Six things that started as a paragraph in an email
A booking flow, a shop, a product with paying users, a tool that does the Tuesday morning job, a site that answers two questions, and a first week people actually finish.
About these
These are the shapes of work we do, written up as case studies. No client is named and no figure here was measured — when we have permission to publish real numbers and real names, they will replace these. The sites we have actually built, live and clickable, are on the projects page.
A booking app for a two-person clinic
Two practitioners, one shared diary, and a phone that never stopped ringing
They were taking every appointment by phone and losing evenings to rearranging them. They wanted people to book themselves without the clinic losing control of the diary.
A booking flow that takes about twenty seconds on a phone, with rules underneath it: how long each treatment takes, who can do which one, and the gaps they wanted kept free. Reminders go out the day before.
Most bookings now happen outside working hours, which was the whole point — those were the ones being missed.
A shop for a maker who keeps selling out
One person, a workshop, and a following that arrives all at once
Drops were being run through social media comments and a spreadsheet. Half the work was working out who had actually paid.
A shop built around the drop rather than around a catalogue: a countdown, a queue that holds an item while somebody checks out, and stock that cannot go below zero however many people press at once.
A drop that used to take a weekend of admin now takes the afternoon it is live.
A product taken to its first paying users
A founder with a spreadsheet that forty people were already begging to use
The spreadsheet worked and did not scale. They needed the smallest version that people could pay for, quickly, without betting a year on it.
Accounts, subscriptions, the two screens that did the actual work, and an admin area the founder could run the whole thing from. Everything else was deliberately left out of version one.
Live in ten weeks with paying users on it, and the second version was shaped by what those users did rather than by guesswork.
The Tuesday morning job nobody wanted
A small trade business with a very patient office manager
Every week, four hours of copying job sheets into invoices and chasing the ones nobody had paid.
A small tool that reads the job sheets, raises the invoices, and sends the polite reminders on a schedule. It flags the three or four that genuinely need a human and leaves the rest alone.
The four hours became about twenty minutes, and the chasing happens whether or not anybody remembers.
A site for people choosing between three firms
A practice whose enquiries all started with "I found you and then I found two others"
Their old site said what they did. It did not say why anybody should pick them, and it took eleven seconds to load on a phone.
Five pages, written with them rather than for them, built around the two questions every enquiry had turned out to be asking. Under a second on a phone.
The enquiries that arrive now already know what the answer to those two questions is.
A first week that people actually finish
A product with plenty of sign-ups and very few second visits
People were signing up and never coming back. Nobody knew where they were dropping out, only that they were.
We watched ten people use it, cut the sign-up from nine steps to three, and built the first-run screens around the one thing that made somebody come back — which turned out not to be the feature the team was most proud of.
Far more people reach the point where the product is useful. That is the only number that mattered here.
None of these people knew what they wanted when they wrote to us
Every one of them arrived with a problem and a vague sense of the shape of the answer. Working that out is the first station, and it is the part we are best at.