The honest version of how a project runs.
Including the parts most agencies leave out: what happens when an estimate is wrong, how you can stop, and what you walk away with if you do.
Six stages. Two of them are free or fixed-fee.
You find out whether we are any good before the expensive part begins. That ordering is deliberate. It is the only way a founder can reasonably commit to a build with a team they have not worked with.
- 01
Build review
30 minutes, freeA call with the engineer who would lead your project, not a salesperson. You describe what you are trying to do; we tell you what we would build, what we would not, and roughly what it costs. If we are the wrong team you will know by the end of the call.
A straight answer and a ballpark range - 02
Scoping sprint
1–2 weeks, fixed feeBefore anyone quotes a six-figure build, we spend a week or two turning the idea into an actual plan: architecture, data model, screen inventory, risks, and a milestone schedule with prices attached. You own the output whatever you decide next.
A technical plan, a fixed quote, and no obligation - 03
Design & architecture
2–5 weeksFlows, interface and the design system, built in code from the start. In parallel the foundations go in (repository, environments, CI, error tracking) so that the first feature lands on infrastructure instead of waiting for it.
Clickable product and a deployed skeleton - 04
Build in two-week increments
The bulk of the workEvery two weeks: a deployed environment you can use, a demo of what changed, and an honest note on anything that slipped. No status reports written to look good. You see the real thing on a real URL from the second week onward.
Working software every fortnight - 05
Launch & handover
1–2 weeksLoad testing, accessibility pass, security review, analytics, and the documentation a future in-house engineer needs on their first day. The repository, the accounts and the infrastructure are in your name. That is true from day one, not granted at the end.
Live product, documented, fully yours - 06
Afterwards
Monthly, cancel anytimeMost clients keep us on a retainer for new features, monitoring and the occasional 2am problem. Some take it in-house and we help them hire. Both are fine outcomes and we will tell you which one we think you should pick.
Support, or a clean exit
Estimates are wrong sometimes. Here is what happens then.
Any team telling you their estimates are always right is either inexperienced or not being straight with you.
We underestimated
On a fixed-scope milestone, the overrun is ours. We quoted it, we carry it. What we will not do is quietly absorb it by shipping something worse than we agreed. If the only honest options are more time or less scope, you will hear both and choose.
The scope changed
Normal, and usually a sign you learned something. We re-estimate the affected milestone in writing before any new work starts, so there is never a surprise invoice. Nothing gets built on a verbal 'while you're in there'.
A dependency failed
A payment provider changes an API, a client's legacy system is not what the documentation claims. We flag it the day we find it with options and costs attached, rather than burning a sprint hoping it resolves itself.
You want to stop
Monthly contracts take 30 days' notice; milestone contracts end at the current milestone. You keep everything built to that point, in your own repository, with the documentation written so far. There is no exit fee and nothing to negotiate.
How we behave when nobody is watching.
These are the ones we will put in the contract, which is the only meaningful test of a principle.
You talk to the engineers building it
There is no account manager between you and the work. The person on your kickoff call is the person writing the code, and they stay on the project until it ships.
Senior only
Everyone on the team has shipped production software for at least six years. Nobody is learning the fundamentals on your budget, and nobody needs a week to understand your codebase.
You own everything from day one
Repository, cloud accounts, domains, design files: all in your name from the first commit. No hostage situations, no licence that expires when the relationship does.
Working software every two weeks
Not a slide deck, not a percentage complete. A URL you can open and use. If something slipped, we say so in the same email as the demo.
We will talk you out of things
Half of what founders ask us to build first is not the thing that makes money. Saying so costs us billable hours and is the main reason clients stay for years.
Boring technology, deliberately
We pick stacks with ten-year track records and big hiring pools, because in three years you may want to hire for this codebase. Novelty is a cost, paid by whoever maintains it.
The awkward questions, answered.
It starts smaller than you probably expect. A written plan is $1,900. A single two-week sprint is $3,900 and you can buy exactly one. A launch-ready store is around $6.5k, a working MVP with accounts and payments around $11k, and a large multi-year platform runs well into six figures. We publish all of it because the alternative wastes your time and ours.
Tell us what you are trying to build.
Thirty minutes with the engineer who would lead the project. You will leave with an opinion on what to build, what to skip, and roughly what it costs, whether or not you hire us.
2 build slots open · Next start: early next month
What happens next
- 1You send a few lines about the product and the deadline
- 2We reply within one working day with times
- 3A 30-minute call, no deck, no sales script
- 4A written summary and a ballpark range within 48 hours