SAP Closing

Closing brings a period to a clean, final state: completing postings, running periodic programs (depreciation, recurring entries, accruals, foreign-currency valuation, GR/IR clearing), reconciling sub-ledgers to the G/L, closing posting periods, and producing financial statements. Year-end adds extra steps (balance carryforward).

One distinction to make early: closing a posting period and closing the fiscal year are different acts with different controls. The period close is monthly and reversible. The year close is annual, carries the balance forward, and reopening a closed year is a serious matter involving whoever signed the accounts.

• Complete and reconcile sub-ledger (AP/AR/AA) postings.
• Run depreciation, accruals, recurring entries.
• Foreign-currency valuation and GR/IR clearing.
• Close posting periods; produce statements.

A close is a sequence, and each step depends on the one before it. Running them out of order produces numbers that have to be redone rather than an error.

Stop the inflow first. Logistics postings complete: goods receipts, goods issues, billing. Anything still arriving changes what everything downstream calculates.

Then the sub-ledgers. Payables and receivables posted and reconciled to their reconciliation accounts. Asset accounting runs depreciation, which must come after all acquisitions are in, because depreciation is calculated on what exists.

Then inventory. Goods receipt and invoice receipt clearing analysed, work in progress calculated, production orders settled. This is where logistics hands its result to finance.

Then the periodic finance postings. Accruals, deferrals, foreign currency valuation, recurring entries. These take the sub-ledger results as input.

Then controlling. Overhead allocation, assessment and distribution cycles, settlement. Allocating before all costs have arrived allocates the wrong number.

Then close the period with OB52, and produce the statements.

The reason to know the sequence rather than the list: when somebody asks why a figure changed after they looked at it, the answer is nearly always that they looked between two of these steps.

An hour, and it tells you whether a close is under control without attending a single meeting.

1. Check OB52. Which periods are open, for which account types, and for whom. A prior period open to everybody is the first finding.
2. Run the goods receipt and invoice receipt clearing analysis. An aged balance means receipts and invoices are not matching, and it is a real accrual question.
3. List production orders technically complete and unsettled. Cost sitting on orders.
4. Check that depreciation ran for every company code in the period.
5. Look for parked documents. Known liabilities not in the ledger.
6. Reconcile each sub-ledger to its reconciliation account. A difference should be impossible.

Six checks, and between them they find most of what makes a close late.

A fast, reliable close is a hallmark of a well-run finance function. SAP tools (and in S/4HANA, real-time postings and the universal journal, plus Advanced Financial Closing) reduce reconciliation effort and accelerate the close. Strong period-control discipline prevents postings to closed periods.

The levers that actually shorten a close, in rough order of effect. Move work out of the close: reconciliations done weekly rather than at month end. Automate the repetitive: recurring entries, automatic clearing, scheduled valuation runs. Close sub-ledgers earlier than the general ledger, so logistics stops posting while finance is still working. And fix the causes of manual adjustment, because a close with twenty topside journals is a close correcting something that happened wrongly upstream.

• How the period is controlled. Posting period variants let you close for most users while a small group finishes, which is the practical compromise every month end needs.
• Hard close or soft close. A hard close prevents any further posting; a soft close permits it for a named group. Whichever, somebody owns the decision to reopen.
• What is accrued and what waits. A materiality threshold agreed with finance, or every close is an argument about small amounts.
• Who signs off each step. A close task list with owners and dependencies, rather than a shared understanding, is what makes it repeatable when somebody is on holiday.

A year-end close is a month-end close plus a set of activities that only happen once, and each has a sequence of its own.

Balance carry-forward. Balance sheet accounts roll their closing balance into the new year as an opening balance. Profit and loss accounts roll into retained earnings and start at zero. This runs as a program, it must run before the new year can be reported on properly, and it can be repeated if postings arrive into the old year afterwards.

Asset fiscal year change. Asset accounting opens the new year separately, and its year-end close program checks that depreciation is complete for every asset before the old year can be closed. One incomplete asset blocks the whole company code, which is why the report exists rather than guessing.

Number range extension. Document number ranges are often year-dependent, and a range that has no interval for the new year stops posting on the first working day. It is a five-minute task that causes a memorable outage when forgotten.

Foreign currency valuation and provisions at the year-end rates, usually with more scrutiny than the monthly equivalents.

Statutory reporting and audit support, which is where the financial statement version and the quality of the underlying postings are judged by somebody external.

The scheduling consequence: year end is not a busier month end, it is a different task list, and planning it as the former is why it overruns.

The measures a finance function is judged on here are few, and knowing them tells you what a close project is trying to move.

Days to close. Working days from period end to statements signed off. A well-run close is a handful of days; twenty is a symptom rather than a schedule.

Number of manual journals. Each one is a correction of something that happened wrongly upstream, or a calculation the system should be doing. A close with dozens is telling you where the design gaps are.

Reopened periods. Each is a reported figure that changed after it was reported.

Late adjustments. Entries posted after the numbers were first circulated.

The useful reframing for a consultant: these are all downstream symptoms. Shortening a close is mostly not a finance project, it is fixing the goods receipt clearing account, the unsettled orders and the master data that generate the corrections in the first place. Attacking the close directly produces a faster version of the same problems.

• Postings to a period during/after close, control periods tightly.
• Unreconciled sub-ledgers at close.
• Missed periodic programs (accruals, valuation).

• Reopening a closed period without a record. Reported figures change and nobody can explain when or why.
• Depreciation missed for one company code. Silent until somebody compares to last month.
• Allocation run before all costs arrived. See the general ledger, bank accounting, FI configuration and FI examples for the pieces this sequences.

• A close task list that lives in one person's head. It works until they are away, and then the sequence is reconstructed under time pressure.

Running a step is mechanical, and building a close calendar with its dependencies and owners so the month ends on time is the part you do in the course.

The reframing worth carrying: a slow close is usually not a finance problem. It is unsettled orders, an unanalysed clearing account and master data producing corrections, and fixing those shortens the close more than anything done during it.