The cost of detective work at close

Ask any finance team what slows down the close, and reconciliation itself is rarely the honest answer anymore, especially where matching is largely automated. The real time sink is the small, recurring set of line items that don't match cleanly and require someone to actually investigate — checking an old email, calling a colleague, cross-referencing a system nobody else opens — before the close can finish.

Why this cost is easy to underestimate

Because it's distributed across dozens of small investigations rather than concentrated in one visible task. Ten minutes here tracing an unusual deposit, twenty minutes there confirming why a variance exists, five minutes confirming an old exception is still active. None of it looks significant in isolation. Added up across a monthly close, over a year, it's a meaningful chunk of a finance team's time spent not on analysis or planning, but on detective work that a slightly better-documented business wouldn't require at all.

What makes an item require detective work in the first place

Almost always the same underlying cause: something happened — an exception was granted, a judgement call was made, a configuration was set a certain way — and the reasoning behind it was never recorded anywhere close to the record it explains. The close doesn't fail because the transaction is wrong. It slows down because confirming the transaction is right requires reconstructing context that should have been captured once, at the time, instead of every month, indefinitely.

This is the same underlying gap this cluster has approached from several directions — see what your ERP does not record — showing up here as a specific, recurring tax on the close process rather than an occasional audit or board-level cost.

The specific pattern worth noticing

If the same type of item requires detective work every month — a particular customer's non-standard terms, a recurring supplier adjustment, a specific account that never quite reconciles cleanly — that's not bad luck. That's a standing signal that the underlying reasoning was never captured, and the business is choosing, month after month, to pay the reconstruction cost repeatedly rather than pay the documentation cost once.

Why "we'll just get faster at the detective work" is the wrong fix

Teams often respond to this by getting more efficient at the investigation itself — better search habits, a faster way to track someone down, a personal shortcut for the usual suspects. This helps at the margin and misses the actual fix, because it treats the investigation as inevitable rather than as a symptom of something upstream that could be prevented. The investigation exists because a reason wasn't recorded. Getting faster at investigating doesn't address that; it just makes living with the gap slightly less painful.

Closing this specific leak

Track which line items require investigation, close after close, and treat repeat offenders as a documentation priority, not just a recurring annoyance to be efficiently handled again next month.

Attach the resolved reasoning to the record once it's found, so the next close doesn't require redoing the same investigation from scratch. If this month's detective work uncovers why a particular account never quite ties out, write that down against the account, immediately, before the insight is lost again.

Push documentation upstream, to the point the exception or judgement call is first made, rather than accepting recurring investigation as a permanent cost of doing business — see documenting exceptions as they happen not after. The close is the wrong place to first discover why something happened; it should be the place that simply confirms what was already recorded.

Common questions

Why is detective work during close such an underestimated cost?

Because it's spread across many small investigations rather than concentrated in one visible task, so no single instance looks significant. Added up over a year of monthly closes, the total time spent reconstructing context that was never recorded can be substantial, even though it never appears as its own line item anywhere.

What usually causes a line item to require investigation during close?

Almost always a missing reason: an exception, judgement call, or configuration decision that was made without the reasoning being recorded anywhere near the record it explains. The transaction itself is usually correct — confirming that it's correct is what takes the time, because the context has to be reconstructed rather than simply looked up.

Why isn't getting faster at the investigation itself a real fix?

Because it treats the investigation as an unavoidable part of closing the books, rather than as a symptom of a documentation gap that could be prevented. Getting faster at reconstruction reduces the pain slightly but doesn't address why the reconstruction is necessary in the first place.

How can a finance team reduce recurring detective work at close?

Track which specific line items require investigation every close and treat repeat offenders as a documentation priority, attach the reasoning to the record permanently once it's finally found, and push the habit of documenting exceptions upstream to the point they're first made, rather than accepting reconstruction as a routine part of every close.

Related: what your erp does not record · documenting exceptions as they happen not after · why audit season keeps taking longer than the numbers justify