A’s hire A’s, B’s hire C’s, and C’s hire D’s.” Excellent people want to work for excellent people! And excellent people have high standards, make them known and understand how to get the best work out of others. I hope this gives you a solid framework of how to identify excellent product marketers onto your team!

What are some tips you would add to this framework? Or – taking a lesson of humility from our “excellent” product marketers – how would you change it? Let us know in the comments!

Subscribe now

Every week we’re trying something new and _curating_ a list of roles for Dev-first jobs. Got one? Send it or comment.

This week:

For those Just getting Started

Altis is looking for an Associate on contract if you want to break into VC

Mistral is looking for a Technical Marketing Engineer. Could be an interesting way to break into AI-first companies.

Jobs off the beaten path

OpenClaw needs a Community Manager/DevRel in Europe

Lago needs a Growth Chief of Staff to work directly with the founder

Jobs we’d want in another life

Runlayer (who we mentioned in a prior post) needs a Head of DevRel

Docker needs a PM to run growth

Cheers!

I’m a serial connector, which means if I’m interested in something you’re working on I WILL dm you and become your friend. In fact, this is how I met Francesca. As a result of my extracurricular activities, I often feel like my iMessage is a clearinghouse for growth convos (and SaaS gossip). I figure why not reuse my friendships and monetize them as content, right? Just kidding, we know I’m not getting paid for this. Nevertheless, this week I discussed how to get trials to convert “better” and put some more in the bank for later issues of this newsletter. Stay tuned.

Names and companies have been hidden to protect the innocent.

“We’re great at getting trials… but they’re not converting”

One friend reached out with a very specific (and very common) trial conversion problem.

The context: The current setup they described:

• The product is seat-based and a tops-down purchase, and this is kind of a PLG motion inside of it. This PLG motion is new and my texting friend is leading it.

• They only run trials for cross-sell / upsell. This means that they run them existing customers who already have a credit card on file.

• A trial start signals something beyond high intent for a prospect, something akin to “opportunity created” for their internal teams.

• 14-day uncapped trial (they recently shortened it from longer)

• They’re generating plenty of trial starts

• But a lot of those trials aren’t converting and the users aren’t engaging with sales either

• The trial is OF COURSE an AI feature

My pushback:

What should we actually count here as a “Trial Start”?

A self-serve attempt to use a new product is truly not the same as a new opportunity from a sales-led world. I suggested not only a handraise but some real usage of the new AI product should count as a trial start. If we cut the data that way does the trial conversion look better?

Change Trial Start signal to a handraise vs. an opportunity creation

This thread is a great example of a very specific growth trap: PLG high trial volume + low conversion can be really frustrating at a sales-led org that expects product-led users to be qualified sales leads like marketing feeds them. These are not marketing leads! The product trial should have its own criteria for sales engagement. These often fall into two categories

• Product data: Engagement metrics like usage. Some companies also use antipatterns as outreach signals. For example, if a user gets stuck or runs into an expensive error, get someone to reach out and offer help.

• Firmographic and User data: Identify ICP accounts and titles qualify those for sales outreach. This is incredibly easy to set up these days thanks to tools like Clay.

Be the Human In the Loop: Ask them what they need?

This friend needs to be more aggressive in a few ways:

Product outreach: This is a new product. The product manager who created it and is owning it absolutely needs to be paired with this team to understand what is and is not working to quickly iterate alongside the team. Without that, GTM is lipstick on a pig. They need to be connecting with these prospective users or doing actual user interviews too.

Driving trial conversion: For example, I’d try reaching out to people who requested the trial but aren’t using the product and telling them the trial will be turned off if they don’t start engaging, and that they can get it back by reaching out again. (This is sometimes the nudge you need!)

Product Positioning (AKA not everything needs to be AI)

Very few folks were interacting in this “trial.” The audience for this trial is back office ops folks. One hypothesis that I threw out is that the product is an AI plastered textbox that required loads of imagination to use. Does this persona like to type their wishes and desires into empty boxes? Is this persona a bit scared of AI? If these things are true, I suggested that they should experiment with a less “AI-forward” approach and instead come up with a few templated asks for the trailers to choose from in order to drive an “aha” moment.

Right Message, Right person, Right time

Many of my growth suggestions included emails triggered off of a user’s action in inaction. My texting friend mentioned a broader comms fatigue problem: sales is already very aggressive with outreach, and marketing had been sending so many emails that they had to pull back to avoid burning out customers (and almost burning their sending domain).

I recommended pulling back on that but we all know how sales can be.

In my experience, the fewer recipients of an email in a workflow, the higher the open, clickthrough, and opportunity rate. The days of spray and pray are long gone. At ngrok, we target our emails to be focused on helping a person through a more friction-filled part of the journey like an error message, or helping them discover a feature that’s aligned to their use case.

Stuck on a growth question? Text us.

That’s it for this week’s batch. If you’ve got a gnarly growth question you’ve been texting your friends about, send it my way — I might feature it next time.

Jobs with one of us!

Head of Growth at ngrok: Work with Sam to help make the growth loop tighter, faster, and smarter. ngrok is entirely product-led and this role will have a massive impact over the strategy and tactics that drive that topline growth.

Jobs for those getting started

MKT1 Intern Intern for one of the coolest Marketing Leaders and Substackers in the game: Emily Kramer. The MKT1 newsletter, read by over 80,000 marketers, is a staple of every startup and enterprise marketers media diet. You know this role is going to be cool because of the application process: just submit a LinkedIn, a video you made, and a Claude Skill you’ve built.

Jobs off the beaten path

Developer/Community Advocate, Americas at Clickhouse. Great for a future road warrior. This team is phenomenal: I personally think the Head of Marketing and Product at Clickhouse, Tanya, is amazing, and a few former colleagues I LOVE have gone to Clickhouse as well.

Founding Developer Advocate at ConfidentAI: They’re looking for a high-agency unicorn who wants to help build infrastructure for AI trust.

Developer Relations at WorkOS This came in directly from the CEO, Michael. Authorization and Authorization (Auth* IYKYK) have always been critical for B2B SaaS going up-market, but in the age of agents, it’s even more critical to have your Auth on lock. WorkOS is building the backbone for this and they’re looking for an events-loving DevRel leader to join their team.

