Our method
Diagnose, contain, fix, prove
We start on the floor, not in the boardroom, because that's where a handing error, a wrong fixing or a missing part begins its journey from order to site. Containment in weeks. Fixes in months. Proof throughout.
- Diagnose
- Contain
- Fix
- Prove
Day 30: review the numbers, then carry on or stop.
-
01 · Weeks 1–3
Diagnose
Know where the money is going, in pounds, and why.
We walk the warehouse, the office and a live site, read every failure record, and group them into families by cause and cost. For each family we ask two questions: why did it happen, and why didn’t our checks catch it?
-
02 · From week 3
Contain
Stop the bleeding in weeks, not after a six-month project.
A few deliberate checks (at goods-in, at dispatch, a hold point before more work is made) and one channel for site issues, so the problem stops reaching customers while the real fix is built.
-
03 · Months 2–3
Fix
A few fixes that remove whole categories of problem.
Each change is aimed at a whole family of failures, not a single incident. Cheapest control first: a one-page specification record, a survey hold point, written supplier codes. Every check has to earn its place; nothing becomes a second system to fill in.
-
04 · Throughout
Prove
You’ll see it in the numbers, or you can stop.
Every fix has a measure and a target, and an agreed, conservative cost per incident the board believes. A short weekly update, and a review at day 30 where you decide whether to carry on.
What makes this different
-
We start with the actual work, not the org chart
We walk the floor, read the records and talk to fitters and the office team. If you've had consultants hand you a slide deck, you'll notice the difference.
-
We've done the jobs
Production and finishing, an assembly team, specification of access equipment and ironmongery, fire-protection surveys and remediation, quality management in manufacturing. We know what a goods-in bay, a site survey and a supplier argument look like.
-
No blame
“Operator error”, “supplier fault” and “more training” are not accepted as causes. Your team tells us the truth because nobody is being hunted.
-
Lean by design
Every check has to earn its place. Nothing we add becomes a second system to fill in, because a business this size can't afford the baggage.
-
Proof before closure
A fix isn't done until the numbers show it worked. That's what separates this from “we wrote you some procedures”.
-
Front-loaded, with a break point
Most of the effort is in the first month, and there is a review at day 30 where you can stop. That takes the risk out of saying yes.
Two points, not one
Most repeat errors have two causes: the point where the error is made, and the point where it should have been caught but wasn't. We find both, and fix both with your team.
Where software fits
We use AI and our own quality platform, Doctilla, to speed up the analysis and keep progress visible. You never need to buy software to work with us.
What this looks like in a real business
The method is the same on every engagement; the work is not. Here is the shape of it, so you know what you are agreeing to before the first conversation.
-
We spend time where the work happens
Generally on the floor, in the warehouse, at the desk where the quoting is done, or with the supplier in question. We talk to the people doing the job, not only the managers describing it, and we follow real examples through the business rather than reading the process document. The process document is usually what is supposed to happen.
-
We get the numbers before proposing anything
How often does this happen, what does it cost when it does, and where does it start? Until those are answered, any proposed fix is a guess. This is the part most businesses have never done, and it is usually the part that changes what everyone thought the problem was.
-
We change the process with your team, not for them
The people who do the work are the ones who know why the workaround exists. Changes are made with them, tested on real jobs, and adjusted when the first version is wrong, because it usually is.
-
We measure before and after, and report it
The measure is agreed at the start, against where the business actually is now. Progress is reported regularly, and nothing is closed as fixed until the data says it worked, not until everyone is tired of talking about it.
What it requires from you
Being straight about this saves everyone time. An engagement works when the business can provide:
- Access to the people doing the work, and permission for them to speak plainly.
- Whatever records exist: error logs, complaints, remedial visits, credit notes, scrap reports.
- A decision-maker who can agree a change without a committee.
- Willingness to find that the stated problem is not the real one.
- Support for the people who have to adopt the change afterwards.
What it does not require is a specification, a business case, or an existing quality function. Most engagements start with a description of what is going wrong.
Questions about the method
Do you blame our people?
No. Most repeat errors are caused by a process that lets the error happen and then lets it through. We fix the process, not the person.
How do you measure success?
Before and after. We report weekly and close nothing until the data says it worked.
What if it is not working?
Most of the effort is in the first month, and there is a review at day 30. If the numbers are not moving, you can stop there.
Will this add admin for my team?
No. Every check has to earn its place, and nothing becomes a second system to fill in.
See what your rework really costs.
Start with a free Margin Leak Review: about an hour on a sample of your own records. If the numbers justify it, a three-week Rework Cost Diagnostic puts a £ figure on every family of problems.
Free. About an hour. You keep the findings.