How Much of Your Business Is Actually Documented?

You are in the majority, and it is not close. When we gap-analyzed 16 small and mid-size businesses across 68 roles and 461 process areas, the average process area came back 27% documented, half of all role areas had nothing written at all, and 82% of teams scored below 50%. So if you are asking how common it is to not have written procedures at a company with 40 people, the honest answer is that having them is the exception. The more useful question is not how much of your business is documented, but which parts.

How Common Is It to Not Have Written Procedures?

Very common, and the number is worse than owners guess. Owners tend to overestimate, because they are counting the documents they know exist: the employee handbook, the safety binder, the onboarding checklist somebody built two years ago. Measured against the actual process areas a role touches, the average lands at 27%.

Half of the role areas we scored had zero. Not thin, not stale. Nothing.

That is the part that catches people. A company can have a real training platform, a real handbook, and a real onboarding flow, and still have entire roles where not one working procedure exists in writing. The knowledge is there, it is just stored in people rather than in the business, which is a fine arrangement right up until somebody gives notice.

Coverage Is Lopsided, Not Just Low

The gaps are not spread evenly. They cluster, and they cluster in the same place at nearly every company we walk into.

Take a multi-site physical therapy group we worked with. Onboarding, payroll and insurance verification were documented to the step, clean and current. Clinical treatment protocols, technician daily workflow, front-desk prioritization, open and close procedures, and equipment maintenance were all undocumented. Every single one.

Look at those two lists again. The documented side is everything with a compliance deadline, an auditor, or a government form attached. The undocumented side is everything a patient actually touches.

A company that documents payroll and skips the treatment room has not documented its business, it has documented its paperwork.

Documentation follows external pressure, not internal risk. Payroll gets written down because somebody outside the company will ask for it. Front-desk prioritization never does, because the only person who will ever ask is the new hire on day three, and by then the answer arrives verbally from whoever is standing closest. That asymmetry is why total coverage percentage is a weak number on its own, and why sequencing matters more than volume when you decide which processes actually deserve the effort.

How Do You Measure Documentation Coverage?

You measure it per role, against a list of questions a complete version of that role would answer. Company-wide percentages are almost useless, because a company at 40% is usually two teams at 80% and three teams at zero, and the average hides exactly the thing you needed to see.

Here is the mechanic we use. Every team carries an interview guide, and the guide keeps a live count of answered versus unanswered questions. On one account we ran a gap analysis pass across 35 historical session transcripts and auto-checked 91 of 175 guide questions across four teams, then published the coverage per team rather than one number for the company. Four separate scores, four separate decisions about where to go next.

Coverage alone still is not enough to steer by, so we watch three things together.

The third one is the tiebreaker. Ask any role what stops if they are gone for two weeks, and the honest answer is a list. If that list is shorter than it was last quarter, the documentation is working. If the percentage climbed and the list did not shrink, you documented the easy things.

A structured audit of what you already have will usually tell you which of the two happened.

The Processes That Always Get Documented, and the Ones That Never Do

There is a predictable order to this, and you can check yourself against it in about five minutes.

Almost always written down: onboarding paperwork, payroll, benefits enrollment, insurance verification, safety and OSHA material, anything a franchisor or a regulator requires. Almost never written down: how a job actually gets done, how the front desk decides who gets seen first, what a technician does between the first task and lunch, how somebody decides a quote is worth chasing, the judgment calls that separate your best person from your third-best person.

The second list is the business. The first list is the cost of being allowed to operate one.

This is also why the people holding the most valuable knowledge are almost always invisible on a documentation report. They score well, because the paperwork attached to their role is filed and current. Run a proper knowledge risk assessment instead and the same person shows up as the single point of failure they have quietly been for six years.

---

What Does Good Coverage Look Like at Your Size?

There is no percentage that makes a 40-person company safe, and anyone who hands you one is guessing. Our measurement is per role area, not per headcount, and a company with 40 people might carry 6 roles or 16 depending on how the work is split. What travels across sizes is the shape, not the score.

Good coverage means no role sits at zero, the documented set includes the work customers touch and not only the work auditors touch, and the two-week list is getting shorter. A company at 35% with those three things true is in better condition than a company at 60% that documented five versions of onboarding.

And coverage is not a virtue on its own. We asked one auto repair operations manager which of his processes were broken and he said none, flatly: great processes, great employees, and 90 to 95 percent of the time everything gets done the way it is supposed to be. Almost none of it was written down anywhere.

He was not wrong about his shop. He was measuring compliance today, which was excellent, while the owner was worried about deterioration, a revision history, and what happens the day two long-tenured guys leave in the same quarter. That is a different problem with a different reason to solve it, which is worth understanding before you decide that nothing is broken so nothing needs writing.

Coverage Falls When You Start Asking Better Questions

Expect your number to drop, and treat it as good news when it does.

Every real interview turns up a question the guide never thought to ask. When it does, we write the new question into the live guide, and the coverage percentage falls, because a question nobody has answered is exactly what a gap is. The number went down because the map got bigger.

A coverage score that only ever climbs is measuring nothing.

That is the most useful reframe for an owner watching a dashboard. If your metric never regresses, nobody is discovering anything, which means the guide is measuring itself rather than the business. The first six weeks of honest work usually look like backsliding on paper. It is not.

What Should You Document Next?

Do not answer that with a percentage. Answer it with a role.

1. Pick the one role. Whichever role, if it vanished for two weeks, would stop the most revenue. Not the busiest role, the one that stops things.
2. List its process areas. Everything that role owns, written as areas rather than tasks, ten to twenty lines. Mark each one documented, partial, or nothing.
3. Strike the compliance items. Payroll, onboarding paperwork and safety are probably already done. Ignore them for now, they are not what breaks.
4. Start at the customer. From what remains, take the area a customer or a paying job touches first, and document that one.
5. Rerun the two-week question. After the first three, ask the role again what stops without them. A shorter list is the proof, not the percentage.

That order is deliberate, and it is close to how we sequence engagements at The Systems Effect when a client wants to know what to document first with a limited number of hours. It also survives contact with a real week, which "document everything" never does.

Pick one role this week and write its process areas on a single page, then mark the ones with nothing behind them. You will have a coverage number in about twenty minutes, and it will be more useful than any benchmark.

Frequently Asked Questions

How common is it for a company with 40 employees to have no written procedures?

It is the normal condition, not the exception. Across 16 businesses, 68 roles and 461 process areas we measured, the average process area was 27% documented, half of all role areas had nothing written at all, and 82% of teams scored below 50%. Companies that size almost always have some documentation, usually onboarding, payroll and safety, and almost always have entire roles where no working procedure exists in writing.

How much of a business should be documented?

There is no safe percentage, and any number offered by headcount is a guess. The useful test is shape: no role sitting at zero, the work customers touch documented and not just the work auditors require, and a shrinking list of what stops when a given person is gone for two weeks. A company at 35% with those three things true is in better shape than one at 60% without them.

How do you measure documentation coverage?

Score it per role against a guide of questions a complete version of that role would answer, and publish the number per team rather than per company. On one account we checked 91 of 175 guide questions across four teams by running a gap analysis over 35 past session transcripts, which gave four separate coverage scores and four separate decisions. Company-wide averages hide the teams at zero, which are the only teams that matter.

Is 100 percent documentation the goal?

No, and chasing it wastes months. Coverage is a diagnostic that tells you where the holes are, not a score to maximize, and a number that only ever climbs means nobody is asking new questions. Expect coverage to fall when a conversation surfaces a question your guide never thought to ask, because that is the map getting bigger rather than the work getting worse.