Subscribe now

Director of Integrated Marketing Campaigns at Temporal I met the founders of Temporal years ago and was so impressed with their vision and experience. They’ve been building versions of Temporal for years at Amazon and Uber. The push to open source the tech was worth it. They’re doing incredibly well and are a critical component of agentic workflows for teams at OpenAI, Replit and Cursor, to name a few. I also love what the CMO, Clare Byrd, has been doing at Temporal (they recently became a sponsor of Crystal Palace FC!) so I think this is an excellent role on a team that’s doing things a bit differently!

Jobs we’d want in another life

Insert Job You Excel At at Etched, who are building frontier inference clusters (a new type of hardware for AI workloads) and just raised $800M. They’re a team from NVIDIA, Broadcom, TSMC and others. This is a cool company to watch as the demand for hardware grows. Honestly I kick myself everyday for not buying TSMC stock (I thought about it 5 years ago!)

Editorial Director, On With Kara Swisher at VoxMedia I mean…how cool?

Drop your role in the comments if you want your job posting featured in next week’s Hot Jobs! And if you’re on the hunt for a DevFirst job, send us a message on Substack!

• Tools and Governance

• The Hive Mind

• Hacking and Experimentation-focused rollout

For context, I run a full-funnel marketing team of 6 responsible for pipeline and revenue outcomes. All of them are non-technical marketers and do not know what a bash command is! I wanted my team to take advantage of the power of AI and not be left behind as teams shift to AI-driven workflows.

I’ll be candid: I stress about this a little too much. I feel like I’m trying to prepare my albatross for their long flight! These investments dbt Labs made helped my non-technical team — and countless others — evolve the way they work with AI. In this post, I’ll talk about three things that drove my team’s AI usage and show some examples of tools we’ve built.

Identity and governance: Get your MCPs to connect safely

All new process rollouts are more successful with a paved road, reducing the adoption friction. We rolled out an MCP catalog, which streamlined the adoption of AI tools for my team at dbt Labs.

MCP, the Model Context Protocol, is a framework for connecting LLMs to other systems – like connecting your LLM to Figma so you can translate text or code into designs. Most business systems you use today likely have an MCP you can hook into. Setting up your MCP often involves connecting your API key to an AI Harness, like Claude, but creates what is called “Shadow AI.” Just like “Shadow IT” before it, this refers to the sprawl of unmanaged MCPs in an organization, which could lead to vulnerability exposures and data leaks that become challenging to trace and remediate.

At dbt Labs, our security team rolled out the Runlayer catalog, which I like to think of as Okta access management for your MCP servers. With this in place everyone internally can access blessed with secure MCPs from the catalog so we can easily connect to them.

This was a huge unlock for my team. They were able to quickly connect to internal systems, like Notion, and see the opportunities from connecting Claude, their AI Harness, to these tools. Instead of creating content or doing research with AI, they were able to build apps, workflows and reusable tools (some examples later).

Runlayer also hosts our organization-wide repository of skills so we don’t have to send around Markdown files to adopt those modules. Instead we have a managed library of those official skills that we can discover and easily port into our AI harnesses.

How it works in practice

To set up the Omni MCP, which is an approved connector, I requested access in Runlayer and then ran a command in Claude Code to complete the set up. That’s it. I didn’t need to deploy the MCP to a machine or a cloud instance. Overall, I think this is a must have for any organization who wants to accelerate AI adoption.

The Hive Mind

A repository of structured data and existing workflows can be a critical context for your internal AI adoption. Everyone at dbt Labs is a Notion super-user, and we quickly convert new teammates who are Google Docs stans into Notion fantatics. I like to think of Notion as an internal knowledge system powered by structured context that is easily integrated into other tools. This can integrate beautifully into LLMs as a corpus of organized context, structured data and existing automations.

How it works in practice

Pre-AI adoption, my team has documented nearly every process in Notion. They also have a number of workflows orchestrated through Notion databases. One of these workflows, our Global Marketing Calendar, (GMC) is the central repository of all our marketing activities, and powers workflows in Salesforce and Asana. This context has helped us build more advanced orchestration using AI. Some examples

• ToFu Pacing Dashboard: I built an artifact that tracks our current pipeline pacing, budget and projected end of quarter outcomes using data from our GMC, Google Slides, and Omni. This is a great example of a proof of concept that I built that I will ask our data team to build into a dashboard. You can see an anonymized example in the image below.

• Campaign Tracker: We pull in campaign member data from Omni to understand whether we’re effectively targeting our ICP. The Notion page updates every Monday and sends a Slack message to my team’s Slack channel.

Capturing this data on a regular basis for reporting would be very time-consuming. But I can’t imagine how challenging this would have been without all of our structured context from Notion. We are now reaping the investments we’ve made over the years to relentlessly document and automate everything, and that encourages us to keep up the good habits.

Hacking and Experimentation-focused rollout

Hackathons are common amongst development teams to encourage innovation and experimentation, but it’s rare for a marketing team to host a Hackathon. Our marketing team ran a series of smaller Hackathons to encourage dedicated hands-on-keyboard time for building. Open experimentation time are especially valuable when you establish constraints for your team. Research from Patricia Stokes, a psychology professor at Barnard who studied how artists create new ideas, shows that innovation occurs when artists keep certain variables fixed but alter the others. This theory, called “solution by substitution,” helps artists find new ways to represent their visions.

The same is true for experimentation with AI. You can’t start with “build anything with AI.” Help them establish those constraints and have AI tools be the workflow innovation.

How it works in practice

Take AI completely out of the equation. Help your team identify small tasks that they can automate. My team used an Eisenhower matrix, a task management tool, to list out our work for the week. We filled this out asynchronously and then met as a group to review the submissions and vote on the top two projects to build. Some of our current automated workflows, like the Campaign Tracker, were inspired by this exercise.

The Eisenhower matrix worked because it helped us surface repetitive tasks that AI could automate or accelerate. It also helped us distinguish between tasks to automate and tasks that require human judgement.

With any AI adoption rollout, I wouldn’t say there is a perfect formula. Every organization is unique and only you can understand the right process and technology to match your people and culture. Most teams are still figuring this stuff out (we are too!) and I am so curious to understand what you’re doing to drive AI usage internally. Share what you’re doing in the comments.

