Why Process Improvement Projects Stall After the First Workshop
The workshop goes well, then nothing changes. Here is why process improvement projects lose momentum, and the practical habits that keep them moving.
The first workshop usually goes well. People fill a wall with sticky notes, name the bottlenecks everyone already suspected, and leave with a sense that things will finally change. Six weeks later the photos of that wall sit in someone’s phone and the process works exactly as it did before.
This is rarely a motivation problem. Process improvement projects stall for predictable, structural reasons, and most of them are settled before the workshop even starts.
What a stalled project looks like from the inside
Picture a building materials distributor in Amman with around 80 staff. Order processing is slow. A sales rep takes the order by phone or WhatsApp, types it into a spreadsheet and sends it to the warehouse, and the warehouse calls back when something is out of stock. Customers complain about late deliveries and wrong quantities.
Management runs a half-day workshop with sales, warehouse, finance and delivery. The group agrees on a new flow: orders go straight into the system, stock is checked at the point of order, and finance releases credit holds within two hours. Everyone nods. Then Sunday arrives, order volume is high, and the reps do what they have always done, because the new flow exists only on a flipchart.
Nobody decided who owned the change. Nobody adjusted the system to support it, and nobody measured whether anything was different. The workshop produced agreement, not a plan.
Five reasons improvement work loses momentum
- The workshop mapped the official process, not the real one. People describe how work is supposed to happen, especially when their manager is in the room. The workarounds, side spreadsheets and phone calls that actually keep orders moving never reach the wall, so the redesign fixes a process that does not exist.
- There is no single owner. Improvements that cross departments belong to everyone, which in practice means no one. If the sales manager thinks the warehouse is driving it and the warehouse thinks finance is, the project dies quietly.
- Actions are too big to start. “Implement order tracking” is not an action. It is a project hiding inside a bullet point, and nobody can begin it on a Sunday morning, so nobody begins it at all.
- The day job wins. Improvement work is added on top of a full workload with no time set aside. When a large customer calls, the improvement task is the first thing to slip, every week.
- Nobody measured the before. If no one recorded how long an order took before the change, no one can show that it is faster afterwards. Without visible progress, energy fades and the sceptics win the argument by default.
Start before the workshop, not in it
The fix begins before anyone books a meeting room. Spend a few days watching the work and collecting evidence. Follow a handful of orders from the first call to delivery and note every handoff, every wait and every time the same data is typed again. Ask the people doing the work what they do when the system does not cooperate.
This is Know, the first stage of the KEMLAT Method. With traced orders on the table, the group discusses real examples instead of trading opinions. “This order waited two days for a credit approval that took three minutes once someone looked at it” is far harder to argue with than “approvals are slow”.
Leave the room with actions someone can start next week
A good workshop ends with fewer, smaller commitments than a bad one. The test for each action is simple: can a named person begin it in the next working week without waiting for anyone’s permission?
| Weak workshop output | Strong workshop output |
|---|---|
| Improve order accuracy | The sales coordinator adds a mandatory stock check column to the order sheet by Thursday and trains the four reps on Sunday |
| Speed up credit approvals | Finance agrees a two-hour response target, and the finance manager reviews the waiting list at 10:00 and 14:00 daily for one month |
| Better communication between sales and warehouse | A ten-minute standing check at 08:30 between the sales coordinator and the warehouse supervisor, starting next week |
| Implement a new system | The IT (information technology) lead lists the three changes the current system needs and gets a cost estimate within two weeks |
The strong actions each carry a name, a date and something you can observe. They also accept that some changes will be manual at first. A manual step that works is a better foundation for later automation than an automated step nobody trusts.
Keeping momentum for the first ninety days
Most of the real work happens in the weeks after the workshop. These habits keep it moving:
- A short weekly review with the same people at the same time. Thirty minutes is enough to check each action as done, in progress or blocked. Blocked items get a decision in the room, not another meeting.
- One or two measures, tracked where people can see them. For the distributor, that might be time from order to dispatch and the number of orders corrected after dispatch. A simple chart on the wall or in the team chat is enough. You do not need a dashboard yet.
- Protected time for the owner. If the process owner has no hours set aside for the change, the change will not happen. Agree on a few hours a week and defend them when the busy season arrives.
- Update the written procedure as soon as a change sticks. Otherwise new staff learn the old way from colleagues and the process drifts back within months.
- Share small, real results. Telling the team that credit holds now clear the same day, with the numbers to show it, does more for adoption than any launch email.
Do not let the project wait for a tool
Many projects stall because they reach for software too early. Someone suggests a new ERP (enterprise resource planning) module or an automated workflow, and the improvement waits months while the tool is selected, budgeted and configured. Meanwhile the underlying problems carry on untouched.
The better sequence is to stabilise the process manually, measure it, and only then decide which parts deserve a system change or automation. Once the distributor’s reps have used the stock check consistently for a month, automating it is low risk, because everyone understands the step and the data is cleaner. Automating a process that people have not agreed on simply produces wrong orders faster.
Turning the workshop into the start of real change
If a workshop is already on your calendar, check three things before it happens: that someone has traced real transactions, that one person will own the outcome, and that the first two measures have a baseline. If any of the three is missing, move the date rather than run the session without them.
This is how our Operations Transformation work runs: trace how work really happens, redesign it with the people who do it, give each change a named owner, and track the measures until the new way is simply how things are done.