Small commitment first. Always.
Nobody can price a build honestly before understanding what it has to do, so we sell the understanding first. No phase depends on you committing to the next, and everything we produce is yours whether or not you continue.
Four stages, and what each one produces.
The output column is the point. A phase that ends in a document nobody can act on has not ended.
No chargeConversation
You describe what you want built. We ask about volumes, exceptions and who does what today, and say plainly whether we are a fit.
You end up with: An honest answer, and a rough sense of shape and size.
Fixed price, 1-3 weeksDiscovery
We map the requirement, model the data, identify the exceptions and design the first release. Time with the people who will use the thing is essential here.
You end up with: A written scope, a data model, screen flows, a delivery plan and a fixed price for the first release. Yours to keep either way.
Fixed scopeFirst release
We build the narrowest version real users can do real work in, admin side included. You see working software weekly, not status reports.
You end up with: A running system, the repository, the database and the deployment - in your accounts.
MonthlyOperate and extend
Hosting, support, monitoring and the next round of work - prioritised by what the first users actually did rather than what the plan assumed.
You end up with: A maintained system and a backlog driven by evidence.
On every project: you own the code, data, domains and cloud accounts from day one; you see working software weekly; scope and price are fixed per stage; and if something is going to be late you hear it in the week it becomes likely, not the week it is due.
We start from the problem, not the platform.
These are the patterns we see most often. If one of them describes your week, the conversation is usually short.
IdentityYou cannot prove who is who.
Documents arrive over chat, verification lives in someone’s head, and nobody can reconstruct who approved what. We build verification with private document storage and an audit trail that survives the argument.
CoordinationYour system stops at the office door.
The people doing the work are on the road, on site, or on a shop floor. The system was never built for them, so the real state of the operation lives in phone calls and nobody can see it.
ScaleThe spreadsheet became the system.
It works until two people edit it at once, or until the person who built it leaves. Moving it into software is a data-modelling problem long before it is a screen problem.
FragmentationEvery system holds a different version of the truth.
Accounting says one thing, operations says another, and the customer-facing app says a third. Somebody reconciles them by hand every month, and everybody trusts the result a little less each time.
ReportingYou find out too late to do anything.
The numbers exist but arrive after the decision was needed - and a blank dashboard looks identical to a broken one. We make failure states explicit, so an empty chart never gets mistaken for good news.
InheritanceSomeone else built it, then left.
A half-finished codebase with no documentation and no tests, and nobody willing to touch it. We take these on - but only after reading it properly and telling you the truth about what we found.
Trust is an engineering decision, not a promise.
Private documents stay in private storage with short-lived signed access. Permissions are enforced in the database, not just hidden in the interface. Sensitive actions are logged with who did them and when. Your data and your code are yours - at the end of a project you hold the repository, the database and the accounts.