Approach

Approach is the word agencies use to sell a methodology. We use it for something duller and more useful: knowing what happens next, who is doing it, and what it costs to find out. Predictability is the product. A project that surprises you is usually one where somebody deferred a decision they should have taken in the second week.

There is one lifecycle. Discover — the company, the market, the customers, the processes, the risks. Define — strategy, product, experience, architecture, and a plan with costs and sequencing set against it. Build — design, development, integration and testing as one coordinated process rather than four handoffs. Launch and grow — ship it, measure it, and keep developing against what real use turns out to demand.

The first two stages are the ones you buy on their own, and that is deliberate. Everything after them is priced against what they produce, so quoting a build before the scope exists is quoting an assumption. The discovery covers Discover and Define, and it ends in a written scope, an architecture direction, a sequenced plan and a number — documents that are yours whether you continue with us or not.

Discover is not a reading exercise, and it is not only about the company. Half of it is the systems and the processes; the other half is the people the company sells to — who they are, how they search, how much time they will give it and what convinces them. The message and the design are built from those answers rather than from a preference expressed in a room. Chosen the other way they are a template with better typography, and the difference does not become visible until the traffic arrives and does nothing.

The Custom pages name two of these stages differently, and that difference is deliberate rather than an inconsistency. Define is called Architecture there, because on a bespoke system the act of defining is architectural: what the data actually is, which systems have to agree, what will be expensive to reverse. Launch and grow is called Scale, because a platform that already has users does not launch — it grows while people are logged in. One lifecycle, named for the work in front of it.

The four disciplines run across that lifecycle rather than sitting inside one stage of it. Strategy does not stop when the build starts; it is what the build is judged against. Growth is not a phase bolted to the end; the measurement it needs is configured before there is any traffic to measure. That is what one accountable team gives you and four suppliers cannot — the handoffs are internal, and so are the arguments about them.

What none of this promises is a fixed price for work nobody has scoped. Where the scope is defined — a site, a shop, an automation, someone keeping it running — the threshold is published and the date is real. Where it is not, the discovery is what makes it so, and anything found along the way is quoted before it is started rather than absorbed quietly and mentioned afterwards.