A wise man, Brian Kotlyar, VP of Marketing at Hightouch, gave me excellent advice a few years ago: be the expert on your customer. As a marketer it is your job to understand the customers jobs to be done, their pains, their fears and how to reach them! On top of that, it’s a excellent way to build credibility with Sales and Product. I’ve seen many marketing teams assume product marketing is the only team that should understand the customer, an unfortunate outcome of extreme specialization in growing organizations.

But with AI tooling on our side and the right data we have the ability to distill more customer insights more frequently and turn our marketing programs into flywheels: fueling the engine with customer insights to increase reach.

Brian used recordings of customer and prospect calls in Gong as his primary resource to stay informed on the customer. Now, we don’t need to watch 20 Gong calls, we can summarize a huge number of transcripts. In my experience, LLMs are excellent at organizing this type of content into themes. We have MCPs for a number of our sales enablement, productivity and forecasting tools, including Gong and Glean, that give us deep understanding of the customer journey and better insight into the language customers are using.

The Campaign Flywheel

You run a Campaign with the goal of acquiring 20 new customers each month. Let’s say you acquire 25 customers in a month. Exceptional work! You are driving 25% more customers than you had originally planned. That’s the first value of running this campaign. But the customer insight around why this campaign is so powerful is the second value. This is best expressed using the sociological framework of manifest and latent functions.

Your campaigns now have two jobs. The manifest functions are the outcomes we expect, the intended outcomes. The latent functions are the unintended outcomes. With the Campaign Flywheel, you deploy marketing dollars, design resources, content and execution budget. Your manifest function is acquiring those customer logos. The latent function of this are the customer insights you capture. While you run the campaign, you can review customer data signals, those sales tools we referenced above and capture those insights to inform your next campaign.

Analyzing Customer Insights

My teammate, Mary Alati, did this exact exercise for our Webinar program. We’re in the middle of Q2 and I wanted to understand how our overall programs are performing and what we could learn. Specifically, I wanted to understand why these prospects were looking to buy right now. She used our Omni MCP and hooked Claude into Glean to build the analysis.

New Hire Triggers

What we saw: One of our evergreen campaigns has gotten very good at acquiring teams new to dbt. We’ve recognized there’s a theme: when a new data leader comes in, they often want to modernize the data stack and dbt sits alongside Cloud Data Warehouses as the first things they invest in.

What we learned: This shouldn’t be news, these external triggers are sometimes the strongest indicator of buying intent and new leaders will often make purchase new software in their first 100 days to drive new strategy and execute quick wins. But, we hadn’t thought of it yet. This was the signal to get something kicked off!

What we’re doing: We’re taking this insight into the next iteration: using Clay to target job switchers of this persona with content about dbt.

Subscribe now

The mAIn event

What we saw: We have a ton of folks engaging with our AI content — some from new personas! — but when it came to the deal cycle, it seems like our prospects are early in the AI maturity curve and are engaging with our content to see what great looks like. But implementation is a far way out.

What we learned: AI is obviously a hot topic and has high ROI, but teams are much earlier in the adoption process than we anticipated.

What we’re doing: Investing in some content that map to earlier in the AI maturity curve to capture more of that market.

Sharing the wealth

We’ve shared this project with our peers in Revenue Marketing to help them see the importance of incorporating customer insight into their campaign proposals. It’s a shared skill in Claude CoWork that anyone can pick up at dbt.

Your end users are the ones that matter, and in a less “woo woo” way, your customer knows more about how their organization adopts and purchases software than you do. our customers are always telling us what they think. Listening can be one of the most effective tools for we can improving your programs.

How are you using customer insights to power your marketing campaigns? How do you keep your marketers in touch with the needs of the customer? Share your best practices and let’s start the conversation.

Why do people decide to angel invest?

To quote me, myself and I from a recent group chat “Angel Investing is a really stupid way to make money”. Any investment portfolio should balance risk and security, and Angel Investing is among the riskier investments you can make. Why? You invest a sizable amount of cash in a company at a very early stage with a low probability of return. Your investment risk will decline as the company matures, but with each subsequent round, your potential upside will decrease. No risk, no reward baby!

Ergo, I would never suggest that anyone start angel investing until they have some foundational investments, like a 401K, mutual funds and/or a rainy day fund. Those are not the things that will get you “rich,” but they will be the financial cushion you need to protect against baseline uncertainty.

Every Angel Investor has their own motivations for investing. Some see it as a high-risk portfolio of bets that they will someday reap! Others, like me, do it because it’s interesting, fun, exciting!

What’s my why?

I started angel investing as a way to learn about the industry and prepare for a pivot into Venture Investing. I had been curious about this area for years, but never realized I could try out this career path on my own. When I was at MongoDB early in my career I got to meet tons of startups who were marketing to developers and wanted feedback on their strategy. I loved working with these Founders, learning about their challenges and exchanging advice. I found it helped me remain agile and curious in my day job.

Angel Investing was not something I considered as an engagement option, but I wish I had understood the entire ecosystem sooner. As an Angel Investor you stand with the founder early in their journey, understand their challenges and support them using your network and expertise. I understood my “why”, now it was time to figure out my pitch to founders. Why should I get the privilege of investing in your company?

Why do I belong on a CAP table?

One thing I didn’t realize about investing is that it’s incredibly competitive. You see this in the news today, with so many Venture Investors going after a small subset of standout companies for investment. It’s because out of the thousands of companies Venture firms meet with each year only a small subset of those will have an outsized exit that will make back the investment on the fund and give a return to their investors (Limited Partners).

All to say, you can’t just go to a founder with a check and get on a capitalization (CAP) table. You’ve likey seen it on Shark Tank: The more investors, the more complexity. The valuation grows with the amount of investment you take on, and founders want to balance the total valuation and maintain as much of their founder equity as possible.

It’s much easier to negotiate a place on the CAP table if the founder is a friend or former coworker. But if you’re not, you need to understand and sell your differentiation. Why do you as an investor stand out among all the other folks who want to contribute capital? Here are the areas where I spike and add differentiation, which I illustrate to help you identify your own spikes.

1. I’m a woman. Founders and VCs love getting female investors on CAP tables.

2. My Marketing experience is a value add to founders since most General Partners at Venture Firms do not have marketing expertise. Emily Kramer and Kathleen Estreich started MKT1 fund with this same premise! On top of that I have experience in Product Led Growth and that perspective is often prioritized for founders of Developer and Infrastructure tools.

