For NHS trusts
You already paid someone to tell you what was wrong.
The review happened. The findings were accepted at board. The recommendations went into a plan, and then into the queue behind everything else competing for the same few people. Two years later the same issue is on the same board report.
The advice was not wrong. There was simply never anyone free to do it, every day, forever. That is the job we take.
The pattern
Analysis is a product. Delivery is a different product.
The NHS spends a great deal every year on advisory work, and most of it is competent. Reviews identify real problems and recommend sensible things. The recommendations get formally accepted — in board papers, in business cases, in published strategies.
Then delivery requires operational capacity the trust does not have. Not budget: capacity. A named person, doing a specific task, repeatedly, alongside everything else they already do.
So the recommendation stays a document. And because it was formally accepted, the trust is now accountable for an outcome it has no route to deliver.
We were not there for the analysis and we are not going to repeat it. We take the finding as settled and build the thing that carries it out.
| An advisory engagement | What we do |
|---|---|
| Identifies what should change | Carries out the change, daily |
| Ends with a document | Ends with software running |
| Evidence rebuilt per inspection | Evidence produced as a by-product |
| Priced by the day | Priced by the outcome, fixed |
| Needs your people to act on it | Needs your people to approve it |
| Ends when the report lands | Still running in three years |
The three objections, answered
What a CIO usually says in the first ten minutes
“We have paid for this once already.”
You paid for the analysis. It was delivered. What was never bought was the operational capacity to act on it — because that was not what you were buying.
You are not replacing that work. You are finishing it.
“How is this not just another supplier?”
Because the first engagement is small enough to approve without a business case, and it either meets the acceptance criteria you set or it does not.
You find out in four to six weeks, not eighteen months.
“Our systems will not support it.”
We work through the same screens your staff use — the remote desktop, the browser, the terminal window. Where a supplier publishes an interface we use it. Where none exists we do not need one.
No integration project. No vendor negotiation.
On the accountability point. The executive accountable for the outcome is usually accountable whether or not we are involved — that accountability came with the accepted recommendation. We are not asking anyone to take on something new. We are making an existing obligation deliverable.
How the engagement is shaped
Small first, and structured so you can stop
The programme is laid out in full from the start, so you can see where it goes. But you commit one stage at a time, and each stage has to earn the next.
| Stage | Duration | What you get, and what you decide |
|---|---|---|
| Pilot from £12,000 |
4–6 weeks | One workflow, working, against your own data. You set the acceptance criteria in week zero. If it does not meet them, the engagement stops here — and that is a designed exit, not a failure. |
| Departmental | 10–14 weeks | The same workflow across a department, or several workflows for one team. Scoped after the pilot, when we both know what we are dealing with. |
| Enterprise | 6–10 months | Trust-wide, integrated into your operational reporting, with the quarterly evidence that shows the original recommendation is now being delivered. |
| Ongoing | Annual | Monitoring, alerting, configuration changes, and the platform itself. |
We will give you the later figures at the first call — before the pilot, not after it. Nobody should agree to a first stage without knowing roughly what the second costs. And the pilot price is fixed: if the workflow turns out not to fit a pattern we have built before, we tell you at discovery and you pay nothing.
Before you call
Four questions that decide whether this is worth a meeting
We ask these on the first call anyway. Answering them now saves both of us the meeting if the answer is no.
Is it documented?
A board paper, an inspection finding, an accepted recommendation, a known backlog. Something written down that somebody already owns.
Is there an operational sponsor?
Not an interested director — the person who runs the team doing the task today and wants it to stop.
Is it a person reading a screen?
Repetitive, rule-based and currently manual, we can almost certainly do. Requiring clinical judgement, we will not.
Is now the right time?
Mid-restructure, mid-inspection or mid-leadership-change usually means defer. We would rather say so than bill you for a stalled engagement.
Three yes answers make a thirty-minute call worth your time. Fewer than that, tell us anyway — we will say so, and you will have lost half an hour.
The boundary we do not move
We do not build software that takes clinical action without a person in the loop, and we do not automate clinical decision-making. That is permanent, not a roadmap position. Everything we build stops and asks when it is not certain — and a second system independently checks that it did.