Operational friction: what it is and how to measure it

What operational friction is

Operational friction is work that consumes capacity without producing the outcome it was supposed to support. It is not laziness and it is rarely one broken thing. It is the accumulated drag of a process that has grown around constraints nobody has revisited: work repeated because something upstream was wrong, time spent waiting on an approval, effort lost in handovers, information hunted for, data re-keyed into a second system.

Every one of those activities is real work. It is performed diligently, it appears in someone’s day, and it is almost always invisible in cost reporting because it is spread thinly across everyone who does it.

Why it stays invisible

• It is distributed. Ten minutes a day across forty people is a full-time role nobody noticed creating.
• It looks like the job. Re-keying an order between systems is described as processing an order, not as rework.
• Nobody logs it. It is not in a system of record, so it does not appear in reports built from systems of record.
• It is politically quiet. Naming friction can imply a previous decision was wrong, so it tends to be tolerated rather than raised.

How to measure it

Friction becomes manageable once it is expressed in units that finance and operations both accept. The method is straightforward, and the discipline is in not skipping steps.

1. Define the unit. Decide whether you are counting time, frequency, cost or delay, and be consistent. Mixing units is how a friction analysis loses its audience.
2. Capture at the point of work. Record the activity as it happens, with enough context to say what caused it. Retrospective estimation produces numbers that cannot survive a challenge.
3. Attribute to a cause. Each occurrence needs a stated reason: missing information, a system that does not talk to another, an approval with no service level, a rework loop. “Other” should be small, and its size is itself a finding.
4. Quantify in money, with the basis shown. Time multiplied by a stated rate is an estimate, and that is fine, as long as the rate and the hours are both visible.
5. Validate with the people doing the work. If the operators do not recognise the finding, the number is wrong or the framing is. Either way, fix it before it reaches a steering committee.

What to do with the result

A friction analysis is worth nothing unless it changes something. The output should be a short list of causes ranked by effort consumed, each with an owner and a proposed change, and enough evidence attached that the ranking can be defended. Everything else is commentary.

DELT4’s Zunox readouts are built for that purpose: they show where capacity is consumed, what the friction costs, and what sits underneath each figure.

Where friction usually hides

In practice, the largest sources tend to appear in a small number of recurring places: duplicate data entry between systems, approvals waiting in queues, work re-done because a requirement arrived incomplete, searching for the current version of a document, and status updates maintained by hand to compensate for systems that do not reconcile. If you are starting an analysis, start there.

Common questions

What is operational friction in one sentence?

Work that consumes real capacity without producing the outcome it was meant to support - rework, waiting, handovers, searching and re-keying, spread thinly enough that no report shows it.

Can operational friction be measured without new software?

Yes, at small scale: pick one process, define the unit, and have the people doing the work record occurrences as they happen. The difficulty is sustaining it, and comparing it across processes, which is where a capture tool earns its place.

Is friction the same as waste?

It overlaps. Lean describes waste as activity the customer would not pay for. Friction is narrower and more useful operationally: it is capacity consumed by the process itself rather than by the outcome.

How much friction is normal?

There is no universal figure, and any consultant who quotes one without measuring your process is guessing. The useful comparison is your own process against itself over time, or against a similar process in the same organisation.

See what your organisation’s work actually looks like.

Zunox captures activity where it happens, attributes it, and turns it into evidence you can act on.