Information architecture
Turning a business model into something a person can navigate. Sitemaps, content models, taxonomies, and the nav structure that lives on top.
What we do · One of three lenses
The product people actually use. Most "training problems" aren't really training problems. Users keep clicking the wrong button, support tickets cluster around the same handful of screens, and every new feature seems to break an old workflow. UX is the layer where most of those failures actually live.
Inside an enterprise product, every business rule, every workflow step, and every regulatory constraint either becomes legible on a screen or becomes a wall. UX is the layer that decides which one happens.
It's not decoration. It's not "make it pretty." For people who use this software all day, the choice between two button labels can turn a thirty-second task into a three-minute one. Multiplied across hundreds of users and tens of thousands of transactions, that's a real number. We design with that number in mind.
Each artefact answers one risky question. We pick the cheapest one that does the job, and stop when the question is answered.
Turning a business model into something a person can navigate. Sitemaps, content models, taxonomies, and the nav structure that lives on top.
Black-and-white structure, before colour and polish become a distraction. Used when we need to argue about logic, not look-and-feel.
Running designs you can put in front of real users. Increasingly that's running code, prototyped alongside your engineers.
The visual layer once the structure is settled. Pixel decisions, type, motion, and the small bits that make a product feel finished.
Moderated, contextual, hypothesis-driven. We watch real users meet the work and feed findings back to your engineers and ours.
So the next twenty screens stay coherent. Tokens, components, and the documentation that keeps the system honest as it grows.
The three lenses we use, UX, business process, and service design, each look at a different layer of the same product. UX is the layer the user touches. The other two are the workflow that lives behind the screens, and the broader experience around them.
The product people actually use.
Sibling lensHow the work moves behind the product. The roles, steps, and handoffs the screens have to support.
Sibling lensThe whole experience, end to end. The people, channels, and support around the product.
A great workflow can still produce a frustrating UX if the screens ask for the same thing three times. Great UX laid over a broken process is wallpaper. We use whichever lens the risk asks for, and most engagements end up needing all three.
The same five-phase process underpins all three of our lenses. We've spent over two decades refining it to surface the right problems early and keep users in the loop until launch. Here it is with the UX work and artefacts in the foreground.
We start with the business question. The success criteria we agree on here become the yardstick for every later choice.
We talk to your users in their environment, and to your business about market and constraints. The output is a hypothesis list.
We diverge wide before we converge. Sketches and journeys, multiple competing directions, until the room can argue with the work.
We pick the right fidelity for the question. Increasingly that's running code, prototyped alongside your engineers in their codebase.
Moderated, contextual, hypothesis-driven. We watch real users meet the work and feed findings back to your engineers and ours.
Engineers are in the room from Plan through Test. We loop back through Gather until users tell us it's right.
Tell us what you're seeing. We're good at finding the screen-level shape behind a vague brief.
Talk to us