What a Blocker Hint Saves You on Every Late Job

Finding late jobs is trivial: it is a date comparison the system does instantly. Working out why each one is late is where a review turns into an afternoon, because it means answering five separate questions per row from scratch. A hint that routes you to the right report collapses the first question, which is the one you were paying for. EDGEBIC by User Solutions prints one beside every late job.

This post covers one return: planner and supervisor hours currently spent establishing what happened rather than deciding what to do. It sits under the EDGEBIC results guide and follows late versus overdue, which splits the same list by failure type.

Detection Is Cheap, Diagnosis Is Not

The late jobs report is a snapshot of every job whose scheduled end exceeds its due date, ranked worst-first, carrying the due date, the scheduled end, days late, percent complete, and a status of not started, in progress, or behind.

Producing that list costs nothing. The comparison is arithmetic and the ranking is a sort.

Now take one row and ask why. To answer honestly you need to know:

1. Has it started at all?
2. If it started, how far has it got?
3. Which operation is holding it?
4. Did the plan move underneath it, and if so who moved it and why?
5. Was the station it needs overloaded?

Five questions. The usual method for answering them is a walk to the floor and a couple of phone calls, which is slow, depends on who is available, and produces an answer nobody can check later. Multiply by twenty rows and you have your Monday morning.

Notice that none of those five questions is the interesting one. The interesting question is what to do, and it comes after all five have been answered.

The Mechanism: A Hint Is a Routing Decision

Each late job row carries a short blocker hint saying where to dig first. Three documented examples cover most rows: no actuals meaning the job never started, significantly behind plan, and schedule slipped.

The critical thing to understand is what a hint is and is not. It is not a cause. It does not tell you why the job is late. What it does is answer question one from the list above, which means you stop having to check the other four to find out which situation you are in.

Three hints, three different reports, three different conversations. Without the hint you open all three for every row to find out which one applies. With it you open one.

The Ambiguity Inside the First Hint

The no-actuals hint deserves separate treatment because it is the one that misleads, and it misleads in a predictable direction.

The hint reports that no actual hours exist. There are two ways for that to be true. The work has not started. Or the work has started and nobody captured it.

Those require opposite responses. A genuinely unstarted job needs a queue or material or priority intervention. An unrecorded running job needs nothing done to it at all, and someone chasing it will find an operator halfway through the operation wondering what the fuss is.

So the first check on any such row is capture, not production, and in plants where punching is inconsistent a meaningful share of these rows turn out to be data gaps. That ratio is worth measuring in its own right, because a late list inflated by capture gaps burns attention on jobs that are fine while masking the ones that are not. Kiosk actuals closing the planning loop is what shrinks that ambiguity.

There is a related trap across reports that catches people out. On the earned value report, a job that has not started shows a schedule performance index of exactly 1.00, because nothing was planned to be done yet and there is nothing to compare against. That reads as on schedule and it is not: it means nothing to compare yet. A job can therefore look untroubled on one report and sit near the top of the late list on another, and both are behaving correctly.

The Causal Chain to Saved Hours

1. The list is produced automatically, ranked worst-first, so the sequence of attention is decided before anyone sits down.
2. Each row's first question is pre-answered by its hint.
3. One targeted drill-down replaces a search. The step report, the reschedule trail, or a capture check, rather than all three.
4. The cause is found in the right report. This stage is genuinely irreducible: a slow step might be a routing that was always wrong, a machine problem, an operator pulled elsewhere, or late material, and only evidence separates those.
5. The decision is made with numbers rather than recollection. Move a job, add capacity for a day, accept the date, or renegotiate.
6. The review finishes before the standup instead of after lunch.

The documented worked example in the product guide follows exactly this path: the late list flags the job, progress confirms it, the step report finds the operation, the reschedule trail shows the push, the swap audit rules out a machine change, and the capacity report exposes two rush jobs inserted ahead of it. Ten minutes, four reports, no guessing. The alternative is the same conclusion reached at lunchtime by phone, if at all.

Measuring the Saving in Your Own Plant

This is one of the easier returns to quantify because the unit is time and you can measure it directly.

1. Time your current diagnosis, honestly, for five late jobs. Start the clock when somebody asks why, stop it when the cause is established. Include the waiting: the callback, the walk, the person who was in a meeting.
2. Count your typical late population per review. Not per year, per review, because that is the unit of the meeting.
3. Multiply. That product is the diagnosis cost of one review cycle, in hours of the people who can least afford it.
4. Annualize by review frequency. Weekly reviews make this a large number quickly.
5. Time the hint-routed version on the same five jobs. Read the hint, open the one report it points to, find the cause. Compare against step one.
6. Separately, measure the capture ratio. Of your no-actuals rows, how many were genuinely unstarted and how many were running but unpunched? A high unpunched share means your late list is partly fiction and your first project is capture, not expediting.
7. Value the displaced work. The hours saved are not the whole return. They are hours a planner spends on the plan instead of on reconstruction, which is where the compounding sits.

For documented outcomes rather than typical ones, the User Solutions and RMDB lineage includes Homestead Furniture cutting schedule build from 40 hours per week to 2, and GE Railcar moving from 30 percent to 90 percent on-time delivery. Those belong to their engagements and are quoted as heritage, not as forecasts for your review meeting.

Where the Hint Does Not Help

Five limits, and the first two are the ones that catch people.

A job with no due date never appears at all. The list is defined by a scheduled end exceeding a due date, so an order with an empty due date cannot be late by construction. If your late population feels small relative to how the shop is behaving, audit for missing due dates before congratulating anybody.

One row gets one hint, and a job can have several problems. A job can be behind plan and have had its schedule slip and be short of material. The hint names the most prominent signal, not the complete picture, and the drill-down is where you find the rest.

Percent complete depends on logged hours. It is hours-based, falling back to pieces only when no hours exist, so a job where nothing has been captured reads as zero progress whether or not work happened. Read percent complete and the no-actuals hint together rather than separately, because they share a dependency.

The hint cannot tell you the cause, only the category. Anybody who treats a hint as a root cause will occasionally be right and will build a habit that eventually costs them. It narrows the search; it does not conclude it.

Faster diagnosis does not reduce lateness by itself. It reduces the cost of dealing with lateness and it makes the response earlier, which matters, but the jobs are still late. Reducing the count is a capacity, quoting, and sequencing question that reducing expediting and the true cost of a single late job both address.

The takeaway

The expensive part of a late-job review is not finding the late jobs, it is establishing what happened to each one, and that cost is paid per row by the people you would rather have working on the plan. A hint that answers the first of five questions turns a search into a lookup, and the ranked list decides the order of attention before anyone sits down. Time your own diagnosis on five jobs before you assume this is small, and check what share of your no-actuals rows are capture gaps rather than stalled work. To see the hints against your own late list, book a demo of EDGEBIC, and if you are running the older platform, the move from RMDB to EDGEBIC brings the reporting with it. Read this next to how EDGEBIC shortens the planning cycle and how customer service answers where is my order.