◆
Projects start immediately
A new client project is set up in a day, so a fixed-price engagement does not lose its first week.
Agency economics punish setup time twice: once in unbilled hours and once in the delivery days it takes from a fixed-price engagement. Both are infrastructure problems masquerading as process problems.
Used for client project delivery, preview environments and handover.
What this looks like in practice
A new client project is set up in a day, so a fixed-price engagement does not lose its first week.
Separation by arrangement, which matters more here than almost anywhere: the failure is visible to a client.
A URL per branch that a client can click, which removes most of the feedback cycle friction.
Ending an engagement cleanly, so a client can run it themselves if they choose.
For a studio, infrastructure setup is pure cost: the client is not paying for it, and in fixed-price work it comes directly out of margin.
On a fixed-price engagement, setup time comes straight out of margin, and on time-and-materials it is the line a client is most likely to question. Removing it improves the commercials and the first impression simultaneously, which is a rare combination.
Review links per branch, so feedback is on the thing rather than on a picture of it.
Separation strong enough that one client’s work cannot affect another’s.
Campaign sites that survive the day the client has been promoting for a month.
An ending that leaves the client with something they can run, which is easier to sell than a dependency.
Setup used to cost us the first week of every project. On fixed-price work that was most of the margin.
Same day in most cases. For fixed-price work that week is margin, so it is usually the first thing studios notice.
Yes, with a link per branch that they can click. Feedback on something running is both faster and better than feedback on screenshots.
Structurally separate. In this business a separation failure is visible to a client, which makes convention-based approaches a poor bet.
You export it and the client can run it. Being easy to leave turns out to be commercially useful more often than being hard to leave.
Start on something that will still be right at a hundred times your size, and pay now what a startup should pay.
Move in stages with the old arrangement still running, and cut over only once the replacement has carried real traffic.
Get capacity when the experiment is ready, keep what produced a result, and put models in front of users without a separate platform.
Keep everything at full fidelity, query it without a project, and stop deciding today what you will want to ask next year.
Ship the application your users touch, on an address that stays put, with every version one click behind you.
Build the tools your own teams need, at a cost that makes small improvements worth doing.
Begin with one workload. Keep what you have running. Decide whether to continue once you have seen it carry something real.
No migration required to begin. Nothing to tear down if you stop.