We shipped an app to total silence — turns out we'd made every SEO mistake a technical founder can make
A couple of years ago, two developers and a designer (me being one of the devs) shipped Heap — a mobile app that turns voice notes, photos of business cards, and scribbled notes into proper CRM-ready leads. We were proud of it. It worked.
Then: no downloads, no traffic, nothing. Not "slow growth" — nothing.
Turns out we'd made the exact mistakes you'd expect from a team that can build software but has never thought about how anyone finds it: no keyword targeting, no content beyond a single landing page, no search intent alignment. We were writing about what we'd built, not what anyone was actually searching for.
We went looking for tools to fix it and hit a different wall — every SEO tool assumed we already spoke the language. Crawl reports we could run but not interpret. Audits that dumped 200 issues with no sense of what to fix first. Eventually we just built what we needed ourselves — that part's a separate story.
Full story (and the two failure modes that killed our growth before it started) here: https://blog.snippetflow.io/en-us/seo-for-startups-a-founders-guide-from-real-mistakes/
Curious if other technical founders here hit the same wall — what tripped you up, and did anything actually work to fix it?
Heap had three distinct search jobs hiding inside one product: business-card-to-CRM, voice-note-to-lead, and scribbled-note digitization. I’d give each job one focused page before building a broad content library, then use impressions as the first gate. No impressions means the language or intent is wrong; impressions without clicks means the promise is weak; clicks without installs moves the problem back into product onboarding.
Did you have a look at Snippetflow by any chance? Any suggestion?
I've only reviewed the positioning visible here, not a full Snippetflow product pass, so I won't pretend otherwise. The first suggestion is to make the landing page demonstrate one SERP-aware decision end to end: query, competing-result pattern, recommended change, and expected metric, using one real example. If the home page can't show why that output beats a generic audit, publishing more articles will send traffic into the same ambiguity.
You are absolutely right. And the landing page needs a lot of attention. At the moment we are focused on optimising the blog articles (just one for now, but we have quite a few ready to post). As a side hustle we don't have much time to invest on it, so we need to prioritise what we want to do carefully.
This resonates more than most SEO postmortems because you named the actual problem precisely: writing about what you built instead of what anyone searches for.
Technical founders almost universally fall into this trap because the product is the thing they understand most deeply. So the landing page becomes a features list, the blog becomes a changelog, and none of it maps to the language a potential user types into Google when they have the problem your product solves.
The CRM lead capture angle is interesting because the search demand almost certainly lives in problem-first language. Nobody searches "voice note to CRM app." They search "how to log leads from business cards" or "capture leads from phone calls automatically." Those are different pages with different copy and different intent, and each one is a separate acquisition channel if you treat it that way.
The 200 issue audit problem you described is something I see constantly. A crawler report is not an SEO strategy. The most valuable thing any audit can do is answer one question: what is the single structural issue most responsible for suppressing performance right now. Everything else is maintenance. Most tools give you everything except that answer.
One question worth asking in retrospect: did you have Google Search Console connected from day one? Even with zero traffic it starts collecting query data immediately, and that data would have shown you within weeks exactly what language real people were using to search for your solution. Free, built in, and almost always the most honest signal available to an early stage product.
Good writeup. The founders who document the real mistakes rather than the polished lessons are the useful ones.
The shift from "what we built" to "what people are searching for" is important. I'd keep validating whether SEO is really the bottleneck now, or whether searchers still need a stronger reason to choose your solution once they find it. Ranking gets attention; positioning earns adoption.