// Product development
From ideato something live
Discovery through to a product in production, with one team accountable for the whole thing rather than a design studio handing a file to a development shop.
Timeline
8 to 16 weeks
You get
MVP in production, repo, docs
Scope
Fixed after discovery
Platforms
Web and mobile
The expensive failure in product work is not building the thing badly. It is building the right thing badly scoped, or the wrong thing beautifully, and finding out after the budget is spent.
Discovery exists to make that cheap. It is the part that can end with us telling you not to build it, which is worth more than a proposal that agrees with everything.
What the work covers
Discovery and scoping
A short paid phase that ends with a scope, a fixed price and a written argument for what to cut. Sometimes it ends with a recommendation not to build.
Prototypes and MVPs
Something real enough to put in front of users, built so the parts worth keeping survive into production instead of being thrown away.
Mobile and web builds
Flutter for mobile, web where the web is the better answer. The choice is argued from your distribution and your users, not from what we prefer.
Design systems and interface work
Components, states and rules rather than screens, so the product can grow after we leave without drifting into five kinds of button.
Infrastructure, CI and release
Environments, pipelines and deployment set up as part of the build, so shipping on the last day is not the first time anyone tries it.
Handover and documentation
Written so an engineer who was not there can run the project on their second day. Your repository and accounts throughout.
What goes wrong with product builds
Not a list of things other people do badly. These are the ones worth checking before anybody quotes you.
- 01Discovery skipped to save two weeks, which is how a build finds its real requirements in month three, at ten times the price.
- 02Design and engineering split across two suppliers. Every gap between them becomes your problem to manage, and neither one owns the result.
- 03An MVP scoped as a small version of everything rather than a complete version of one thing. It launches doing nine things badly and proves nothing.
- 04No instrumentation, so after launch there is no way to tell which parts are used. The next round of decisions is then made on opinion.
- 05Deployment left until the end. The last week of a project is the worst possible time to discover how the thing gets to production.
Questions we get asked
A short paid phase, usually one to two weeks. It ends with a scope, a fixed price, and a written argument about what to cut. It can also end with us saying the project should not go ahead, and that has happened.
Often paired with
Your next product starts here.
Tell us what you're building, right here. Attach a brief if you have one — no commitment, no pressure, just a real reply.