3. I’ve worked at a few high-growth Developer-focused companies and that roller coaster experience is something founders want to understand. This is the most important piece of your brand as an Angel Investor. If you are a successful founder or an executive at a high growth company, you are a higher priority for a founder to include in their round.

When you meet with founders, it’s important to articulate why you’d be an asset on the CAP table. Communicate your value so they can vouch for you. When I was early in my journey, I took this for granted and thought my experience spoke for itself. Sadly, it didn’t and I lost out on a deal with an incredible founder because the he didn’t understand how I could help the company in their early stage of growth. I learned a lot from that lesson and now I have a clear pitch I deliver (more on that in a future post) so it’s very clear to founders how I operate and why I am a unique value add.

From First Check to Angel Track

My first investment was in Pocus, the AI GTM intelligence platform. I had gotten to know the founder, Alexa Grabel as she built the community for Pocus, which was initially focused on companies with a PLG motion. I loved what Pocus was building – and if you’ve met Alexa, you know, she’s got a talent for galvanizing support and has an impeccable eye for talent. In short, no matter what she does in her career, she’s going to win.

After that I got the opportunity to enroll in Angel Track, a program run by FirstRound capital that I call the Hogwarts of Angel investing. Many people start angel investing like I did: they’re curious, they met a founder they believe in and then that snowballs into a fun “side of your desk” job that requires skill — most of which you’re learning on the spot. Maybe it’s like Max Weber’s beruf, or “a calling”, something that has both rational economic value but is also spiritually motivated. That calling deserves care!

Before Angel Track, the only way to get better at Angel Investing was to keep doing it. But Angel Track teaches you the practices that can help you evolve into a better investor and helps you hone the special skills you bring to the table.

Angel Track was one of the best investments I made in myself. I got to connect with other Angels who have different perspectives and experience. I left with a playbook for investing and even though I haven’t been actively investing since I had a baby last October, I know once she’s a bit older, I have my own playbook to dive back in and meet founders who excite me!

Next to my time at MongoDB, no other network has been as valuable to me as angel track. I have a group of folks I connect with regularly from my Angel Track cohort who have taught me so much about investing, hiring, marketing and life. If you want to dig deeper into your Angel Investing skill set, or if you ever have been curious about starting your own fund, I highly recommend it. Follow FirstRound and Partner Brett Berson (AT’s Dumbledore) on LinkedIn to keep tabs on future cohorts.

What next?

This is the primer of why. Becoming an Angel has given me access to new communities, and widened my aperture of the venture world. If you’re an Angel, I would love to hear your story in the comments. What is your “why”?

Next in this series I’ll share my approach to sourcing Angel opportunities.

Under the Fig Tree

What I’m reading and loving.

Books

I just started True Story: Murder, Memoir, Mea Culpa by Michael Finkel. The story starts with the mea culpa: the author is disgraced after a scandal got him fired from New York Times magazine. At the same time, a man from Michigan, Christian Longo, takes on Michael Finkels identity, runs away to Mexico after, allegedly, drowning his wife and children. The book chronicles the author's return to investigative journalism following the murder and incarceration of Longo. It's an oldie and was turned into a film starting James Franco and Jonah Hill. CRAZY and delish.

Articles, etc.

I Let Claude Code Autonomously Run Ads for a Month a great case study by Technically

A visual guide to getting out of a creative slump from Lenny’s Newsletter. This post showing up in my inbox was kismet. I almost publish this post! But after reading this I remembered the advice I’ve given to so many friends: it’s about the reps. Get something out there and move on!

The power of an accountability buddy to get things done from NPR’s LifeKit. Shoutout to my friend from the Flesh Report (follow it!) for setting the example as a writing accountability buddy. Sometimes you need the loving pressure!

Tons of companies want to know how to sell to developers. I’ve never sold to anyone but developers and I do think that developers have less patience with salespeople than folks in other roles (why? I think the same things I mentioned in Show and tell). But there’s been an exciting transition in the past few years: the Product Led Sales motion that my friends at Pocus have espoused over the past three years has become a reality for companies outside of the Developer Tools space and many companies like Ngrok, Loom, Linear, Productboard and Hubspot are using a hybrid sales model to grow their businesses, often with Product capturing leads for the sales led motion.

From Pocus’s 2022 Product Led Sales Benchmark Report

GTM teams are always looking to increase net dollar retention, improve sales efficiency and CAC payback. Product Led Sales is an additional sales engine you can add on that can improve those measures significantly. It’s also the best approach to selling to software developers for two reasons:

• The Product-Led Sales motion rests on having end users trying out and actively using your product and software developers love test-driving new software. This process inspires belief in your product you couldn’t get from the incumbent enterprise sales model where time-boxed demo with a sales rep was their only path to understanding your software.

  • Additionally, in a business with a strong Self-Service customer base, you can build a massive customer base within key accounts before introducing a Product Led Sales motion.

• Because users are actively using your SaaS product your team has insight into their habits and challenges, something that is much more challenging to do in the enterprise sales model. This allows for smarter more targeted outreach that developers and their managers will appreciate.

  • Products like Pocus make it really easy to surface this data from your data warehouse in a way that’s easily actionable by your sales reps and account managers.

  • When the time comes to pitch, you don’t need to call the developers directly to sell them a product they already use. You can call their boss, someone who is more senior and has larger budget and team oversight who is looking to solve big problems and will be more willing to speak with you. You already have adoption amongst their team, so it is easier to start these conversations.

Share

To summarize, Product Led Sales is a tactic that’s being used for developer tools companies precisely because product usage gives a very strong signal of intent. Once you have that product signal, it becomes easier to surmise the needs and wants of a customer to faciliate deeper conversations.

Some things to get right before you launch a Product Led Sales motion. Some of this might seem basic, but as someone who has worked in the range early stage startups to public companies, I now understand that nothing should be taken for granted. Everyone grows across different vectors at different rates, so what you think is basic or obvious advice might be valuable or novel to someone else. I love this post from my friend Alexa Grabel on how to launch a PLS motion using a Crawl, Walk, Run approach. Some of our advice is similar, but she paints the picture very clearly!

One of my first principles is parsimony, or simplicity. I always guide people to start as simple as possible and then grow more complex over time. When starting in the crawl stage, don’t overthink it, but be intentional. Nothing will be perfect when you start, but you have to think of the simplest approach you can take to learn something meaningful.

