A person readsevery message.

It depends on two things: how many systems we have to connect, and how settled your process already is. That is why we start with a two-week first step rather than a number off a call. You leave it with a business case and a written estimate you can take to finance, and you keep everything we produced whichever way you decide.

The plan takes two weeks: ten working days from the first conversation to your decision. What comes after depends on scope. One process is measured in weeks; connecting several systems or migrating data is measured in months. The estimate states the range and the assumptions behind it, in writing, before you commit to anything.

Yes. Yours or ours, before any access is granted. A data processing agreement and an IP assignment are available on request as well. Named client work is only ever discussed in conversation, under the same terms.

You do, from the first commit, not on final payment. Code lives in your repositories and infrastructure runs in your accounts wherever possible. If you stop mid-project, you take everything built so far, with the documentation to run it.

For a defined scope — one process whose requirements we can pin down in the first two weeks — we agree the scope in writing and carry the estimate risk; changes go through a change order. Where the unknowns live in someone else's system, we propose capped time and materials instead, with a ceiling you approve and weekly visibility of spend. We tell you which fits before you ask.