In short
Operational friction is the unpaid tax a business pays every time work has to stop, wait, be re-done or be explained again. You can put a number on it by measuring three things: touch time versus elapsed time on your core workflows, rework rate, and the hours senior people spend on work only they can do. Multiply the recovered hours by loaded salary cost and you have a defensible figure — usually somewhere between 8% and 20% of payroll for a growing business that has never formalised its operations.
Key takeaways
- Friction is invisible in accounting because it is absorbed into salaries you are already paying.
- The cheapest measurement is the ratio of touch time to elapsed time: if a two-hour job takes six days, the gap is the friction.
- Rework is the most expensive category and the least tracked. Count it for two weeks and the number will surprise you.
- Senior-person bottlenecks cost more than junior inefficiency, because the opportunity cost is strategy work that never happens.
- You do not need perfect data. A defensible estimate that triggers action beats a precise one that arrives next quarter.
What operational friction actually is
Most businesses know something is wrong long before they can name it. Projects take longer than they should. The same questions get asked every month. A simple approval crosses four inboxes. Nobody is lazy, nobody is incompetent, and yet the work moves like it is wading through something.
- Operational friction
- The gap between the work your business does and the work your business has to do to get that work done. It shows up as waiting, re-explaining, re-doing, chasing, checking, and reconstructing information that already existed somewhere.
The reason friction survives for years is that it never appears as a cost. Nobody submits an invoice for “three hours reconstructing a client brief because the original was in a Slack thread.” That cost is absorbed into a salary you were already paying, so it passes through every budget review untouched while genuinely smaller costs get scrutinised to death.
This is the central accounting problem of operations work. A $400 software subscription gets an approval meeting. A workflow that burns forty person-hours a month gets nothing, because those hours are already bought and paid for.
Friction is not the same as being busy
A busy business can be a healthy one. The signal to look for is not volume, it is the ratio of productive work to the work required to enable it. When that ratio degrades, everyone feels it before anyone can prove it, and the usual response is to hire, which imports the friction into more salaries rather than removing it.
Why the cost stays invisible
Three structural reasons, and they compound.
First, it is distributed. Friction rarely costs one person a day. It costs eleven people twenty minutes, which nobody notices and nobody reports. The aggregate is enormous; the individual experience is mild irritation.
Second, it is normalised. People who joined after the workaround was invented do not experience it as a workaround. They experience it as the job. Ask a two-year employee why the handoff between sales and delivery involves a manually maintained spreadsheet and you will often get a shrug: that is how it works here.
Third, the people best placed to see it are the busiest. The person who could describe the whole broken workflow end to end is usually the person holding it together, and they have no time to write it down. This is not a coincidence. It is the same underlying problem.
The people who understand the friction best are the ones with the least capacity to fix it, because understanding it is what is consuming their capacity.
The practical consequence is that friction gets discussed in the language of feeling rather than the language of finance. “Things are chaotic.” “We need to get organised.” Those sentences do not survive contact with a budget conversation. A number does.
Three measurements that turn friction into a number
You do not need a time-tracking rollout or a consultant with a clipboard. You need three measurements, all of which can be taken in a fortnight by someone who already works there.
1. Touch time versus elapsed time
Pick one workflow that matters — client onboarding, quote to contract, hire to first day. Ask two questions about a recent instance:
- Touch time: how many hours of actual human work did it contain?
- Elapsed time: how many working days passed between the first action and the last?
A healthy ratio depends on the process, but the gap is always instructive. When a job containing six hours of work takes eleven days to complete, the other eighty-two hours were spent waiting, chasing, or sitting in a queue nobody owns. That is your friction, expressed in the most legible unit there is.
A rough alarm threshold. When elapsed time exceeds touch time by more than roughly six to one on a routine internal process, the delay is structural rather than incidental — it is caused by the design of the workflow, not by the workload.
2. Rework rate
Rework is the most expensive friction there is, because you pay for the work twice and often pay a third time in relationship damage. It is also almost never measured. For two weeks, ask everyone doing client-facing or production work to log one thing: did this need to be redone, and why? Three categories cover most of it:
| Category | What it looks like | Usual root cause |
|---|---|---|
| Wrong input | Work started from stale, partial or contradictory information | No single source of truth; handoff carries no structured brief |
| Missed requirement | Deliverable was complete but did not meet an unstated expectation | Requirements live in someone's head or an old email |
| Late change | Approval or feedback arrived after the work was finished | Approval step sits at the end rather than at the decision point |
3. Senior-person bottleneck hours
Count the hours your two or three most senior operational people spend each week on work that only they can do — not because it is strategically important, but because the knowledge or the access lives only with them. Password resets. Answering the same client question. Approving a routine purchase. This is the most expensive hour in the building being spent on the cheapest possible task.
Doing the maths without pretending to be precise
Here is the calculation, deliberately crude, because a crude number that gets acted on is worth more than a precise one that arrives after the decision.
Step one: recoverable hours. For each of the three measurements, estimate the hours per month you would recover if the underlying cause were fixed. Be conservative. If you think a documented handoff would eliminate 70% of the rework, model 40%.
