A message landed in my inbox eight days after we launched.
“Hey. Just finished the audit you sent. Genuinely useful, thank you. Two questions though. Where do I see this next month? Like, can I check back in and see what’s changed? And, slightly weird ask, but is there any way I can plug this into Claude? I run most of my workflow inside Claude Code now and switching to a dashboard feels like going backward. Anyway, no pressure, the audit was great as is.”
It wasn’t the only one.
In the past two weeks, dozens of messages like it have rewritten the roadmap of the company I just shipped.
This essay, and Episode 2 of Cited, is the live receipt of that.
SEOABLE is fourteen days old. It is already not the product I thought it was going to be. And I think the reasons why are interesting beyond just my company.
The bet, two weeks ago
Two weeks ago, I shipped SEOABLE. A single offer. Ninety-nine dollars, one time, no subscription. You buy it, we run a full AEO audit on your site, you get a report, you get a content plan, you get the schema and structure recommendations, you get the things you need to do this month to be more citable. Done. Move on.
The reasoning was simple.
The SEO industry has been built, for thirty years, on retainer logic. You hire an agency. You pay them three thousand, five thousand, ten thousand dollars a month. They send you a monthly report. You’re never quite sure what they did. The pricing model is structurally misaligned with your actual interests. The agency wants you to keep paying. You want the problem solved.
Ninety-nine dollars one-time was a deliberate inversion of that. Pay once. Get the work done. Move on. If the audit is good, you’ll come back when you genuinely need another one. If it’s not, we don’t deserve your money anyway.
Ninety-nine dollars was never about being cheap. It was about telling founders the truth: most of what they’d been told they needed was a recurring fee for a one-time problem.
That was the bet.
Then customers started using it. And the messages started coming in.
Three signals in fourteen days
What I want to walk through is not a story about SEOABLE. It’s a story about how quickly a product can change shape once it meets real customers, and what to do when that happens.
Three signals showed up in the first two weeks. Each one reshaped what we’re building, in real time.
Signal 1: They want continuous insight, not a one-time report
A customer buys the audit. We run it. They get the report. They love it. Some of them share screenshots on Twitter.
Then, about a week later, they come back.
“This was great. Where do I see it next month?”
“Can I see if anything has changed?”
“I made some of the changes you recommended. Is there a way to track if it’s working?”
“I love the blog post recommendations. I sent the first three to my writer. Can we do more?”
What I heard, underneath all of that, was something I had not planned for.
Founders don’t just want an audit. They want a relationship with the data. They want to know where they stand. They want to know if they’re getting better. They want to be able to take what we give them and put it to work.
The job-to-be-done, it turned out, wasn’t “audit my SEO.” It was “help me understand where I stand, stay current, and use the output.”
That sounds obvious in hindsight. It was not obvious when I shipped this two weeks ago. I genuinely thought a one-time audit was the whole product.
Signal 2: They want this inside their AI assistant, not in another dashboard
The second signal was the one I didn’t see coming.
A specific kind of message started appearing. Not from traditional SEO buyers. From a different kind of founder.
Founders who run their entire workflow inside Claude. Or ChatGPT. Or Cursor. Or their own custom agents.
“Can I use SEOABLE with Claude Code?”
“Is there an API? I want to pipe this into my own pipeline.”
“I don’t really use dashboards anymore. Everything I do, I do inside my assistant. Can SEOABLE just be a tool my agent calls?”
The first time I got that message, I almost dismissed it. I thought it was niche. A small slice of customers.
By the fifth message, I realized something else was happening.
The way founders work has shifted, faster than the SEO industry has noticed, faster than I had noticed. A whole generation of founders, especially technical founders, no longer log into tools. They have a primary assistant. Claude, GPT, Cursor, whatever. And every other piece of software in their stack either shows up inside that assistant, or it doesn’t get used.
For these founders, “log into our dashboard” is friction. For their agentic workflows, it’s a non-starter. You cannot ask an autonomous agent to log into a website, navigate a UI, click around, and copy a report into a chat window. That’s not how any of this works.
