Services
Four things we are asked to do
Most engagements combine two or three of these. We price the outcome, not the hours, and we tell you up front which combination your problem actually needs.
2 wks
To a written recommendation
The first engagement is deliberately small: we read everything, then put what we think in writing — including the case for doing nothing.
Fixed price, agreed before we start
Set against an outcome we have both written down, not against hours and not against a discovery phase that quietly becomes the project.
Your repos, day one
Commits, credentials and pipelines live in your accounts from the first day, not at handover.
Senior only
The people in your standups are the people writing the code.
We price the outcome, not the hours.
01 — Strategy
Product strategy
Two to four weeks. You get a written recommendation whether or not you hire us for the build.
We start by arguing with the brief. Most projects arrive with a solution attached — a platform to adopt, a framework to migrate, a feature list to deliver — and the expensive part is discovering that the problem was somewhere else.
So the first engagement is deliberately small and deliberately arguable. We interview the people who live with the current situation, read the code, look at what the data says, and come back with what we would do and what we would explicitly not do.
- A written recommendation you own, including the case for doing nothing
- Costed options rather than a single number
- A sequenced roadmap with the first slice sized to ship inside a month
02 — Engineering
Product engineering
A senior team inside your standups, not beside them. Go, TypeScript and Postgres.
We build the thing, in your repositories and your cloud, on the stack your team can maintain. The bias is toward boring, typed, well-tested systems that a new hire understands in a week — not toward novelty that needs its author forever.
You keep every commit, every credential and every deployment pipeline. There is no licence to renew when we leave, because there is nothing we hold.
- Your repos and your infrastructure from the first commit
- Typed end to end, with tests where they actually pay
- Runbooks and pairing so your team owns it when we go
03 — Data
Data and growth
Instrumentation first, dashboards second. Most teams have the second and are guessing about the first.
A dashboard nobody trusts is worse than no dashboard, because it makes a confident wrong answer feel like a fact. We build the measurement layer while the product is still cheap to change, then let the data settle the arguments that are currently being settled by seniority.
- Event instrumentation designed with the product, not after it
- Experimentation that reports uncertainty honestly
- Attribution that survives a change to the marketing site
04 — Platform
Platform and tooling
The work that makes every later project cheaper than the last one.
Design systems, content platforms, internal tools, developer environments. Unglamorous, and usually the highest-return work available to a team whose product team is blocked on something manual.
We scope these as products with their own users, because that is what they are — and the internal team is a real customer who will tell you when it is not pulling its weight.
- Design systems that survive their second and third designer
- Content platforms your marketing team can actually operate
- Internal tools that replace a spreadsheet someone maintains at weekends
Next step
Not sure which one you need?
That is the normal place to start. Describe the problem and we will tell you which of these it actually is — including when the answer is none of them.
We reply to every enquiry within two working days.