Most people who call us already know something is wrong. They just can’t name it.
The symptoms are familiar. Jobs take longer than they’re quoted at. The same problems keep appearing and nobody can say why. Everybody is busy, nobody has capacity, and the monthly numbers never quite explain where the time went.
What they don’t have is a diagnosis. That’s the job.
A business process improvement consultant is not there to tell you that you have a problem. You already know. We’re there to find out exactly where it lives.
It starts with measurement, because most businesses aren’t measuring
The single most common thing we find is not a broken process. It’s the absence of any reliable measurement of the process at all.
This isn’t a small-business failing either. The RICS Construction Productivity Report 2026, based on responses from nearly 3,000 construction professionals, found that 22% of UK firms never measure productivity – the highest rate of any region surveyed. Only 13% measure weekly. Just 5% use industry benchmarks at all.
Think about what that means in practice. Four out of five firms have no external reference point for whether their operation is performing well or badly. They are making improvement decisions on instinct.
So the first phase of any process improvement work is establishing what is actually happening, as opposed to what everybody believes is happening. Those two things are almost never the same.
Mapping the process as it really runs
The next step is to map the process end to end – not the version in the handbook, the version people actually follow.
We do this by walking it. Sitting with the people doing the work, following a job from enquiry to invoice, watching where it stops.
What surfaces is almost always the same handful of things:
- Waiting. Work sitting in a queue because it needs a signature, a decision or a piece of information from somebody who doesn’t know it’s waiting.
- Rework. Something done twice because it wasn’t right first time, or because nobody told the next person it had changed.
- Duplication. The same information entered into three systems, because the three systems don’t talk to each other.
- Workarounds. The unofficial process people invented because the official one doesn’t work, which nobody has told management about.
- Handover loss. Detail dropped every time a job passes between departments.
None of these show up in a P&L. All of them cost money.
Quantifying the cost
This is where process improvement stops being an opinion and becomes a business case.
Once the process is mapped, each of those problems gets a number attached. How many hours a week does that waiting cost? How often does that rework happen, and what does it cost each time? What percentage of jobs are affected?
If you cannot put a number on a problem, you cannot justify the cost of fixing it – and you cannot prove you did.
This matters for a reason that goes beyond the immediate project. A business that has quantified its problems can prioritise them. Suddenly it’s obvious that the thing everybody complains about costs £4,000 a year, and the thing nobody mentions costs £40,000.
What we ask in the first conversation
Before any engagement, there are five questions that usually establish whether there is a case to answer.
- Which process, if it ran perfectly, would make the biggest difference to your numbers?
- How do you currently know whether it is running well or badly?
- When something goes wrong in it, who finds out, and how long does that take?
- What is the workaround your team uses when the official process doesn’t fit?
- If you fixed it tomorrow, what would you do with the capacity that freed up?
Question four is the revealing one. There is almost always a workaround, and almost nobody at management level knows what it is. It exists because the official process fails in a specific, repeatable way that nobody has escalated — usually because raising it once produced nothing.
Question five matters too. If there is no answer, the improvement will not be prioritised, because nobody will feel the benefit.
Redesigning, then proving it worked
The redesign itself is usually less dramatic than people expect.
Good process improvement rarely means replacing everything. More often it means removing three unnecessary approval steps, moving one check earlier so errors are caught before work is done rather than after, and giving one person clear ownership of a handover that previously belonged to nobody.
Then it gets measured again, against the baseline established at the start. That final step is the one most often skipped, and it’s the one that determines whether the improvement survives.
A worked example of the numbers
To make this concrete, take a process problem we see regularly, information arriving late, so work starts on the wrong version.
Suppose it affects one job in eight. Suppose each occurrence costs half a day of labour to correct plus a delay to the following stage. If a business runs two hundred jobs a year, that is twenty-five occurrences, twelve and a half days of avoidable labour, and twenty-five delayed downstream stages.
Attach a day rate to it and the number stops being an operational irritation and becomes a line item that justifies attention.
That calculation is not sophisticated. What makes it valuable is that most businesses have never done it, which is why the problem has survived so long. It has always been an annoyance and never a cost.
What it isn’t
Worth being direct about a few things.
It isn’t a software project. Technology sometimes forms part of the answer, but digitising a broken process gives you a faster broken process. The order matters, understand the process, fix the process, then decide what technology supports it.
It isn’t a redundancy exercise. In our experience the capacity released by better process almost always gets absorbed by work the business was already struggling to keep up with.
And it isn’t a report. A consultant who hands over a document and leaves has done half the job. The value is in the implementation and the proof that it held.
How long it takes
For a single process in a small to mid-sized business, the diagnostic and mapping phase is typically two to four weeks. Redesign and implementation depends entirely on scope, but the first measurable improvements usually land within a quarter.
The full benefit takes longer, because processes only stay improved if the behaviour around them changes. That is a different piece of work, and one we cover in our writing on embedding and sustaining change.
Where to start
If you recognise the symptoms at the top of this article, the useful first step is not to commission a project. It’s to pick one process, the one that causes the most friction, and establish two things. What does it actually cost you, and how would you know if it got better?
If you can answer both, you have the beginnings of a business case. If you can’t, that’s exactly where we’d start.
Take a look at our business process improvement → /business-improvement service, or read our case studies → /case-studies to see what the work looks like in practice.