• Easily Accessible Data: A unicorn, I know! It’s really challenging to give your team access to the right data that they can meaningfully interpret and use. Ultimately, as long as you have a team that can pull the data for you and help you understand what the data means, there’s hope here! Ideally you will have a data warehouse, a product analytics BI tool like Amplitude or Mixpanel that can help you reason through the trends and identify groups of users to reach out to.

• Ideal Customer Profile: What are the common traits of users who buy your software? This doesn’t have to be perfect, but understanding one or two attributes here will help you immensely!

  • Company size or industry

  • Technology overlap, for example, the perfect marriage of DBT and Snowflake, Node.js and MongoDB back in 2012

• Product Signals: Try to understand common engagements with the product that signal buying intent. Look beyond basic metrics such as clicks or page views and dig deeper into the customer journey. By identifying key touchpoints and common engagements with the product, you can gain valuable insights into how customers interact with our brand and what drives their decision-making process. Lots of companies will use an “activation” metric, and it’s likely your product team has developed a paradigm that aligns to this already. If you’d like more insight here, Amplitude has an awesome guide to Product Metrics that serves as a great primer. Again, this doesn’t need to be perfect, just good enough to start for now.

• Prioritization: Based on the above, you’ll be able to put in some type of prioritization approach for your teams that will help them sort through all the leads!

This week’s links

• Love the Problem: Live for opportunity, not for solutions. Wise words by Julie Zhou

• This essay on using GPT-4 to understand how time passes in written text is so well written and the TL;DR that LLMs are very cool because they’re capable of explaining to you how they thought about a problem is 🤯

• The cognitive load developer’s handbook, for anyone who wants to develop empathy for the developer and the impact of context switching on their work, look no further than this post which articulates why software engineering can be so mentally taxing. Everyone in DevTools is trying to make this simpler!

Today, I got an exciting announcement: I’ve started a new job as the Head of Marketing for Oso, the platform that makes it easy to add RBAC, sharing, fine-grained access, or any other permissions model you can think of to your apps. The permissions-as-a-service space is relatively new as the trend of offloading developer toil continues to grow in popularity, following the giants like Stripe and Twilio. 

Oso makes it possible for developers to work on the parts of their app that deliver customer value, not the tedious parts of access that take time and energy away from projects and delay release dates. Oso makes authorization simpler so you can build better collaboration experiences for your customers.

Why Oso? 

As Rolland Garros said, victory belongs to the most tenacious. Graham Neray, who I worked with at MongoDB for 5 or 6 years – who is counting?-- is the CEO of Oso, and although he and the team gave me a hard sell, it was a very easy decision for me to make. Graham is probably the hardest working person I know, he’s relentless in his desire to do things right and he and I share a lot of the same values that we developed early in our career at MongoDB. While he’s incredibly accomplished in his own right, no work is too low for him – he’s always down to pick up the trash – a quality I really admire and wish to emulate every day. 

I also saw a strongly realized customer-first attitude on this team. Lots of companies say they are customer first, but then you realize that they only care about their paying customers or they just say they’re customer-first so they can sign deals. Oso really lives the value of “work for the customer". Engineers spend 20% of their time doing customer-facing work, and everyone on the team sits on customer calls. In the conversations I’ve had with team members, few have spoken without referencing the voice of the customer, something that is very challenging to emulate. 

Subscribe now

The Tido Framework 

As referenced above, there were a lot of pros when it came to Oso. I had pre-existing trust built with the founder and the overlap of development and software security was appealing to me. Still, it’s important to have a framework for evaluating your options and I was thankful to have a great teacher in Tido Carriero, who offered me his framework for evaluating big career opportunities, which I found extremely helpful and relevant for just about anyone looking at a new role:  

• The Reporting Line: Do you believe that your manager and your manager’s boss are people you can learn from? Do you have conviction in them as leaders? Your reporting line is your support system. They hire your peers, and the best people hire the greatest talent. If this line is not strong, you end up working with people who you are less likely to learn from, who won’t be effective advocates for you or your career, or folks who don’t have the background to have strong instincts that make teams and companies successful. 

• The Francesca-shaped hole: Is there a big problem that you can uniquely solve? Is it something that you’ve done before, or an area where you feel uniquely qualified to figure out over a 6-18 month period? This, along with your reporting line, are your markers for success. If the problem was not Francesca shaped – for example a few companies I spoke to were looking for brand experts who could help with product marketing and category creation, areas of marketing where I have less expertise. With Oso, I have conviction that I can help them drive more awareness and usage of Oso Cloud and I have the leadership skills and expertise to build out a marketing team to support their next stage of growth. 

• The ROI: Is there an outsized opportunity that will come from this – monetary or otherwise – that motivates you to do an amazing job? For me, money is important, I’d love to own an amazing ski house, but the opportunity to learn from Graham and work alongside folks who have the same hard working ethos was critical for me. I know that working with folks of this caliber on a problem as big as authorization would be incredibly rewarding and important for my career trajectory. Additionally, I know this job is experience that will help me in my side gig as an angel investor – running an entire marketing org is a step up from running Growth, Dev Rel and Demand Gen teams, which will help me advocate for allocation in companies I want to invest in and future board seats. 

I hope this framework is helpful for you as you think about future career transitions. I’m super excited to be back with my MongoDB family and I’m thrilled that I get to share some of my journeys at Oso with this community. 

I’ve never been able to accomplish anything meaningful without community – and I am so fortunate to have an incredible network of friends who supported me as I explored the next stage in my career.. Thank you so much to my people, namely Ed Sim, Lenny Pruss, Ian Livingstone, Samantha Richard and Meg Goetch! 

Specifically, I’d like to talk about a delightful model for content development and recycling that I call “The Content Galaxy.” I know that many organizations struggle to determine how much content to create, identify content creators, and articulate challenges to their users. The Content Galaxy can help you build a framework for engaging your content creators and building brand equity around core topics that are important to your audience. More specifically I have used the Content Galaxy model with DevRel teams in the past to help them prioritize their workload and focus on longer-term projects that drive performance (e.g. impressions and lead gen) 

By using the Content Galaxy model you’ll see how to structure content opportunities for your DevRel organization. First I will introduce the Content Galaxy framework and then I’ll walk through an example. Finally, I’ll offer some guidance on how to use the framework in your own DevRel organization. 

