How developer tool companies find companies worth selling to.
Devtool sales has a strange shape. Your buyer is an engineer who hates being sold to, your best signal is buried in a job description, and your real competitor is usually a script somebody wrote in 2021. The good news: engineering teams document their stack in public more than any other function, which makes this the easiest ICP type to prospect for with evidence.
Prospecting for developer tool companies, the way we do it: infer your ideal customer profile from your own site, pull candidates from the sources where your buyers leave signals, screen them cheaply, then read each survivor’s pages against a fixed set of checks. You get 8–25 companies, ranked, with a quote and a URL behind every claim. Try it on find-customer for free.
What your ICP usually looks like
- How it's defined
- tool- or role-defined, the best case (40–60% of crawled companies pass qualification)
- Size band
- 20–500 engineers; below that there is nobody to own the tool, above it procurement changes the motion
- Likely buyers
- Head of Platform · Staff / Principal Engineer · VP Engineering · DevOps or SRE lead
How we infer it from your site
On a devtool site we read the docs index (which languages and frameworks you support says who you're for), the integrations page (every integration is a technographic signal we can search for), customer logos and case studies (we seed lookalikes from these), and the 'built for' language on the homepage. A page that says 'for teams running Kubernetes' becomes a tech_adoption check; a case study with a Series B company becomes the size band.
Typical sellers on this profile: observability, CI/CD, database-as-a-service, auth and identity, feature flags, API platforms. You see the inferred fields before any research starts and can correct them.
Where your buyers leave signals
| Signal | Where we read it | What it proves |
|---|---|---|
| Tech adoption | Job descriptions that name the stack; HTTP Archive / Wappalyzer detections for web-visible tech | Presence of the complement (they run k8s, you sell k8s cost tooling) or the competitor is need evidence, not just timing |
| Hiring | Greenhouse, Lever and Ashby public boards, filtered to platform, SRE, DevOps and infra titles | Growth in the function that would own your tool, and the posting text is quotable |
| Lookalikes | Exa findSimilar seeded with your customer logos | Semantic similarity to a company that already paid beats any keyword |
| Stage and type | YC directory, Wellfound, GitHub orgs with public repos | Proves company type and stage cheaply; a public repo also reveals the stack |
| Stated goals | Engineering blogs and 'what we're working on' posts | 'We're migrating off X' on their own blog is the strongest problem evidence there is |
Typical fit rate: 40–60% of crawled companies pass qualification when the ICP is tool- or role-defined. Sources whose membership already proves something beat generic search, which has a 5–10% prior; the reasoning is in how to build a target account list from your website.
Queries that find them
- site:boards.greenhouse.io "kubernetes" "platform engineer"
- site:jobs.lever.co "terraform" "SRE"
- "engineering blog" "migrating" "observability" 2026
- "we use postgres" careers "series a"
What problem evidence looks like
A check passes only with a quote from a page we fetched. For developer tool companies, the lines that pass most often look like these:
- “We run our own Prometheus and it's becoming a full-time job”
- “You'll own our homegrown deploy scripts and help us move to something maintainable”
- “Currently on a self-hosted CI runner that goes down weekly”
Default for this profile: yes. For most devtools, a company already using a competitor is a good sign: they've bought in the category and can be displaced. We default displace_competitors to yes for this persona; flip it if you sell something that only makes sense greenfield.
Disqualifiers we apply
- Agencies and consultancies: they might use your tool but they aren't the buyer
- Companies that sell what you sell (competitor check)
- Pre-seed teams with no engineering hires listed (nobody to own it)
- Enterprises with a platform team of 40 (different sales motion; exclude unless that's you)
What a qualified card looks like
Illustrative example, not a real company
Series B fintech with a 30-person engineering org, hiring two platform engineers, job post says they run self-hosted Prometheus and it's 'becoming a full-time job'. Engineering blog announced a Kubernetes migration in Q2.
- “30 engineers across three product teams” — /about
- 2 open Platform Engineer roles — /careers
- “self-hosted Prometheus … becoming a full-time job” — /careers/platform-engineer
- “moving to Kubernetes this quarter” — /blog/infra-2026
- Funding — Not found on site; press release dated 2025-11 — /news
Two platform roles posted this quarter and a public migration in progress
Head of Platform
The score is arithmetic on check verdicts, not a model’s opinion; the weights are in our ICP scoring formula.
Questions devtools ask
- Does find-customer detect what technologies a company uses?
- Partly. We read technology mentions on the company's own pages and in its job posts, and quote them with the URL. Web-visible technographics at scale (HTTP Archive detections) are on the roadmap. We never infer a stack from a logo wall.
- Can it find companies using a specific competitor of mine?
- Yes, when the competitor is named in public: job posts, integration pages, engineering blogs. Tell us the competitor in the ICP and whether their presence should count for or against the company.
- Do you return engineer contact details?
- No. Company-level results only: the company, why it fits, the likely buyer role and the evidence. You decide how to reach out.