The plan

Each pillar carries the same four fields: current state, intervention, how I’d know it worked, and what I don’t know yet. The fourth field is what makes the rest credible, and it’s the one that’s tempting to drop.

One exception to the badges, stated once. Every number about OpenRouter on this page carries a provenance grade. The ones about my own prior work don’t — the vocabulary grades what I could measure on your surfaces, and none of it reaches my own record. Where those appear I’ve said whose they are and left them unbadged, rather than dress them in a grade they haven’t earned.

Self-serve growth engine

Current. QUOTED 10M users, 250k+ apps, 200T+ tokens/month, “nearly all of it starting in self-serve.” UNKNOWN Signup→first-purchase conversion, ARPU, and gross margin per routed dollar. Those three govern every target below and I have none of them.

Intervention. Instrument end to end before buying against it: signup, key creation, first API call, first credit purchase, second purchase, 30-day retained spend. Buy against first credit purchase, never signup. A signup with no key isn’t a user.

How I’d know. Blended CAC per credit-purchasing user, payback in months of gross margin, reported against a forecast set at quarter start — not against last quarter’s actual.

UNKNOWN What I don’t know. Whether activation is instrumented at that granularity. And if ARPU is bimodal — a long tail of $5 hobbyists against a short head of production apps — blended CAC is the wrong metric and this section needs rewriting around segments.

Answer-engine and search

Current. Your machine-readable surface is advanced. The gap is that per-model usage data renders and isn’t crawlable.

Intervention, in ascending effort.

1. Expose what already renders. MEASURED Per-model /apps, /activity, /providers, /uptime return 200 and appear in zero sitemap entries. Days of work, not a content programme. Highest value-per-hour item in this document.
2. Harvest the community corpus into canonical public answers. MEASURED The #models census alone is 1,027 threads of people comparing models on real workloads — the experiential signal a comparison template can’t carry.
3. Turn the integration directory into an outcome library. MEASURED 54 tools are named; no page says what any of them achieved.
4. Test the summarisation hypothesis before acting on it, on the panel below.

How I’d know. Citation share, not ranking: how often OpenRouter is named and cited by answer engines across a fixed panel of high-intent routing questions, re-run monthly against a baseline.

What I don’t know. Whether Disallow: /seo/ conceals an existing programmatic effort. And I have no baseline panel yet — building it is week-one work, not a claim I can make today.

UNKNOWN I’d bring an SEO partner for the technical half of this pillar. My depth is the measurement and AI side; thirteen years of enterprise technical SEO is not something I’d claim.

Conversion rate optimisation

Current. MEASURED The anonymous model page offers Try this model, Playground, Compare — and no account or credit CTA; authenticated, Create API Key appears. MEASURED openrouter.ai/signup returns HTTP 200 titled “Author Not Found” (the path collides with the /{author} namespace; the working path is /sign-up). It carries noindex, so the search impact is contained — the conversion impact isn’t.

Intervention. Standing experimentation cadence on the credit-purchase flow first, because it’s the only step where money changes hands and therefore the only step where a lift is unambiguously revenue. Start with the coupon cohort as a natural experiment: what distinguishes the people who came back from the ones who didn’t. UNKNOWN Which of them is which. Nothing visible from outside splits that cohort, and I’d rather start where the answer is missing than where I could already see it.

How I’d know. Hypothesis → test design → readout → decision, logged per test, primary metric and minimum detectable effect pre-registered. Tests that can’t be powered ship as judgment calls and get labelled as such rather than dressed up as experiments.

What I don’t know. Per-step traffic volume, which decides whether these surfaces can be A/B tested at all or whether the honest answer is sequential testing with geo splits.

Prior work. I reduced registration friction 35% through a data-driven intake redesign for first-generation and adult learners, and improved program retention 22% through friction analysis on an 80+ participant annual pipeline. Different building, same problem class.

Paid media, and the $5M

The gap, plainly, once. QUOTED The req names a “paid budget of $5M+ for 2027” and a director-level “player/coach” to own it. I have not run a paid-media P&L.

What I have held is budget accountability at two scales. Owned end to end: a $500K studio buildout — my own capital project, scoped and delivered on time and under budget. I set the allocation, chose the vendors, carried the variance. Operated inside: a campus construction and full-campus relocation programme, where my authority was variance tracking and executive reporting, not approval over allocation. I’d rather state that than let a larger number imply a signature I didn’t hold.

Allocation under constraint, variance discipline, vendor and agency governance, and executive spend reporting all transfer, and the posted paid line explicitly includes QUOTED “sponsorships” — bought, negotiated and governed like vendors. The return-optimisation loop does not transfer: a capital budget has fixed scope, a paid budget is continuously reallocated against a return signal.

So rather than claim the credential, here is the reasoning it’s a proxy for.

Governance I’d set on day one.