Introducing the Content Galaxy framework 

No Developer wants to admit it, but when you are doing DevRel you’re doing advanced content marketing. Of course, there’s more to DevRel – engaging with your community, supporting local developers, being a connector within your community – but when we think about the performance metrics associated with DevRel, many of them are achieved through quality content marketing. Content marketing of this kind can be defined by a few characteristics

• Creating quality content based on what your community’s needs 

• Aligning that content to your product’s core value

• Distributing that content through relevant channels to drive impressions

• Building that content into growth loops to drive evergreen impressions 

• Measuring the impact of the content you create 

Many teams struggle to create a lot of high-quality content. The Content Galaxy framework 

allows your team to focus on building a large asset at the beginning which then leads to creating smaller, more easily digestible assets to distribute across different channels. 

A Content Galaxy includes 

• The Sun: The large content that sets the stage for all the subsequent recycled content. For DevRel organizations, this is often a sample app you create to demonstrate a problem or showcase a new solution. 

• Planets: These are content pieces created out of your original Sun. A tutorial based on the sample app is an example of a planet! You can and should have multiple planets as part of a Content Galaxy

• Moons and Stars: These are short-form content pieces that you can simply and easily create once you have a Sun in place! For example, Twitter threads, TikToks, and Emails. 

Please note, I know this framework has no basis in scientific fact. This is a framework to help you with content, not to show off my knowledge of astronomy. 

My favorite part about this strategy is that it’s both a content-creation tactic and a distribution strategy. From that large Sun that you create for your content Galaxy, you can then create videos for YouTube and TikToks from the same original Sun! 

There are tons of ways to experiment with this as a tool for building equity around specific topics. 

Share

Case Study: Temporal 

For those who aren’t familiar with Temporal, it’s the programming model of the future, based on an open-source project built at Uber called Cadence. Temporal powers microservices at Snap, ensuring seamless delivery of Taco Bell online orders and resiliency of Netflix’s real-time data and CI/CD platforms. 

Temporal is ushering in a completely new mindset around resiliency, which they call Durable Execution. Developers need to understand what Temporal is and how it can evolve their infrastructure – and the fastest way to do this is by articulating different use cases where Temporal can transform your reliability and simplify operations, as I’ve done above 😀. 

One of the cool use cases for Temporal is a simplification of a transaction pattern called Saga. This is a popular pattern for workflows, but often challenging to operate long-term. Temporal provides native support for this pattern, making it simpler to build and operate, so it’s a perfect use case to showcase the power and simplicity of Temporal while building the brand equity of Durable Execution as a concept. 

Temporal has created a Webinar on the Saga topic and is in progress on a tutorial that goes through building a trip booking system and introduces Temporal’s native Saga pattern. Based on this, I created a content Galaxy for Temporal based on the Saga pattern in PHP. 

Introduce the Content Galaxy to your team

I’ve used the Content Galaxy model with DevRel teams in the past as a way to structure a Dev Advocate’s content plan, and I’ve suggested it as a model for many more teams! It gives your team the freedom to work on something they care about, the freedom to explore something new and yet still has structure so you can manage output and, subsequently, content performance. 

When assigning out the Content Galaxy approach, I’ve always emphasized 

1. This is a time-boxed project. This gives folks guardrails around their Sun projects. They should be ambitious but not something that will take three months to build. 

2. Start with your Sun and go from there. You will learn more as you go along. Sometimes, especially when my team was building with new software, my team members discovered application performance challenges, which then became blog posts in themselves! 

3. Focus on something fun and exciting that will motivate you and is still important and interesting to your audience. 

Not all products are positioned to recycle the Saga Galaxy – SORRY – but if you know your customers you probably know some topics that could benefit from a content Galaxy structure. 

For MongoDB, my former employer, there are evergreen topics, like Schema Design, that you can truly pontificate about forever. There are SO MANY ways to design your schemas around different applications, so many perspectives around what works, and endless ways to optimize further. Schema Design as a topic is a massive Content Universe – maybe let’s call it a SuperCluster– and you can spend years building smaller Content Galaxies around that topic. 

Let’s say someone on your DevRel team wanted to take Schema Design as their Galaxy for the quarter. How do you provide them with direction so that they can choose their own adventure AND have enough guidance to create a compelling enough Galaxy? I would look at the following: 

• What schema design challenges have I seen posted to the community this week? This month? 

• Are there new types of applications being built (gfab, generative AI) where a different approach to Schema Design is needed? 

• Are there common challenges with Schema Design you’ve seen recently? 

• Do you have a new perspective on this that you’d like to showcase with a new language you’re learning, like Rust? 

In short, how should you pick your Content Galaxy? 

• Specific, not evergreen 

• Relevant to your community 

• Compelling and motivating to you 

• Timely, if not zeitgesty!

• Articulates your brand promise and increases brand equity  

I hope this framework is helpful to your team. I’d love to know if and how you use the Content Galaxy in your work. 

Sides of my desk 

I’m late to this, but I’m reading the most INCREDIBLE book, Tomorrow, Tomorrow and Tomorrow about two friends who love video games. It’s making me reminisce about the joys of platonic love and is making me want to play Zelda. (Goodreads) 

Also this NYTimes essay on the origins and development of Zelda made me deeply nostalgic and I will probably buy Tears of the Kingdom this week, play it for one hour and give up because I’m actually terrible at video games – and its an area where I have no resolve to keep going. If you have any tips that are not 10,000 hours, please reply! (New York Times) 

Deeply off topic, this Grinder (aka Hero, Hoagie) pasta salad looks like the perfect thing to bring to a piquenique. (Instagram) 

My friend Carrie’s essay on our culture/country’s deep desire for fame. (The New Yorker)

I love these comments because this problem of “wtf do you do” in B2B software is so pervasive that it’s shocking when you can get a clear answer in 1-3 lines. 

It’s so challenging to come up with these pithy and clever lines that we (marketing) ends up delivering confusion when we should be moving towards simplicity. In this post, I share why positioning for Developers is often perceived as so challenging and show you two examples of how to make the process a bit simpler. 

But…Developers hate marketing!

