The rubric · describe one wait, get the fixes that move it

Which fix would
actually move it?

Once you have mapped a process and found where its time goes, the next question is what to do about it — and the honest answer depends on what the wait is made of. Describe one wait below. You will get the mechanisms that act on that kind of wait, ranked, with what doing each one looks like at your maturity level and what it returned across the worked examples.

Before you buy more people, technology, or services — do you know how effectively your existing resources are being converted into outcomes?

PFV shows where an investment would actually improve capacity, flow, throughput or exposure — and where it would not.

Describe the wait

Start from what your step is called
Type the name of the step you are stuck on. The library holds 353 steps from 55 sample processes — if one looks like yours, tap it and the questions below fill in as a starting point you can correct.
1  What is the step waiting on?
The single most useful thing to know. A queue behind another team is a different problem from a queue behind a decision, and almost nothing fixes both.
2  How long does it typically wait, in business days?
Median, not the worst case. If you don't know, that is worth noticing on its own — and worth measuring before acting.
3  Is this the biggest wait in the process?
In the worked examples this mattered more than anything else. One dominant queue was worth more than every other fix on that process combined.
4  Does the work wait for a cadence?
A weekly meeting, a Friday review, a monthly batch. If the wait is suspiciously close to half a week or a fortnight, it usually is.
Before question 5 — test it, do not guess
Is it actually a bottleneck?

A long queue is not proof of a constraint. A queue can be long because the resource is genuinely overloaded — or because work waits for a schedule, an approval, or a priority order that puts your work last. Those look identical from where you stand and they have opposite fixes. Four checks separate them, and three of them are free.

1  Is the queue growing?
Count what arrives at that step and what leaves it over the same period. This is the only check that can prove capacity: a queue that grows without bound has no other explanation.
Per what period? Any unit works — the test is the ratio, so it only matters that both counts use the same one. Long enough to be representative, short enough to act on: a week suits most operational work.
2  Does the next step ever go idle waiting on this one?
A real constraint starves everything downstream of it. That is its signature, and it is why fixing one raises output instead of moving work around.
3  Are they busy on this work, or on other work?
The distinction that catches most people out. A team fully occupied with work that outranks yours looks exactly like a capacity problem and needs a completely different fix.
4  Have you expedited a few items past it?
Push five items past the step and watch total lead time. Reversible, nearly free, and the only check here that tests your process rather than a plausible story about it.

Whatever this says, it feeds question 5 below: permission or capacity. Getting that one wrong is how organisations hire into a queue that was never short of people.

5  Is the blocker permission, or capacity?
Are people waiting to be allowed, or waiting for someone who is genuinely busy? The fixes are opposite — and if you are not sure, run the four checks above rather than guessing.
6  Does work come back after this step, or downstream of it?
Rejected, reopened, sent back for missing detail, redone. This is how you find out whether the wait is causing a defect rather than just costing days.
7  Where is this process today?
CMMI® levels used as vocabulary, not as an appraisal: 2 it works through effort and memory · 3 written down and done the same way twice · 4 measured, so it can be predicted · 5 it improves itself.

What would move it

How the ranking is produced, and what it is not. Each mechanism is scored on whether it acts on your kind of wait at all, whether the wait is big enough for it to matter, whether this is the dominant queue, and the qualitative answers you gave about cadence, blocker and rework. The days figure, where one is shown, is the median return measured across the 42 operational sample processes for a wait of that shape at that transition — it is a modelled reference point from other people's processes, not a prediction about yours. Nothing here has seen your data. Treat the ranking as a shortlist to argue with, not an answer.

If nothing scored well, that is informative. It usually means the delay is not a queue — nothing was looking, nobody had started, the work had not arrived. Flow mechanisms will not touch that, and knowing so early is worth more than a plausible fix aimed at the wrong thing.

Use at your own risk

Process Flow Visibility is a method and a set of estimating tools, not professional advice. Every figure it produces is derived from numbers you enter, most of which are estimates. It does not measure your process, audit your data, or know your business.

Nothing here is legal, financial, security, safety, medical or engineering advice. Do not use these outputs as the sole basis for any decision that carries operational, financial, regulatory or safety consequences. Validate anything that matters against your own records and your own professional judgement.

Provided “as is”, without warranty of any kind, express or implied, including merchantability, fitness for a particular purpose and non-infringement. To the fullest extent permitted by law, Don Woodward accepts no liability for any loss or damage — direct, indirect, incidental, consequential or otherwise — arising from the use of this site, these tools, or any figure they produce. You use them entirely at your own risk, and you agree to hold the author harmless for any outcome that follows.

The worked examples are modelled from industry data, not audited case studies. Named methods and standards are the property of their respective owners and are referenced for description only; no endorsement or affiliation is implied.