When the Client Disputes the Hours

A client questioning a bill asks for the detail. What arrives on their desk is your timesheet data, formatted, and they read it with an attention nobody inside the organisation has ever applied to it.

The record described in “When the Client Disputes the Hours” should be created close enough to the work that people are not reconstructing a polished week from memory. When teams assess the official product page for how to measure employee productivity, they should keep entry, project selection and correction simple, while explaining which optional activity data is collected and how employees can review it.

What they are looking for is not fraud. They are looking for entries they cannot understand, and every one of those is a candidate for removal. The write-off that follows is usually not about hours that were not worked; it is about hours that cannot be explained.

For a separate benchmark relevant to “When the Client Disputes the Hours”, consult the European Commission project-management guidance. Use it to test record quality, approvals, retention, employee rights and exception handling against the real workflow rather than treating a software report as self-explanatory evidence.

What they examine

Narratives, first and hardest. Anything generic comes out. Anything that does not obviously relate to the matter comes out. Anything that looks like internal administration comes out, because most engagement terms exclude it.

Duplication: two people recording the same meeting, which is legitimate and looks like double-billing unless the narratives make the distinction.

Seniority: expensive people doing work that could have been done by cheaper ones. This is an argument about the engagement rather than the record, but it is found in the record.

Round numbers: a long run of neat hours suggests estimation, and estimation suggests the detail was reconstructed.

And timing: entries created weeks after the work, which they can often infer from the pattern even without your metadata.

Why the hours were probably worked

This is worth holding on to in the internal conversation. The common failure in billing disputes is not inflation; it is unexplainability.

A person who worked six hours on a document and wrote "drafting" has made six real hours indefensible. The firm then writes them off, the person is told their realisation is poor, and the actual defect — that nobody required a usable narrative at the time — goes unaddressed. This happens continuously in professional firms and is almost always diagnosed as a performance issue.

What to be able to produce

A time ledger by matter with date, person, duration, narrative and rate, exportable without manual work. Most firms can do this.

Evidence of contemporaneity if asked: the creation dates. Few firms can do this and it is extremely strong when it is available.

The engagement terms and the rounding rule applied, reconciled. If your system rounds differently from what the terms say, find out before the client does.

And a record of the approval, with the name of the person who affirmed it. Clients increasingly ask who reviewed the time, and "it was approved by the matter partner" is a much better answer than silence.

Handling the query itself

One owner, not the individual whose entries are questioned. Asking the person who recorded the time to defend it line by line produces defensiveness, takes days, and places somebody junior in a commercial negotiation.

Review internally before responding, and remove what you would not defend. Conceding the weak entries up front strengthens the position on the rest and shortens the whole thing considerably.

Do not amend the underlying records to match what you end up billing. The record says what was worked; the invoice says what was charged; they are allowed to differ, and making them agree retrospectively destroys the audit trail and the evidence that the hours were real.

Feeding it back

The write-offs are a quality measurement that the organisation has already paid for, and in most firms the information stops at finance.

Reporting, per team and per period, what was written off at query and why — generic narrative, excluded activity, duplication — turns a cost into the only piece of downstream feedback the recording process ever gets. It changes behaviour faster than any instruction about narrative quality, because it is specific and it is about money.

What the engagement letter could have done

Most of what gets argued about at invoice could have been settled at engagement: the increment, the narrative standard, what is excluded, whether travel is chargeable, who may work on the matter, and the notification threshold before exceeding an estimate.

Firms treat these as boilerplate and clients treat them as negotiable, and the result is a set of terms nobody has reconciled against the system that will produce the bills. Reading the engagement terms against the timesheet configuration, once per client type, is a short exercise that removes a recurring category of dispute entirely.