A Fictional Brand, Built So A Real Product Could Be Tried

This is not a client story and it should not be read as one. The brand in it does not exist. We invented it, and the numbers describe our own demonstration environment, not anyone's business results. We are showing it because the problem it solves is one we keep meeting at clients, and because this is the only version of it we are free to describe in full.

The situation was a rebuilt platform with nothing in it. No data, no brand, no story. Demonstrations were running on leftover fixtures from two earlier tenants whose theming and assets had been lost anyway, which is the worst of both worlds: it does not look finished, and what it does show belongs to somebody else. The usual alternative is to point the demo at a real customer's account, which works right up until the moment it does not.

So we built a brand instead. Not a logo and a colour, a whole organisation. A brand guide. A compensation plan, because the product sells through a field of consultants and a demo without one answers no interesting questions. A full site information architecture and product page copy. Six product photographs. A roster of eighteen seeded identities covering demo consultants, quality assurance accounts, a video persona and real early testers. A deterministic database seed, so the environment comes back identically every time.

The part that made it useful rather than decorative is that we promoted the fiction to production infrastructure. It has its own route group inside the application, selected by a hostname switch, its own guest checkout, and a live public domain with a valid certificate. Then a demo readiness document, a reset script and a runbook, so somebody can put it back to a known state ten minutes before a call.

Which meant we could do the thing the whole exercise is for: give two external testers real accounts on the live environment, each with a public storefront and a creator application, and let them try the product rather than watch a recording of it. Those accounts were verified against the deployed environment rather than assumed from the seed.

It also generated an honest list of landmines, which we kept. An office network that sinkholed the domain and needed a phone hotspot. A bare hosting hostname that does not render the brand home page. A reporting view that shows nothing until you widen the date range to all time. A reset script that cheerfully deletes hand added test users. A first demo video set that was rejected before anything was generated and reframed entirely. None of that is in the brand guide, and all of it is what the runbook is for.

The transferable idea is small and we think undersold: build a credible synthetic tenant, to production standards, so a product can be evaluated without exposing a real customer's data. Most teams either demo something obviously fake, or demo something they should not be showing.