• Buy against first credit purchase, never signup.
• Every line gets a CAC ceiling before it gets a dollar. UNKNOWN The ceiling needs ARPU and gross margin, which I don’t have — so the rule is stated here, not the number.
• Kill rule, pre-committed. Two quarters and $100K minimum spend to prove out. Above $100K cumulative, CAC persistently >1.5× blended target ⇒ cut, reallocate to reserve. Written down before launch so it isn’t relitigated when the failing channel is someone’s favourite.
• Incrementality. Podcast/creator and OOH are not click-attributable and must not be judged on last touch. Geo holdout on both, designed before spend, matched controls, readout date fixed in advance.
• Reserve stays unallocated through Q1 on purpose. A plan with no slack can’t respond to its own results.

INFERRED Every number in that table is a starting hypothesis with a kill rule attached, not a forecast. The forecast is the first quarter’s work.

Measurement and attribution

Current. UNKNOWN What exists today.

Intervention. Sequence matters: event schema first, then attribution, then incrementality. Attribution built on an inconsistent schema produces confident wrong answers, which are worse than no answer because they get acted on.

How I’d know. One number for “customers acquired this month,” agreed across marketing, product and finance. The stack is working when nobody keeps a private spreadsheet because they don’t trust the dashboard.

What I don’t know. Whether the warehouse can join marketing touch to usage revenue at account level. If it can’t, that’s the actual first project and it’s an engineering dependency — I’d say so rather than build attribution on sand.

Prior work. Built automated reporting infrastructure that removed 40+ hours of manual aggregation per month, and executive dashboard reporting on milestones, budget variance and operational continuity.

Developer relations

Current. Fast first response on one surface, structural silence on the other — the two failure modes documented in the evidence section.

Intervention.

1. Per-surface response measurement, right-censored. A two-hour-old thread with no reply is new, not abandoned, and any metric that conflates them is wrong in the flattering direction.
2. Post-incentive retention as a named metric. Same coupon spend, cohort tracked to 30-day retained usage.
3. Harvest to publish — route recurring answers into public canonical pages.
4. Staff the moment, not the notice. Treat #events and lu.ma/openrouter as one calendar with a bridge.

How I’d know. Aged no-reply rate per surface trending down. Coupon-cohort 30-day retention trending up. Second-event attendance by first-event attendees — the only event metric that distinguishes a programme from a party.

UNKNOWN What I don’t know. Whether the 77% post-launch no-reply rate is capacity or deliberate triage. Those call for opposite interventions, and I’d want the team’s read before assuming the first. Nothing here is a judgment about anyone’s discipline — a fast-answering community with a measurement blind spot is a different thing from a neglected one, and this is the former.

Channel and partnerships

Current. QUOTED The second posting asks its holder to “Define and execute OpenRouter’s global indirect sales strategy across cloud marketplaces (AWS, GCP, Azure), system integrators, distributors, resellers, and technology partners”, and to “Establish OpenRouter’s presence on major cloud marketplaces”. INFERRED Establish is the operative verb — this is a motion being started, not scaled, which is the same position the marketing function is in. Two zero-to-one functions with one shared dependency, and the dependency runs from partnerships to marketing rather than the other way round; the role section sets out the counts behind that.

Intervention.

1. Treat marketplace listings as an answer-engine surface, not sales collateral. An AWS, GCP or Azure listing is an indexed page with a title, a structured description and a comparison table, read by the same machine readers as a model page. It belongs to the AEO/SEO system above, with the same template decision applied — one owner, one content standard, two indexes.
2. One pipeline definition, before the first partner is signed. Deal registration writes into the same warehouse join as marketing touch, so a deal that is both partner-sourced and marketing-touched is one row with two attributes, not two claims on one number. Doing this after the first contested deal costs an argument nobody wins.
3. Co-marketing as harvest, not campaign. The joint case studies the second posting asks for are the same builder outcome stories the developer-relations pillar already collects. Sourcing them from the community rather than commissioning them makes partner marketing a by-product of work that pays for itself twice.
4. Pre-commit the partner tiering and the kill rule together. Tiers set what a partner earns; the kill rule sets when recruitment stops. Written at the same time, they’re one policy. Written apart, the second never gets written.
5. Name the hire instead of routing around it. QUOTED The second posting asks for “Existing relationships with enterprise GSIs (Accenture, Deloitte, Capgemini) and hyperscaler partner organizations” and a “History of achieving significant partner revenue targets (30%+ of company revenue through indirect channels)”. I have neither. Unlike the paid budget, which I can answer by allocating one on paper and showing the reasoning, relationship capital has no paper substitute — it is the first hire under this pillar, and the marketplace and measurement work above is what makes that hire land into a system rather than a blank page.

How I’d know. Partner-sourced pipeline that survives the same attribution test as every other source — reported net of marketing-touched overlap, not gross. Marketplace listing to signup conversion, measured per listing. Time from partner signature to first transaction, which is the number that separates a signed logo from a working channel.

UNKNOWN What I don’t know, and it may reshape this pillar entirely. Whether an indirect motion built at OpenRouter still makes sense once the Stripe transaction closes. Stripe already has hyperscaler relationships, a partner organisation and a marketplace presence, and the cheapest version of this pillar may be to route through theirs rather than build a parallel one. I can’t see that from outside, and the second posting predates the announcement. It is the first question I’d ask, not something I’d assume either way.