Designing a Group-Wide Lean Management System

The situation

A senior operational excellence leader approached us with a problem that will be familiar to anyone running a multi-site group: the business was not short of good practice. It was short of a way to use it.

Individual plants had built genuinely good improvement capability, each in its own way, each with its own language, measures and documentation. None of it travelled. A method that worked well in one facility could not easily be lifted into another, because there was no common structure to lift it into. Performance information did not link up either, so the group could see site-level numbers without being able to compare them meaningfully or see where the real opportunity sat.

The underlying issue was not a lack of effort or ability. It was the absence of a shared operating system — and the cost of that absence rises every time a group adds another site.

What we found

Three things, and the third mattered most.

Standardisation was missing, not capability. There was no need to teach these sites how to improve. There was a need to give them a common framework so that improvement in one place became available everywhere.

Performance visibility stopped at the site boundary. Each facility could describe its own position. The group could not reliably compare positions, which made it difficult to direct effort to where it would earn the most.

There was no standardised training pathway. This is the one that quietly undermines everything else. A management system that exists only as documentation decays, because the next cohort of supervisors and managers never learns it properly. Without a route from newcomer to practitioner to leader, the system depends on the people who were in the room when it was built.

What we did

We worked with the senior leadership teams across Operations, HR, Quality and Improvement — deliberately not Operations alone, because an operating system that HR has not helped design will not survive contact with how people are actually recruited, developed and promoted.

Together we designed the system end to end: the structure, the processes, the measures, the training that teaches it and the roll-out plan that lands it. That included a maturity assessment so sites could grade their own performance against a defined standard and see, concretely, what the next level required rather than being told they should improve.

The approach was deliberately culture-aware rather than purely technical. You cannot engineer a culture, but you can guide it, nurture it and expose it — and a lean management system is one of the better instruments for doing that, because it makes performance visible and gives people a shared language for talking about it.

We engaged 24 sites across the group and trained 147 people, from shop-floor practitioners through to site leadership. Rather than attempting a simultaneous group-wide launch, we proved the system at four pilot sites first — long enough to find what did not work in practice and fix it before the rest of the group inherited it.

Anna Lavender-Moore and I co-led the work across the UK and Europe over eighteen months. We co-led the internal teams rather than running the design ourselves and handing it over, which was a deliberate choice: the people who would have to operate and embed the system at each facility were the people who built it.

The result

The group replaced multiple localised systems with one common set of processes. That did two things at once. It raised the floor across the existing business, because a method proven at one site could now be adopted at another without translation. And it made the group materially easier to grow — a newly acquired facility could be brought onto a defined operating system rather than being left to invent its own or absorb one by osmosis.

The assessment gave the group something it had not had: a consistent way of answering how good any given site actually was, and what specifically it needed to do next. Sites worked towards the top band of the standard against staged milestones covering financial, quality and delivery performance.

We are deliberately not publishing the client's figures or targets. They are not ours to publish.

What they took away

The system, and the ability to run it. The four pilot sites were not a demonstration — they were the proving ground that made handover possible, because by the time the system went wider it had already survived contact with real operations and been corrected accordingly.

By the end of the roll-out the internal teams were operating and embedding it at their own facilities, training their own people through a defined pathway, and assessing their own maturity.

That was the point. A management system that depends on the consultants who designed it is a dependency, not an asset.

This engagement is also the ancestry of ProgressXL, the self-assessment platform we are bringing to market — built to remove the weakness that site self-assessment always has, by replacing subjective internal grading with a deterministic, published score.

Frequently asked questions

Does a group have to be large for this to be worth doing?

No, but it has to be more than one site, or be planning to be. The value comes from a common structure that lets improvement travel between locations. For a single plant, the same thinking applies at a much lighter weight — this is closer to our business efficiency work.

How long does something like this take?

This one ran eighteen months from design to completed roll-out across two regions. The design is not the long part. Embedding it — training people, running the assessment, letting sites work the standard — is where the time goes, and compressing that is usually a false economy.

Why involve HR in a lean system?

Because competence, progression and recruitment are what make a management system durable. If the pathway from new starter to practitioner to leader is not built into how people are developed, the system lasts exactly as long as the people who were trained on it originally.

Can you do this without naming our business publicly?

Yes, and it is our default. This case study does not name the client, and we do not publish client figures, targets or branding. If you would rather no account of the work appeared at all, that is equally fine.