Website and software company. Portland, Oregon. Working since 2014.
Most teams hire twice — once for the website, once for the product behind it. We do both, so the two never drift apart.
We keep the scope narrow on purpose. Every engagement is one of four shapes, and the people who sold the work are the people who do it. Nothing gets handed to a delivery team you never met.
Structure first, then the surface. We design the information architecture before anyone opens a design tool, so the navigation is something you can explain to a new hire.
The interesting work is rarely the demo. We start with the slow, awkward part of the job and design the software around it, then earn the rest of the scope.
A fifteen-year-old CMS, a spreadsheet that four people edit at once, a vendor who will not document anything. We map it, replace what is worth replacing, and leave the rest alone.
Launch is the easy part. We keep a standing shift on your product, so small changes ship weekly and nothing waits for the next retainer negotiation.
Three projects, shown with the interface rebuilt in the page rather than a screenshot of it.
Tidewater sells technical marine gear to people who know what they need. Their old site treated browsing like a catalogue. We rebuilt the product model so that fitment, stock, and compatibility sit beside the price instead of three clicks away.
Checkout completion went from 41% to 68% in the first quarter, with 74% of that lift on mobile.
Calder runs eleven sites on a booking tool that could not represent a blocked chair, a travel gap between buildings, or the clinician who must be in two places. We rebuilt scheduling as rules the clinics themselves set, and put every automated change on the same screen as the person who authorised it.
No-shows fell 22% across the eleven sites in two quarters. The audit trail passed their compliance review first time.
| Time | Patient | Clinic | State |
|---|---|---|---|
| 09:00 | R. Alvarez | Northgate | seen |
| 09:30 | T. Nakamura | Northgate | in room |
| 10:15 | D. Okafor | Riverside | hold |
| 11:00 | M. Lindqvist | Northgate | confirmed |
Ridgeline grew by acquisition, and every regional office kept its own way of describing a shipment. We built the contract that lets all of them sit behind one interface, and wrote the documentation that let us stop being the only people who understood it.
Eleven systems now answer the same question the same way. Partner integration time went from six weeks to under a week.
Every legacy regional endpoint maps to this contract. Old keys keep working until you turn them off, and we will tell you which ones you can retire.
Every number below is one we can show you the working for. If a figure is not here, it is because we cannot yet defend it.
| Measure | Figure | What it counts |
|---|---|---|
| Launches shipped | 147 | Production releases, not prototypes or pitch demos. |
| Care clients who renewed | 91% | Of clients on a standing team at the end of year one, measured across the last 36 months. |
| Median time to first release | 31 days | From kickoff to something a real user can open. Longer projects ship sooner, not later. |
| Sites still running on our code | 132 | Live and maintained today. Seven have outlived the company that commissioned them. |
Figures describe Halden as a fictional stand-in. Replace with your own before publishing.
Four stages, in this order, every time. We will not start stage three before stage two has told us what stage three should be.
We talk to the people who will use the thing and watch them do the job today. This is where most projects change shape, and it is cheaper to change shape now than after the design is signed off.
A written problem statement you disagree with if it is wrong.
Scope, sequence, and the honest estimate. We show you what we are cutting and why, because a smaller thing that ships is worth more than a complete thing that never does.
A fixed scope, a named team, and a start date.
A staging URL from week three, and a demo every Friday whether or not there is good news. You see progress as software, not as a status document describing progress.
A working URL you can open any day.
Source, credentials, runbooks, and a walkthrough with the team who will own it. If you stay on a standing team, the handover is just paperwork. If you do not, it should still be clean.
Ownership in your name, documented, and tested.
We had been quoted twice the money by a bigger shop and got a better answer in a week. Halden told us the thing we asked for was the wrong thing, and then built the right one.
A paragraph is plenty. If we are not the right people for it, we will say so and usually know who is.
Start a project