How Gonav Ledger cut a two-week close to four days
A finance team was losing the first half of every month to reconciliation. Here is what actually moved the number — and what didn't.
The client closed their books on the 12th business day, every month. Their team of six spent the first two weeks matching bank lines to the ledger by hand.
Where the time actually went
We instrumented one full close before changing anything. The finding surprised everyone: only 18% of the time was matching. The rest was chasing context — which cost centre, which entity, whose approval.
What we changed
- A rules engine that auto-matched the 80% of lines that follow predictable patterns
- A single queue for the exceptions, with the context attached
- A close calendar that made the critical path visible to everyone
What we didn't change
We left the judgement calls with the accountants. The goal was never to remove people from the close — it was to stop the close from removing people from everything else.
The next close landed on day 4. Six months on, it still does.
More insights
Wayfinding, not roadmaps: how we plan across five industries
A roadmap is a hypothesis with a date attached. Here is the planning practice we use instead when the terrain keeps shifting.
What “embedded” really means when Gonav Build joins your team
Staff augmentation and embedded delivery look similar on an invoice. They are not the same thing. A field guide.
Re-planning when the trip breaks: inside Gonav Wander
Anyone can generate an itinerary. The hard part is the second one — the one you need at 6am when the train is cancelled.