People will often say “developers hate marketing,” which is a statement I find so funny. Generally speaking, developers actually love marketing. Some examples: 

• Events: Developers often attend conferences and meetups to learn from other practitioners. 

• Swag: You know that developers LOVE getting cool t-shirts, mugs, backpacks and more to flaunt their favorite tools and frameworks. 

• Streaming and content creation: Developers have become incredibly talented content creators and boast massive followings on platforms like YouTube and Twitch, where they share expertise, curate content, interview other professionals or do live demos or hacks for their audience. 

• Community: This isn’t exactly marketing in my mind, but often community is lumped into a Marketing organization or a strategy. Developers have built countless platforms and tools for connecting with like-minded folks to exchange ideas and support one another. Examples include, IRC channels, Discord, Hacker News. 

As you can see in the Fly.io Hacker News thread I referenced above, when a brand is clear, concise, and straightforward about its product’s utility and value, you make a statement and turn some heads! 

Why are developers skeptical of marketing? 

When doing any type of marketing or product development, you need to understand the day-to-day life, needs, and convictions of your end customer. In the case of developers, most marketers do not have a clear sense of what software development is like. Here are two things that stand out to me about the software development process that could be helpful to marketers trying to understand the challenges of Software Developers: 

• The cognitive challenge: Software development is cognitively challenging and takes focus to execute correctly. For those who have never done any computer programming, it is similar to solving abstract math and logic problems which can be mentally exhausting. “Context switching” is a term often used by programmers to describe the challenges of managing their day-to-day workload, because many require hours of uninterrupted time to focus on accomplishing their tasks. As a result, any new software that they add to their workflow needs to accelerate their ability to execute and lower the cognitive load. 

• The level of responsibility and ownership: Developers build tools used at a very large scale. Their decisions in product design and technology choice can lead to at best an all-nighter and at worst threats to people’s lives. We’ve seen these examples in the news countless times. I know I am dramatic, but if you’ve read This is how they tell me the world ends, you know why. There are repercussions of choosing the wrong software and making those incorrect decisions can have large-scale impact. As a result, developers are very discerning about the products that they use because they want to avoid making those mistakes. 

This is why statements like, “This is better than anything you’ve ever tried” don’t work because they don’t believe that could be true; they’ve been lied to before. This is also why the use of Buzzwords on your homepage, like “Edge Computing,” often frustrates developers because it shows you just know the word and probably doesn’t understand the concept. 

My guidance: speak to them in their language. Help them understand how what you’re doing can help them. Help them understand the problems that you solve and how you do it. 

How to do “Show don’t tell” 

Back to the top: What did Fly.io do so well to get this praise? 

Here’s a reprint of the “elevator pitch” 

“fly.io is really a way to run Docker images on servers in different cities and a global router to connect users to the nearest available instance. We convert your Docker image into a root filesystem, boot tiny VMs using an Amazon project called Firecracker, and then proxy connections to it. As your app gets more traffic, we add VMs in the most popular locations.”

The reason the Hacker News community loved this so much is that Kurt and the Fly.io team wrote a clear and straightforward description of what they do, how they do it and why they’re different from other cloud services. They didn’t use fancy language, anyone who works in Tech can understand these concepts. For example, they describe the term “Edge Computing” without using that terminology – avoiding the Buzzword which could alienate their audience, which I love. 

Breaking it down: 

What they do? fly.io is really a way to run Docker images on servers in different cities and a global router to connect users to the nearest available instance.

How they do it? We convert your Docker image into a root filesystem, boot tiny VMs using an Amazon project called Firecracker, and then proxy connections to it. 

Why are they different? As your app gets more traffic, we add VMs in the most popular locations.

This is a really simple approach to getting down your pitch to developers. Another approach that I love is because it is similarly so simple First Round’s Arielle Jackson's positioning framework.

For (target customer) Who (statement of need or opportunity), (Product name) is a (product category) That (statement of key benefit). Unlike (competing alternative) (Product name)(statement of primary differentiation).

These frameworks are the first step to getting you to a better place when messaging and positioning your project. But, keep in mind, this is not a solo project! As a marketer, you are often not the end user, and you need their input to get the messaging right. 

• Familiarize yourself with the product by talking to your engineers, often! Join their demo days, read their team newsletters, and find a buddy who you can go to for questions. Fortunately at MongoDB, I had MANY buddies – some of whom are invited to my wedding! At Snyk I was embedded in the product organization, but folks like Ian Livingstone, Alex Ley, Ben Williams and Robbie Clutton were always super insightful when I went to them with questions about the user experience and different challenges facing our customers. Approach these conversations with curiosity and with the intention of making your company better and you will be welcomed by these engineers and product folks with open arms. The reason this is so helpful is you will get a bit more familiar with the problems of your users and how your engineers are solving them. 

• Talk to your customers! Get a better understanding of your customers needs by talking to them. I’ve found that happy customers are more than willing to take a call with anyone at your company, especially if you indicate it’s in the interest of making the product better for the community and if you offer a coffee gift card. Make a list of 5 questions to go in with so you feel confident and prepared, but know that the best feedback will come out of you asking “Tell me more about that…”. If you’re not feeling confident, see if one of your solutions architects or your engineering buddy will join you. Some questions to help you get started 

  • How did you find out about us? 

  • What problem are you trying to solve? 

  • How does this help you achieve better/faster/cheaper? 

  • How would you describe this tool to your mom? To your boss? To someone on your team? 

• Conduct user testing! This is one of my favorite approaches for iterations in design and messaging. If you’d like to iterate on your positioning, come up with a way to get users involved in the process. Ask users for feedback on your current homepage, or your current messaging guide

  • What do they love about the homepage/guide?

  • What do they dislike? 

  • How would they describe your product to someone in their role? 

  • What’s an example of a product they love that does this well? 

In closing, there are many ways to approach messaging. These are some tools I’ve found to be helpful to me or I’ve found can be useful to others — but this is not the only way! One thing is certain, though: if you approach things as a learner and not as a knower you will be vastly more successful.

I hope this gives you guidance, and more importantly, confidence in building your company’s messaging! Go forth and do great things and if there’s anyways that I can be helpful just leave a comment here! 

This is one of the most common questions I get from founders I work with. DevRel is not necessarily a new organization, but more companies have started adding in a Developer Relations team as a tactic for driving growth. For Dev-First companies, it’s a no-brainer, you have a team focused on building awareness and excitement amongst your target audience.