What these founders need is for SEOABLE to show up where they already are. Inside the assistant they already trust. As a capability, not a destination.
Signal 3: Sometimes they need a hand, but not a retainer
The third signal was subtle.
Most customers run the self-serve product fine. They buy the audit. They read the report. They make the changes. They move on. That’s the design.
But every so often, a customer hits a moment. A specific question. A specific situation. A piece of work they want help with that isn’t quite a full audit but is more than they want to handle alone.
Some have in-house teams already. They want a second opinion on strategy. Some have an agency partner. They love their agency for execution but want a more focused AEO perspective layered on top. Some are solo. They’ve done the work themselves and want an expert to look at one specific thing for an hour.
What none of them want is a five-thousand-dollar-a-month retainer they didn’t sign up for.
Saying “sorry, that’s not what we do” every time was leaving founders stuck. Worse, it was leaving them with the wrong impression: that SEOABLE was a product built only for self-serve, and once you needed any human in the loop, you had to leave us and find an agency.
That’s not what I want. That was never what I wanted.
What’s rolling out in the next few weeks
Three signals. Three responses. One philosophy.
One. The $99 audit, with proper reporting. Living reports, not static snapshots. Exportable. Shareable. Trackable over time. Founders can hand a SEOABLE report to a teammate, an agency, an assistant, or just keep it as a tracking tool. Same price. Better product.
Two. API access and an MCP server, with usage-based pricing. For founders who run their workflow inside Claude, GPT, Cursor, or their own agents. Same account as the $99 audit. New surface. Pay for what you run.
If MCP is unfamiliar: it stands for Model Context Protocol. It’s a new standard that lets AI assistants like Claude use external tools natively, the way they already use web search or code execution. Without MCP, you copy and paste between your assistant and external tools. With MCP, the assistant just calls the tool directly. SEOABLE shipping as an MCP server means that if you’re working inside Claude Code, you’ll be able to ask Claude to run a SEOABLE audit on a URL, and Claude will just do it. Inside the conversation. As a native tool call.
Three. Light-touch services. Targeted, scoped engagements. One-off support. A single piece of work, defined, delivered, done. Not retainer-by-default. Structured around the customer, not around recurring revenue logic.
The throughline is simple. Pay for value, not for time. Pay for what you extract, not for being on a list. Aligned incentives at every layer.
We’re not adding these because we want to upsell anyone. We’re adding these because two weeks of customer conversations told us where the gaps were. And we’d rather meet founders where they actually are than force them into one shape.
If any of those three sounds like you, seoable.dev is the place. The $99 audit is live there now. The API, MCP server, and light-touch services are rolling out in the next few weeks, and there’s an early access list.
What 14 days has actually taught me
There’s a more interesting piece of this. The lessons fourteen days of running a real product have taught me about building, not just about SEOABLE.
You can’t plan your way to product-market fit. You can only listen your way there.
Every meaningful product decision in the last fourteen days came from a customer message. Not a strategy doc. Not a roadmap. Not a planning session. A customer wrote me an email. I read it. I noticed the pattern when the third or fourth similar email came in. I changed what we were building.
That’s the whole methodology. There is no other methodology that works at this stage of a company.
Founders sometimes ask me how to validate a product before shipping it. I have started giving a different answer than I used to. The answer is, you can’t. You really can’t. You can talk to people. You can run a survey. You can do customer development calls. All useful. But you don’t actually find out what you’re building until people are paying you and using the thing.
The first two weeks of any real product are the most valuable two weeks of feedback you’ll ever get. Faster than user interviews. Sharper than analytics. More honest than any survey.
If you’ve been sitting on something for months, polishing it, waiting for it to be ready: ship it. Fourteen days of customers will teach you more than fourteen months of preparation.
Speed of listening matters more than speed of shipping.
Everyone says ship fast. What people don’t say enough is that shipping fast is only valuable if you’re listening fast.
