Take a close apart and it is 6 or 7 distinct jobs wearing 1 name. Evaluating financial close software works much better when you look at them separately, because the tools in this category do very different things to each one.
Here is the list, roughly in the order it blocks: bank reconciliation, matching payments to bills, chasing missing receipts and documentation, intercompany eliminations, accruals and prepayments, revenue recognition and deferrals, then review and sign-off.
Now the useful question. For each job, does the software you are evaluating coordinate it, or remove it?
days at Tillo across 4 entities and 23 currencies, with prep landing in the first 3 to 4
days to close 4 countries at All Gravy, with nobody working nights to get there
of consolidation at Oper Credits, which now posts as transactions happen
Coordinate or remove
Coordination is what most of this category sells. A checklist with owners and dependencies, a status board, reminders, a sign-off trail. It is genuinely useful when a close involves 9 people across 3 time zones, and it does nothing to the amount of work. A well-coordinated 12 day close is still 12 days.
Removal is a property of the ledger rather than of a tool sitting on top of it. Bank reconciliation runs continuously against live feeds instead of once, so there is no unmatched list to work through. Payments match their bills on arrival while the reference is intact and the amount is unambiguous. Receipts get chased the day they go missing, in Slack, rather than 3 weeks later from a spreadsheet of exceptions. Intercompany eliminations draft as the transaction posts, because both legs are the same record. Accruals get prepared by an agent and left for a person to review.
Each of those removes a task from the list rather than assigning it an owner and a due date.
What survives
Review and judgment, and they should. Estimates. Accruals that need a decision rather than a repeat. Unusual contracts. The question of whether a variance is the forecast being wrong or the business changing.
The difference is what those decisions are made against. On a batched close, judgment happens at the end of a week of reconstruction, when the books have only just become trustworthy and everyone is tired. On a continuous one, the books were current the whole time and the close becomes confirmation instead of assembly.
That is also why the numbers land where they do. Tillo closes 4 entities and 23 currencies in 5 days, down from 12, with prep work now finishing inside the first 3 to 4 days. All Gravy closes 4 countries in 5. Ocean.io cut close time 60%. Alva Labs is targeting a same-day close. Omnio, at 15 days today, is moving onto Light aiming for 1.
Coordination tools
The work gets managed
Tasks, owners, dependencies, status, sign-off. Valuable for a large distributed team, and the volume of work is unchanged.
Ledger-level
The work gets removed
Reconciliation, matching, chasing and eliminations run as transactions post, so there is no accumulated list for month-end to work through.
What is left
Judgment on current books
Estimates, unusual contracts and variance questions, decided against a ledger that is already reconciled.
The questions that separate the two
Where does the bank feed connect? If reconciliation happens in the close tool rather than the ledger, the ledger is still stale between closes and the cash position is a monthly fact.
What happens to an exception in week 2? A tool that collects exceptions for month-end has accepted the batch. One that raises them the day they appear is dissolving it.
Does the tool own any data? Anything that holds its own copy of contract terms, deferral schedules or intercompany balances has become a second ledger, and something now has to reconcile the 2.
Is the sign-off trail the audit trail? If close sign-off lives in the coordination tool and the accounting record lives elsewhere, an auditor gets 2 artefacts and has to bridge them.
What is the close doing on day 3? The most revealing question. On a batched close, day 3 is still retrieval. On a continuous one, day 3 is review, and if the answer is that day 3 does not exist, that is the destination.
The stack question underneath
Continuous close has a precondition that disqualifies most setups claiming it: whatever does the work has to see the whole transaction. If accounts payable lives in 1 system, expenses in another and the ledger in a third, nothing can match, chase and post across the gaps, and batching returns at every boundary. A close that is continuous inside 3 systems and batched between them is a batched close with better dashboards.
This is why the shortest closes belong to teams that consolidated the stack rather than the teams that bought the best close tool. All Gravy replaced 7 systems with 1, across 4 ledgers, a card provider and 2 payment platforms. Oper Credits went from 8 to 3. Tillo collapsed 8 ledgers into 1.
"Our close used to be excruciating. Lots of systems, lots of bookkeepers, lots of places where the whole thing could stall."
Sebastian Sandorff Jacobsen, Head of Finance and Ops, All Gravy
If most of a shortlist is coordination software, the honest comparison is not between those products. It is between buying better management of the batch and removing the reason the batch forms. The second one is a bigger decision and it is the only one where the number keeps falling after year 1.
Read why the close is a batching artifact, see how Light closes continuously, or book a demo.