Step two: loaded cost. Multiply by the fully loaded hourly cost of the people involved, not their salary divided by 2,080. Loaded cost includes employer taxes, benefits, software seats and the overhead they consume. For most small businesses this is salary multiplied by roughly 1.25 to 1.4.
Step three: the capacity question. This is the step most people skip, and it is the one that changes the conversation. Ask: if we got these hours back, what would we do with them? There are only two honest answers. Either you would take on more work with the same headcount, in which case the value is the gross margin on that additional work. Or you would not, in which case the value is the delayed cost of the next hire.
A worked example, using round numbers for a twelve-person services business:
- Rework: 34 hours/month across the delivery team, 40% considered recoverable → 13.6 hours
- Waiting and chasing on onboarding: 22 hours/month, 60% recoverable → 13.2 hours
- Founder answering questions only they can answer: 26 hours/month, 50% recoverable → 13 hours
- Total: roughly 40 hours a month, or a quarter of a full-time person
At a blended loaded cost of $55 an hour, that is about $26,000 a year in direct cost. But thirteen of those hours belong to a founder, and the honest answer to the capacity question is usually that those hours would go to sales or to fixing the next thing. Priced that way, the figure is considerably larger — and it is the version that gets a decision made.
Deciding which friction to remove first
Once you can see the friction, the temptation is to fix all of it. That is how operations improvement projects die. Everything gets started, nothing gets finished, and the organisation concludes that process work does not deliver.
Sort candidates on two axes only:
| Priority | Profile | Action |
|---|---|---|
| Fix now | High cost, low effort. Usually a missing document, an undefined owner, or an approval sitting in the wrong place. | Do it this month. These build the credibility you need for the harder work. |
| Fix next | High cost, high effort. Systems integration, a real documentation programme, restructuring a team's responsibilities. | Scope properly, sequence after a couple of quick wins. |
| Plan | Low cost now, but scales badly. Fine at twelve people, fatal at thirty. | Schedule against a growth trigger, not a date. |
| Leave alone | Low cost, high effort. The workaround is cheaper than the fix. | Write down that you decided this, so nobody re-litigates it in six months. |
That last row matters more than it looks. A deliberate decision to tolerate a piece of friction is a legitimate operational choice. An undocumented one gets rediscovered and re-argued every year.
Start with the constraint, not the irritation
The loudest complaint is rarely the biggest constraint. If your delivery team is slow because they are waiting on scoping decisions, buying them a better project tool will not help. Find the step where work reliably piles up, fix that, and re-measure — because the constraint moves once you relieve it.
Making the number stick
A one-off measurement produces a one-off improvement. What you want is a figure that gets revisited, because friction regenerates — every new hire, tool and client type introduces new gaps.
Three habits keep it honest:
- Re-measure the same workflow quarterly. Same definition, same method. A moving number that everyone trusts beats a perfect number nobody repeats.
- Attach friction to the growth plan. Before hiring into a workflow, ask whether you are hiring capacity or importing friction. Adding a person to a broken handoff usually creates two broken handoffs.
- Give the number an owner. Not a committee. One person whose job includes reporting it, with the standing to say when something has got worse.
The businesses that get this right do not become obsessed with process. They become slightly boring in a specific way: work moves at a predictable pace, the same question gets answered once, and growth stops feeling like it costs more than it earns. That is the whole objective. Not perfection — flow.
If you want the measurement done for you, the Mayim Ops assessment scores ten operational dimensions and returns the friction points ranked by impact, with the evidence attached to specific answers rather than general advice.
Frequently asked questions
What is operational friction in business?
Operational friction is the effort a business spends enabling work rather than doing it: waiting for approvals, re-explaining context, redoing work that started from bad inputs, and chasing information that already exists. It is a real cost, but it is absorbed into salaries you already pay, so it does not appear as a line item in your accounts.
How do you measure operational inefficiency?
The three most practical measures are touch time versus elapsed time on a core workflow, rework rate over a two-week logging period, and the number of hours senior staff spend on tasks only they can perform. Convert recovered hours into money using fully loaded hourly cost, then decide whether those hours become additional capacity or a delayed hire.
How much does operational friction typically cost a small business?
In the assessments we run, growing businesses that have never formalised their operations typically carry friction worth somewhere between 8% and 20% of payroll. The range is wide because it depends heavily on how much of the business runs on undocumented knowledge and how many handoffs sit between teams.
Should I fix operations before hiring?
Usually you should at least measure first. Hiring into a broken workflow tends to multiply the handoffs rather than the output, and the new person spends their first months absorbing undocumented context from the very people who are already overloaded. If a workflow has high rework or a long elapsed-to-touch ratio, fixing it often recovers more capacity than the hire would have added.
What is the difference between an operations audit and a business process review?
A process review examines how a specific workflow is executed and looks for steps to remove or reorder. An operations audit is broader: it examines documentation, ownership, systems, automation, communication and measurement together, because most process problems are actually symptoms of a missing owner or a missing document rather than a badly sequenced step.