Part 2 of our three-part series on audit capacity. Discussed in part 1: Too many auditors and not enough

When I was in practice we used to say audit partners age in sign-off seasons, not years. It was a joke, mostly.

The statutory deadline does not move. The client's board date does not move. The regulator's filing window does not move. The only variable in the whole system is the people, so the people absorb everything the plan got wrong.

The plan never has a line for it

I spent 16 years at Deloitte, and the audit years are the ones this piece is about. I have sat in the planning meetings before busy season where the staffing model looks complete. And I have watched the same known unknowns show up every year. A flu round through the team. A partner reshuffle that moves a senior mid-engagement. A departure. Extra findings on a file that was supposed to be clean. A new regulatory requirement dropped into the season.

There is always more work than the plan accounted for, and every gap gets closed the same way: by the people who are still there, working longer.

Firms usually call this a staffing problem. I see 2 problems stacked on top of each other. The team doesn't have enough raw capacity. And too much of the capacity it does have goes into manual, repetitive work instead of review and judgment. Most firms only plan around the first, because that's the one you can see on an org chart.

Where the hours go

During busy season, an audit team moves through a high volume of documents and runs the same procedure again and again. A single test of details means opening file after file, matching evidence line by line, and re-checking the work before it moves up the chain. Multiply that by dozens of tests per engagement and dozens of engagements in the same window, and the manual part becomes the season.

Nobody has published a reliable measurement of how an auditor's time splits across a test of details, so I'll give you mine. In my experience, roughly 2 in every 3 hours go to gathering and matching, about 1 in 4 to chasing exceptions, and what's left to the part that needs a trained auditor to think.

So the biggest share of the hours goes to the work that needs the least training. The smallest share goes to the work that builds experience and judgment. That is a poor way to develop a junior, and an expensive way to run a firm.

What it costs in people

The profession has known this for years. In 2019 the Dutch Labour Inspectorate compared psychosocial workload across professions, and accountancy scored poorly against doctors, IT professionals and hospitality staff. The NBA's young professionals research with Nyenrode, run in 2018, 2020 and 2023, found workload pressure high in all 3 rounds, despite the interventions in between. Maastricht University's ROA names high workload as one of the main reasons young accountants leave the profession early.

Advancetrack's 2026 index puts current numbers on it: 74% of firm leaders believe current workloads could push people out of the profession, and about 1 in 4 respondents are personally considering leaving.

For anyone who owns a P&L, this is the line nobody puts in the engagement budget. A senior who leaves after busy season takes 3 or 4 years of training with them. The firm pays to recruit, onboard and re-train the replacement, who then runs the same manual sequence into the same wall. Keeping people through the years when they become valuable is a margin question as much as a people question.

Why adding people doesn't add capacity

The instinct when a team runs short is to hire. But if the process stays the same, the new hire inherits the same manual sequence as everyone else. They gather, tie out and send work up for review at the same pace. Every hour they spend on that is an hour a senior or manager spends reviewing it, training on it, or redoing it.

Add headcount to a manual process and the review burden grows with it. You have bought hours, and you still don't have more capacity.

What changes capacity is what the people spend their time on. That is a budgeting question as much as an operations one, and it's where I'll go in the last piece.

How RobotX Flow adds capacity to the team →