A lot of founders ship fast and then stop listening. They have the product they built in their head, and the messages that come in get filtered through that mental model. The signal that doesn’t fit gets dismissed. The signal that does fit gets amplified.
The MCP request, the first time I got it, did not fit my mental model. The continuous reports request, the first time I got it, did not fit my mental model. I almost ignored both. I’m glad I didn’t.
If you’re shipping something soon, do this one thing. When you read your first customer messages, write down every request that surprises you. Every request you don’t immediately know what to do with. Don’t dismiss them. Don’t try to fit them into the product you already have in your head. Just write them down.
Then watch what shows up twice. What shows up three times. What shows up so often you can’t pretend it’s a coincidence anymore.
That’s where your roadmap actually comes from.
Pricing is positioning.
$99 one-time attracts a specific kind of customer. A founder who values aligned incentives. A founder who has been burned by a retainer. A founder who wants to do the work themselves and just needs the right diagnostic.
Usage-based pricing on an API and MCP server attracts a different kind of customer. A more technical founder. A power user. Someone who runs their workflow inside an assistant.
Light-touch services attracts a third kind of customer. A founder who has the resources for human help but is allergic to retainer logic.
Each pricing model is a filter. Each pricing model decides who shows up.
The question I’d push founders to ask is not “what should I charge.” It’s “who do I want to attract, and what pricing model would attract them and repel everyone else.”
Aligned incentives are the only durable moat.
Every other moat erodes. Technology gets copied. Brands get diluted. Networks get unbundled. Distribution channels get saturated.
The one thing that compounds, the one thing that gets stronger over time rather than weaker, is trust. And trust comes from aligned incentives. From being a company that genuinely benefits when its customers benefit. From a pricing model that doesn’t punish customers for not using enough, or reward them for staying when they no longer need you.
That’s not a marketing claim. That’s a business model decision. And it’s the decision I think more founders should be making.
Public building requires public learning.
This essay, and the episode it accompanies, is itself an example. SEOABLE is fourteen days old. The product is changing in real time. I am writing this and recording the audio version while still in the middle of figuring out what I just learned.
That used to feel risky. The kind of thing you don’t say in public. You wait until everything is finished, polished, working. Then you announce it.
I think that mode is over. Not because polished launches don’t work. They still do. But because the alternative, building in public, learning in public, admitting in public, is faster, more credible, and more interesting.
The version of SEOABLE you can buy today is not the version I’m most excited about. The version I’m most excited about is the one our customers are telling us to build. We’re shipping it in the next few weeks.
I’m telling you that because I think you should know. And because I think founders who build in public, who say the unpolished thing, who admit when their plan got rewritten by their customers, are the founders worth listening to.
This is what Cited is going to be. Every episode, every essay, a little more honest than the last.
If you’re shipping something soon
The single most actionable thing in this whole post:
When the first customer messages come in, write down every request that surprises you. Every one that doesn’t fit the product you have in your head. Don’t dismiss them. Don’t try to fit them into your existing model.
Then watch what shows up twice. Three times. So often you can’t pretend it’s a coincidence.
That’s where your roadmap actually comes from.
Where to follow along
The full audio episode is below (or wherever you get podcasts). It covers everything in this essay plus the founder lessons from fourteen days of running a real product.
seoable.dev is the place to follow what’s shipping. The $99 audit is live now. Early access for the API, MCP server, and light-touch services is open on the site.
Subscribe to Cited on Substack if you want each new episode in your inbox the moment it goes live. Next week, Episode 3: how AI engines actually decide who to cite. The technical mechanics. The signals. The five things you can change about your site this weekend that move you from invisible to indexed.
If this resonated, the most useful thing you can do is forward it to one founder shipping something this month. Two weeks of listening will reshape what they think they’re building. Better they hear that now than after a quarter of building the wrong thing.
Andrew Smith, Founder, SEOABLE Host, Cited
Cited is the SEOABLE podcast. Weekly, founder-to-founder, about getting found in the AI era.