Most founders I speak with are curious about how to hire and build out this team and how to measure their impact. DevRel as an organization can have so many different focus areas – Advocacy, Documentation, Education, Support, Content Creation – that often I’ve seen teams go rudderless. What I’ve found is the best DevRel teams have a strong mission and a laser focus on specific outcomes. Picking a core focus area will help you and your team align on the right tactics to achieve those outcomes. 

Ultimately, measuring the impact of your DevRel team comes down to understanding what success looks like for you and your company. Each company will have different goals and objectives, so it’s important to tailor the KPIs and metrics you track to those specific goals. By doing this, you can make sure that your DevRel team is focused on the right things and achieving maximum impact.

What is your DevRel team’s mission? 

Before you begin to measure, you need to determine where you focus. We’ll choose a made up company, Kanjo, building a BaaS for Machine Learning Apps, to help articulate different DevRel missions, why they matter and the KPIs that align (and don’t align) to each. 

• We want to build a strong and vibrant community around Kanjo, our new BaaS for Machine Learning Apps 

• Grow Awareness of Kanjo in the Developer Community 

• We want to build a product that developers love! 

• We want to grow the number of developers building ML applications on Kanjo 

What is the simplest way for you to measure against these outcomes? 

We want to build a strong and vibrant community around Kanjo, our new BaaS for Machine Learning Apps 

In this example, you are looking to build community around the product, Kanjo. Community can be a robust or a flimsy term. For some “Community” means a Slack Channel, for others it is the number of users who have downloaded Kanjo. Community is not simply sharing information on the internet. Community in my definition is 1) An excellent sitcom that was too early for its time 2) Relationships based on mutual values, experiences and interests. I also love the definition from the Stanford Social Innovation Review. 

 Members of a community have a sense of trust, belonging, safety, and caring for each other. They have an individual and collective sense that they can, as part of that community, influence their environments and each other. (source)

Community is, therefore, a set of interrelationships, not a relationship between you and a customer, but their relationships with one another. So what are the best ways to measure this?  

Community Forum Traffic and Engagement 

What is interesting about Developers is that Community Forums are part and parcel of their day-to-day working experience. They are regularly searching StackOverflow for answers to questions, participating in conversations in Slack or Discord groups or passively reading HackerNews threads. 

If you believe that community engagement can be a powerful growth lever for your business and you want to invest in a forum, then you’ll be measuring performance based on different forms of participation in these forums (views, upvotes, posts, comments, feedback). You’ll also want to measure the happiness of users in these forums through NPS or a simple feedback tool like a “was this article helpful” dialogue. 

You can also measure external forum activity – like the AWS forums or StackOverflow – but those don’t give you a good sense of your own community’s engagement. It also doesn’t give you an understanding of the impact of your work. Sure, appearing on other forums at a higher rate is a great signal of success, awareness and widespread usage, but it doesn’t help you understand if your community-building efforts are making an impact. 

How not to measure Community Engagement

• Documentation Visits: If a Developer is using your product, it’s very likely they’re visiting your documentation often. This is not a bad thing, for extremely technical products it’s very common for developers to check and reference uncommon commands, or they could be reviewing documentation for a new feature. In this sense, Developers who are actively using documentation are active product users, but they might not be active members of the community. 

• Downloads or Accounts Created: This is a painful one for me to talk about because I truly do not believe that Account Creations or Open Source downloads are a measure of Community. The number of accounts created gives you a gauge of the product’s awareness in the market. Users who get value from the tool, either by hitting an activation metric or similar, are not members of your community either. They’re just users, and those measures tell you how effective the product is at getting a user to value quickly. So please, don’t use Accounts Created or Open Source downloads as a way to articulate your community size. I know it is very sexy in a board deck or as a way to showcase traction to investors, but they are not the same. 

• Blog Post Views: Just…no. The purpose of creating content on a blog is to drive awareness of your product, educate the developer ecosystem and drive conversions. Website views are not a measure of the community that you create around your product. 

Grow Awareness of Kanjo in the Developer Community 

Awareness is a great mission for DevRel teams. So much of what DevRel teams do is engage with developers as a champion of your product. Activities like conference attendance/presentations, Meetups and Hackathons are great tactics for driving awareness. 

Measuring awareness is SO HARD! There’s many proxies for awareness, but they all come with caveats, so in this case it’s very much a pick your poison. 

1. Organic Website Traffic: This can be a powerful measure of awareness, but if you have a marketing organization, some of the awareness can be attributed to their work, along with the work of DevRel. Additionally, Website traffic is a leading indicator of later stage performance metrics, so many CMOs don’t love using it as a primary metric 

2. Documentation traffic: As referenced above, unique Documentation traffic is a strong indicator of product usage, but it might not be the best indicator for awareness, since awareness happens before the product is used. 

3. Downloads or Accounts created: If you have a free or open source product, this is a great measure of awareness. Why? You’re measuring the signups, not the usage of the product. It’s a good measure that tells you abou general interest in your product. 

We want to build a product that developers love! 

Developer Love is challenging to measure because it’s so subjective. When you’re saying you want developers to love your product you want them to have a great experience and tell everyone about that great experience. Some folks might think this is an obvious goal for DevRel teams, but if you think about the roles DevRel plays in an organization, they’re not often building product. They might be doing support and therefore offering your customers a delightful customer experience there, but overall a Net Promoter Score is measuring the quality of your product and user experience. 

We want to grow the number of developers building ML applications on Kanjo 

1. Activation: This is a measure that’s become more widely discussed in recent years because of the growing popularity of Product Led Growth principles. I love Amplitude’s definition of Activation: “The process of getting new users to the aha moment, and then to the point where you get business value back from them is called activation. Activation matters because users who find value and invest within your product early on will likely stick around longer” How does a DevRel team impact this measure? Typically DevRel teams with this type of goal will own Support, Documentation and Education and have a hand in ushering users through the initial parts of building with the product. (I highly recommend reading Amplitude’s guide to product metrics) 

2. Downloads or Accounts created: This is a great secondary measure of impact because you’ll see how many developers you drive to your product, but you won’t be able to measure their success in using the product through this measure. 

Hopefully this is a good primer on DevRel KPIs that will help you think through your DevRel strategy. I would love to hear thoughts on this piece in the comments!