WIP You Can't See

The floor is full. Machines are running, the assembly bay is busy, material is moving, and everyone is clearly working hard. By every visible sign, things are going well.

Then the MD asks about one specific job. How is order 4471 tracking against plan and against the price we quoted? And the room goes a little quiet, because the honest answer is that nobody knows precisely, and they will not know precisely until it ships. The job feels on track. Feeling on track and being on track are not the same thing.

In a custom-order shop, every order is a one-off. So, too often, is every nasty surprise at the end of it.

Busy is not the same as on plan

Repeatable manufacturing has a quiet advantage that custom manufacturing does not. When you make the same product every day, you know what good looks like, minute by minute. There is a standard time, a standard cost, a standard yield, and any drift shows up quickly against a number everyone already knows.

Engineer-to-order and make-to-order work has none of that comfort. A single job might sit on the floor for weeks or months, moving through design, machining, fabrication, assembly, and testing. It is semi-custom by nature, so there is no familiar standard sitting there to measure it against in real time. Its true status lives in a few people's heads, on a whiteboard, and in a planner's spreadsheet that is only as current as the last time someone remembered to update it.

So the floor looks busy, and busy feels reassuring. But busy tells you that people are working. It does not tell you whether this particular job is ahead, behind, or quietly eating the margin you quoted.

The surprise always arrives late

This is a problem because of when the truth finally shows up. On a long custom job, the real status tends to become clear near the end, at final costing, when the hours are added up and the material is fully booked and the margin turns out to be thinner than the quote promised.

By then, it is too late to do anything about it. A job that was slipping at week three could have been re-sequenced, re-planned, or discussed with the customer while there was still room to move. The same job discovered at week twelve is simply a loss you now absorb. Late deliveries land the same way. The delivery date does not slip loudly. It slips quietly, and you and the customer find out together, which is the worst possible way for either of you to find out.

There is a further cost worth naming, though it is a consequence rather than the main event. Work-in-progress is where a lot of your cash physically sits, out on the floor, in half-finished jobs. When jobs run slow or stall unseen, that cash sits there longer. The visibility problem is the root. The cash tied up is what it quietly costs you downstream.

Good tools that were built for something else

The usual tools each hold a piece of the picture, and none holds the live whole.

The ERP records transactions. Material issued, a purchase order raised, an invoice booked. That is real and useful, but it is not the physical progress of a custom job as it happens. Project management software captures the plan, but it depends on people updating it by hand, so it lags reality by however long it has been since the last update. The planner's spreadsheet is a genuine feat, and it is exactly as fresh as the last person to touch it. The review meeting surfaces status out loud, which means it is both late and coloured by whoever is speaking.

None of these is a bad tool. They were each built to answer a different question. What none of them was built to do is give leadership a live, shared, honest picture of where every custom job actually stands right now, while there is still time to act on it.

See the job while you can still steer it

The shift is to stop reconstructing project status after the fact and start seeing it as the work happens.

Capture progress at the source, as each job moves through each stage on the floor, rather than re-keying it into a report later. Curate that into one consistent picture of work-in-progress and project status across every live order. Then the job that is drifting becomes visible in week three, not week twelve, while re-sequencing, re-planning, or an early conversation with the customer are all still on the table.

It is worth being clear about what this is. It is manufacturing intelligence and continuous monitoring, which is to say decision support. It surfaces the job that is slipping so a human can act early. It does not schedule your floor for you, and it does not run the project. It gives the people running the project the one thing the current setup takes away from them, which is time to react.

Questions for your next project review

• If I asked for the true status of every live custom order right now, could anyone give it, or would we reconstruct it?
• On our last job that lost margin, at what point did we actually know it was slipping, and could we have known sooner?
• How much of our project status lives in one planner's spreadsheet and one planner's memory?
• When a delivery date slips, do we see it coming, or do we find out alongside the customer?
• How long does cash typically sit in half-finished jobs on our floor, and do we know which jobs are holding it?

The order you cannot see yet

In a custom shop, the dangerous order is not the one you can watch going wrong. You can manage that one. The dangerous order is the one that is already drifting and has not surfaced yet, the one whose bad news is still weeks away.

You cannot fix a margin after the truck has left. The shops that hold their margins are not the ones that quote the sharpest. They are the ones that can see a job drifting while there is still time to steer it.

If your work is custom and long-cycle, it may be worth looking at how manufacturers are getting a live view of work-in-progress across every project. See how it works.

FAQ