Operational audit: the bottlenecks nobody sees

An operational audit shows you what your team no longer sees: the manual steps, the waiting and the double entry that have quietly become normal. None of them is a disaster. They are small frictions that, added up week after week, cost more than many projects that do go through a board meeting.

The trouble is that nobody reports them. The person copying an order from an email into the ERP does not see it as a problem; it is just their job. Whoever builds the Monday report rarely asks whether anyone reads it. That is why these leaks never show up on a dashboard until someone goes looking for them on purpose.

What an operational audit is (and what it is not)

An operational audit is a structured review of how work flows through a company: who does what, with which tools, with which data and in which order. Its goal is not to blame anyone or rate people's performance. It is to find the points where a process stalls, gets duplicated or depends on a single person.

It is not a financial audit or a compliance check. It does not verify that the books balance or that you meet an ISO standard. Nor is it strategy consulting: it will not tell you which market to enter, but whether your operations can carry the business you already have.

The practical difference is the deliverable. A good operational audit ends with three things:

• A map of the process as it actually runs today, not as the manual describes it.
• A list of costed frictions: how many hours, how many errors, how much delay.
• A sequence of action: what to change first, what next, and what is not worth touching.

The bottlenecks that keep coming up

Every company has its own, but some patterns repeat in almost every business of 10 to 200 people that has grown quickly.

1. Data re-keyed by hand between tools

The most common one. An order arrives by email, someone enters it in the ERP, someone else logs it in a spreadsheet for tracking and, at month end, a third person reconciles it against invoicing. Four records of the same fact, four chances to get it wrong.

How to spot it: any task where someone has two windows open and moves information from one to the other.

2. Approvals that live in the inbox

A quote that needs sign-off from a director, a purchase finance has to approve, a holiday request a manager must accept. When approval depends on someone reading one email among a hundred, the process stops without anyone noticing. When that person is travelling, it stops completely.

How to spot it: questions like "did you get what I sent?" or "who needs to sign this?".

3. Reports that get built and never used

Many teams spend hours each week assembling spreadsheets or slide decks that leadership skims in two minutes. Sometimes the report is useful but arrives too late; sometimes it answers a question nobody asks anymore.

How to spot it: if the report stopped going out for a week, would anyone ask for it?

4. Knowledge locked in one person's head

"Only Sarah knows how that works." When a critical process depends on one specific person, every sick day, holiday and resignation becomes an operational risk. The issue is not Sarah; it is that the process was never written down.

How to spot it: tasks that slip or go wrong whenever a certain person is out.

5. Several versions of the truth

Sales has one revenue figure, finance has another and the board pack has a third. Each is right according to its source, and none of them match. Meetings are spent arguing about which number is correct instead of deciding what to do about it.

How to spot it: meetings that open with "according to my numbers…".

How an operational audit works, step by step

The method changes with company size and industry, but the phases are the same.

1. Set the scope. You do not audit "the whole company" at once. Pick two or three processes leadership already sees as painful or that carry the most volume: order to invoice, opportunity to contract, ticket to resolution.

2. Talk to the people who do the work. Leadership knows how the process should run; the people executing it know how it really runs. Interviews happen role by role, ideally with their screen in front of you, to see which tools they use and how.

3. Draw the real process. From the interviews you build a map of each process: steps, owners, tools, data in and out, waiting time. This is where the double entry and the steps nobody knew about usually appear.

4. Measure. Each friction gets a cost estimate: hours per week, error rate, waiting time, impact on the customer. Lab-grade precision is not needed; a reliable order of magnitude is enough to compare one friction with another.

5. Prioritise. Two axes: the impact of fixing each friction and the effort it takes. High impact, low effort goes first. Low impact, high effort usually stays as it is.

6. Hand over a roadmap. The output is not a hundred-page report but a short document leadership can read and act on: what to change, in what order, with what resources and what result to expect.

Signs you need an operational audit

You do not have to wait for a crisis. These signs usually mean it is time to look at operations with a method:

• The team is growing, but the administrative load is growing faster than sales.
• Data errors (a mistyped order, a duplicated invoice) are always fixed after the fact.
• Some processes can only be carried out by one person.
• Leadership asks for figures and getting them takes several days of manual work.
• A new tool has been bought, yet the team still does the same work in parallel in a spreadsheet.

If you recognise two or three of these, the audit often pays for itself with what it finds in the first few weeks.

An illustrative example: a company of 40 people notices that month-end closing takes a week because three teams keep their own list of orders. The audit does not start with software. It starts by mapping who creates each list, who reads it and which one is used when the figures disagree. Only then does it become clear whether one list is enough or whether a connection between tools is needed.

How to put a cost on a manual task

The simplest way to cost a friction is to multiply the time it takes by the cost of the person doing it. An illustrative example: if someone who costs the business £35,000 a year spends five hours a week moving orders from email into the ERP, that task takes about 12% of their working week, around £4,000 a year in time alone.

That number is a floor. It leaves out errors (a mistyped order that turns into a return), delay (the customer waits an extra day) and opportunity cost (what that person could be doing instead). It is still enough to compare frictions and decide where to act first.

What to do with the findings

An operational audit is worthless if it ends up in a drawer. The roadmap usually mixes three kinds of action:

• Process changes with no technology: drop a step, change who approves, stop producing a report.
• Automation: connect tools so data is entered once and travels on its own.
• Training: make sure the team uses what it already has well, including AI in their own role.

Order matters. Automating a messy process just makes the mess move faster. Fix the process first, then automate it.

That is what we do in QuO Status: look at how your team really works and leave you, in writing, what to change, in what order and why.

Frequently asked questions

How long does an operational audit take?

It depends on scope. For two or three processes in a mid-sized company, three to six weeks is typical, including interviews, analysis and delivery of the roadmap.

Does the team have to stop work during the audit?

No. Interviews are scheduled in short blocks around each person's calendar. The point is to observe real work, so the team should carry on as normal.

What is the difference between an operational audit and process consulting?

The audit is the diagnosis: it looks at how operations run and says what to change. Process consulting also covers the redesign and, often, support during implementation. Good process consulting always starts with an audit.