Process

A clear path from the first conversation to a system in production.

No long requirements phase, no big-bang launch. A working slice early, real users early, then steady weekly releases until the platform does everything it needs to, and a proper handover at the end.

1. Understand

We start with the people who do the work, not with a product catalogue. That usually means watching the spreadsheet or process in use, asking what breaks, and writing down the rules the business follows, including the ones nobody has written down. Where useful we run those rules over the existing data before building anything; on one scheduling project that surfaced 143 violations the old spreadsheet had never flagged. The output is a short, plain-language brief with a data model, a list of integrations and a first cut of what version one contains.

2. Prototype on real data

Within days there is something to click on, loaded with your real data and behind a real login. It is deliberately rough. The point is to react to it: the screen that is missing, the report that is in the wrong units, the rule that is actually different on Saturdays. Nothing in a prototype is precious, and nothing in it is thrown away either; the production system is built by extending it.

3. Build and ship weekly

The first production release happens early. From there features arrive weekly: a new report, a role, an integration, the phone app. Each release comes with what makes it trustworthy: reconciliations that tie out, tests for the rules, data-freshness indicators and monitoring. You see progress in the product, not in a status report.

4. Harden

Before the platform is relied on, it is reviewed: authentication and roles, token lifetimes and revocation, rate limiting, secrets handling, database exposure, backups, uptime and dead-man's-switch monitoring, data residency where it matters, and privacy rules such as what fields each role can ever receive. Findings are fixed, not filed. For systems that run collectors or feeds, every one of them must tell you when it stops.

5. Hand over and support

You own the source code, the hosting accounts and the documentation. A handover checklist covers how to deploy, back up, add a user and recover from the obvious failures. Support is a simple monthly arrangement, scaled to how much the platform changes. If your own team or another developer takes it on, the stack has been chosen to make that easy.

Commercial terms

Two models. Fixed scope for assessments, prototypes and well-defined builds, with a written output and a price agreed up front. Monthly for platform builds and ongoing support: a higher monthly fee during the build phase stepping down to a lower fee for support, with no setup fee and no long lock-in. Hosting runs in your own accounts and typically costs tens of dollars a month.

We will tell you early if we think the problem does not need custom software, or if it needs more than we can sensibly deliver.

Have something you need built?

Tell us the problem, the systems you already use and what a good outcome looks like. You will get a plain-language reply within one business day, usually with a first sketch of how we would build it.