Running an improvement programme: pipeline, funding and management attention

Somebody has to own the pipeline without owning the improvement

A coordinator or small central team is genuinely useful for triage, scheduling, chasing overdue actions, collecting evidence and making sure two areas are not solving the same problem separately. The trap is that the same team gradually becomes the department that does improvement, at which point line managers hand problems over and go back to firefighting. Draw the line explicitly: the improvement function facilitates, teaches and tracks, while the area manager owns the result and reports on it. If the coordinator is presenting an area's results on its behalf, ownership has already moved to the wrong place.

A triaged backlog rather than a suggestion box that fills

Everything raised needs sorting into a small number of routes: change it now with the team, needs an engineer or a fitter, needs money, needs a decision from another function, or declined with a reason. Each route has a different owner and a different expected pace. Keep the backlog visible on a board in the area rather than in a spreadsheet only the coordinator opens, and show what has moved. The single most damaging failure is a growing list with no visible movement, because it demonstrates that contributing achieves nothing and the flow of contributions stops.

Small money quickly beats large money slowly

A great many worthwhile changes need a bracket fabricated, a bin bought, a bench modified or a fitter for an afternoon. Routing those through a capital process guarantees they die of delay. Give supervisors a modest discretionary spend they can commit without approval, and a standing allocation of maintenance hours for improvement work. Anything above that threshold goes through the normal route, but the threshold should be set high enough that most workstation-level changes fall below it. Plants that get this right are recognisable by how many small physical modifications appear between visits.

The cadence that keeps work moving

Improvement dies in the gaps between reviews. A short standing session at the board, in the area, at a fixed time, with the actions and their owners visible, does more than an elaborate monthly report. The agenda should be the same every time: what moved, what is stuck, what is needed from outside the area. The scarce resource being spent here is management attention, and it is the honest measure of how serious the organisation is. An operations director who attends most weeks is funding the programme more effectively than any budget line.

Reporting benefits without manufacturing savings

Claimed savings that never appear in the accounts destroy a programme's credibility faster than admitting a project failed. Be disciplined about what is being asserted: a change in the plant's cost base is different from an avoided cost, which is different again from freed capacity that has not yet been used for anything. Do not add up hours saved across many small changes and present the total as money. Agree the counting rules with finance in advance, report physical measures such as elapsed time, stock level and defect frequency alongside any financial claim, and never count the same benefit in two projects.

Frequently asked questions

How many people should a plant dedicate to improvement work?

Fewer than most organisations expect, because the work is meant to be done by line teams rather than by specialists. A single capable facilitator supporting several areas is a reasonable starting shape for a medium plant, with additional capacity added only when areas are genuinely competing for support. A large central team is usually a symptom that line management has delegated the responsibility, and it tends to produce impressive documentation alongside a floor that has not changed.

Should improvement projects have a formal business case?

Scale the requirement to the spend. Anything a team can do with materials on hand and an afternoon should need nothing more than the supervisor's agreement. Anything requiring capital should go through the same process as any other investment, without special pleading because it carries an improvement label. The damaging middle ground is requiring a written case for small changes, which consumes more effort in justification than in doing the work and quietly stops people trying.

What is the right way to report programme progress to a board?

Show a small number of physical measures the board can connect to commercial outcomes, such as delivery performance, elapsed time through the plant, defect escapes and stock held, along with a short account of what physically changed and what is currently blocked. Avoid presenting activity counts, training numbers or audit scores as results. Boards lose interest in improvement programmes when the reporting is about the programme rather than about the business it was meant to change.