Shipping and specialties
Developer funnels
A north star that reflects production value, and the cliffs between signup, first call and live traffic.
Define a north star that reflects production value, not vanity signups (product metrics). If 1,000 developers sign up and 40 make a production call, you do not have a marketing problem at the top — you have a DX cliff between key creation and first success, or between test mode and going live.
Concept
Work the funnel with the same discipline you would for a consumer activation path. Interview five developers who stalled; you will usually hear the same two blockers.
Funnel metrics
| Stage | Metric | Healthy signal |
|---|---|---|
| Acquisition | Doc visits → signups | Clear CTA, low bounce on pricing |
| Activation | Signup → first successful API call | Minutes; high conversion |
| Integration | First call → sustained test-mode traffic | They are building, not touring |
| Production | Test → live mode with real volume | Billing and value begin |
| Expansion | Second API product or higher tier | Platform, not feature |
Also track: error rate by endpoint, time-to-resolve for integration blockers, % of calls via SDK vs raw HTTP, webhook success rate, and documentation search queries with zero results.
Walking through a stalled cohort
1,000 signups, 40 production calls. Interviews show two blockers: unclear KYC to go live, and missing webhook SSL guidance. Fix those two. Do not buy more doc traffic until the path you already have converts.
Important
Free-tier abuse and credential leaks are product issues. Rate limits, clear quotas and dashboard alerts protect both you and the developer. Limits that surprise in production without dashboard visibility feel like betrayal.
Security UX belongs here too: scoped keys, easy rotation, dashboard alerts on anomalous traffic, and docs that teach "never commit the secret" without shame. A leaked key that racks up a five-figure bill is both an infra incident and a product-trust incident. Your response playbook (cap, notify, forgive first offense when reasonable) is part of the offering.
A strong story at work sounds like: "Median time to first 200 was 40 minutes; we cut it to 8 by fixing X; production activation rose Y%." That is product sense applied to APIs.
Common mistakes
- North star = signups. Production calls are the value.
- Surprise rate limits in production. Publish numbers, show remaining quota, educate before you block.
- Treating leaked keys as "user error" only. The playbook is part of the product.
Check yourself
1,000 signups, 40 live callers. Buy more ads? No. Interview the stallers. You likely have two repeated blockers on the path to production.
A leaked key racks up a five-figure bill. Whose problem is that? Yours and theirs. Cap, notify, and forgive first offense when reasonable. That playbook is DX.
Key takeaways
- Instrument the developer funnel through production usage, not just signups.
- Two repeated blockers, found in five interviews, usually beat another endpoint.
- Rate limits, quotas and leaked-key response are product surfaces.
Further learning
Next: API pricing and rate limits. Map: developer and API products.
Related
Developer and API products
Treating APIs, SDKs and docs as products: developer experience and adoption.
Product metrics and the north star
Choosing a north star that reflects customer value, decomposing it into input metrics teams own, and pairing it with guardrails. Start here for the map.
Developer experience
Developers adopt a path from curiosity to production, measured in time-to-first-call, not endpoint count.