How Much of Your Business Have You Actually Mapped?

Process mapping coverage is the proportion of your business that your map describes on evidence rather than on assumption, and it is the number almost nobody tracks, which is why so many mapping exercises end with a confident diagram of a company that has only been half looked at.

A map with no coverage figure attached is not wrong. It is just unmarked, and unmarked means every part of it gets read with the same confidence, including the parts you invented on a Thursday afternoon because they seemed obvious.

Where maps come from, and why it matters

Every box on a process map arrived by one of four routes, and they are not equally trustworthy.

Somebody watched the work happen. This is the strongest, and the rarest.

Somebody who does the work described it. Strong, with the known caveats about memory and about people describing the designed version rather than the real one.

Somebody who manages the work described it. Considerably weaker. Management accounts are accurate about intent and structure, and systematically miss waiting, checking, rework and workarounds, because those are invisible from one level up.

Or somebody inferred it. The map needed a step between two others, so a step was drawn. This is guesswork, and on most maps it is unlabelled and looks identical to the rest.

If you record nothing else, record which of those four produced each part. It takes one column and it changes how the map gets used.

The departments that get skipped

Coverage is never evenly distributed, and it fails in a predictable pattern.

Sales and delivery get mapped, because that is where the revenue is and where the pain is loudest. Finance gets mapped partially, usually the invoicing half and not the reconciliation half. Support gets mapped if there is a ticketing system to read from.

What gets missed is everything that happens between departments, everything that only one person does, and everything that happens rarely but expensively. Offboarding a client. The annual renewal cycle. What happens when a supplier fails. These have low frequency and high cost, which is precisely the combination that makes them worth mapping and also the combination that makes everyone forget.

The other consistent gap is the founder. In companies under about thirty people a surprising amount of process runs through one person who is not thinking of themselves as a process, and they map everybody else.

Four questions that measure coverage honestly

You do not need a methodology for this. You need to be willing to answer four questions without flattering yourself.

Which departments have you spoken to somebody in? Not sent a form to. Spoken to.

For each mapped process, how many separate people described it? One account is a claim. Two agreeing accounts is evidence. Two disagreeing accounts is a finding, which is better than either.

Which processes on the map has nobody confirmed? Every map has some. Find them and mark them.

And what is on the map only because it was on the org chart? Reporting structure is not process, and steps derived from it are inference wearing a suit.

Marking the gaps is the useful part

The instinct when you find a gap is to fill it. Resist that for a moment, because an acknowledged gap is already doing work.

A map that says "we do not know what happens between the signed contract and the first invoice" is more useful than a map that guesses, because it tells the reader exactly where not to rely on it, and it tells you exactly what the next hour of effort should be spent on. Confidence markings turn a static picture into a work queue.

This is why coverage by department is reported alongside the map in abi. Clone, so you can see which parts are evidence and which are still guesswork rather than reading the whole thing at one level of confidence. Building the map by hand is free on any account, with no card. The interviews that raise coverage are the paid part, because they are the part that costs anything to run.

Complete is the wrong target

Nobody needs a fully mapped business, and chasing one is how these projects die.

The map of a fifty-person company that covers everything is a six-week project that produces a document too large to read and out of date by the time it is finished. That is not a hypothetical failure mode, it is the standard one, and it is why so many people have a bad association with process mapping in the first place.

Aim instead for high coverage on the few processes that carry the revenue and the risk, and explicitly zero coverage, clearly marked, on everything else. Eighty per cent confidence on your quote-to-cash process is worth more than thirty per cent confidence on all fourteen of your processes.

The honest limit of the number

Coverage measures how much you have asked about. It does not measure whether the answers were true.

You can achieve high coverage quickly by interviewing managers, and produce a map that is confidently wrong in exactly the ways management accounts are always wrong. The number will look excellent. Two people described every process, everything is marked as evidence, and the whole thing still describes the official version.

So coverage is a necessary measure and not a sufficient one. It has to be read alongside who the accounts came from, and an account from the person doing the work outweighs three from people watching it.

Start by marking what you already have

If you have a map already, do not extend it this week. Go through it and mark every element with where it came from.

The exercise takes an hour and it is consistently uncomfortable, because most people discover that a third of their map is inference they had stopped thinking of as inference. That third is where to spend the next conversation, and knowing which third it is was the point.