Where managing financial policies and procedures usually breaks

This process breaks where a policy leaves its authors. Approval limits drift from what the system enforces, published policies lag behind practice, service agreements go unrevisited, and exceptions get granted by email. Each break leaves a trace in the ledger and the approval history.

From the policy author to the people who apply it

Accounting policies are usually written well. The trouble starts at publication. A revised capitalisation threshold or accrual rule gets approved, then sits in a document library while the people posting transactions keep working from an old desk note or a spreadsheet passed down by a predecessor.

The evidence shows up in the general ledger. Look for manual journal vouchers that reclassify items between expense and asset accounts after the fact. If the same type of correction recurs, the published rule and the applied rule have separated. Another clue: staff who cite a policy by its old name, or who cannot say where the current version lives.

Approval limits kept in separate places

Delegations of authority are typically approved by a board or executive committee and recorded in a policy. Separately, someone configures routing rules in the purchasing, invoice and journal systems. These are rarely updated together.

When they diverge, invoices route to the wrong approver, or to nobody. People then forward items manually so payments are not held up. Signs of this include approvals logged by someone outside the expected chain, invoices sitting in a queue owned by a departed employee, and journals approved by their own preparer. Pull the system routing table and compare it line by line with the signed delegation. Gaps are common, and auditors find them before finance does.

Service agreements between finance and its internal customers

Agreements with shared service centres or business units often define turnaround for invoice processing, close inputs and reporting packs. They tend to be negotiated once and then ignored. Nobody measures against them, so both sides quietly invent their own expectations.

You can see this when business units chase finance through informal channels, or when finance sets cutoffs that the agreement never mentions. Late accrual submissions are a strong indicator. If the close calendar and the agreement disagree about deadlines, the calendar has won and the agreement is fiction.

Common systems configured differently by each unit

A single chart of accounts and shared ledger are meant to make reporting consistent. In practice, units add local accounts, attributes or cost centre structures to suit their own needs. Each change looks harmless.

The cost arrives at reporting time. Consolidation needs mapping tables, eliminations need manual adjustment, and reports cannot be traced cleanly to ledger balances. Watch for crosswalk spreadsheets maintained outside the system, and for accounts that exist in one entity with no equivalent elsewhere. When a reporting analyst is the only person who understands the mapping, the common system has fragmented.

Exceptions that become the real policy

Every policy needs an exception route. The failure is when exceptions are approved by email and never logged. Over time the approved exception becomes standard practice, and the written policy describes something that no longer happens.

To detect it, ask for the exception register. If there is none, search approval emails for phrases like "just this once" or "pending policy update." A high share of payments or journals flagged as urgent or out of process points the same way.

Audit findings that never reach the policy owner

Internal control reviews and external audits identify needed adjustments. Those adjustments are booked, the finding is closed, and the underlying procedure is left unchanged. The same finding then reappears in the next cycle under different wording.

Cycle memos are the best evidence. If they describe controls that staff do not recognise, or have not been revised since a system change, the documentation is decorative. Compare open and closed findings across audit cycles and look for repeats.

Questions to ask the people who run it

What staff do often differs from what is documented. These questions tend to surface the difference:

• When a new accounting policy is issued, how do you find out, and what do you change in your own work?
• Which document do you actually open when you are unsure how to treat a transaction?
• Has an invoice or journal ever reached you that you were not supposed to approve? What did you do with it?
• Who do you ask when the system will not let you post something you believe is correct?
• Which deadlines do you work to, and where did you learn them?
• Do you keep any spreadsheet that the reporting depends on? Who else knows how it works?
• What happens when a requester needs something outside policy? Who says yes, and where is that recorded?
• After the last audit, did anything about your procedure change?
• Which part of the process would you fix first if nobody needed to sign it off?

Ask these individually and away from managers. Answers from preparers and approvers often contradict each other, and that contradiction is where the redesign work should begin.

APQC's Process Classification Framework® (PCF) is an open standard developed by APQC, a nonprofit that promotes benchmarking and best practices worldwide. To download the full PCF or to view definitions and measures, please visit www.apqc.org/pcf.