On this page Jump to section
Abstract. Delegated QuickBooks work often ends in a queue waiting for the owner. This paper defines owner approval latency, describes the timestamp and censoring problems that distort it, and proposes a monthly report that separates volume, decision difficulty, and review capacity.
The protocol measures local workflow. It does not benchmark one business against another or claim that shorter approval times are always better.
Why approval latency matters
A bookkeeping queue can look healthy while work is stuck. Bills are coded, receipts are attached, and the reconciliation is ready, but the owner has not approved the payment or the reclassification.
The visible symptom is a late close or a missed payment. The measurable cause is the wait between ready and decided.
Measuring that wait turns an interpersonal problem into an operational one. The fix may be a scheduled review window, a clearer delegation limit, or fewer decisions that require the owner at all.
Define the endpoints
The start is the moment an item becomes genuinely ready for a decision: coding complete, support attached, amount confirmed, and the specific question stated.
The end is the owner's recorded decision: approved, rejected with a reason, or returned for more evidence. A read receipt is not a decision.
Write both definitions into the report. A latency metric that starts when the assistant first thought about the item measures something else.
Use one clock source
Timestamps from chat, email, the accounting system, and the assistant's own notes will disagree. Pick one system of record for the start and one for the decision, and record the timezone.
If the decision arrives in chat but is recorded in QuickBooks later, use the chat time and note that the ledger was updated separately. Otherwise you measure data entry, not approval.
Record the source of each timestamp. A metric assembled from memory is not reproducible.
Handle censoring
Items still waiting at the reporting cutoff have no latency. Excluding them makes the queue look faster exactly when it is backing up.
Report them as open at cutoff, with their age. Compute the latency distribution only for items decided within a fixed window, and report the share of the cohort that was still open.
This is the same structure used for response-time measurement in support queues: the unanswered items are part of the result.
Segment by decision type
An owner approves a routine bill, a payment release, a reclassification, and a refund differently. Pooling them produces a number that describes no decision.
Use a small taxonomy: routine coding, payment release, correction, policy question, and exception. Report the count and median latency per type.
A long wait on routine coding suggests a scheduling problem. A long wait on policy questions suggests the question was not ready when it was sent.
Measure the review load
Latency is partly a function of how many items arrive at once. Record the number of decisions requested per day or week, and the owner's stated review window.
If the queue spikes after the assistant batches everything into one message, the batching rule is the problem. Spreading submissions across the week may reduce the median wait without changing capacity.
Report the shape, not just the average
Report the median, the 75th percentile, and the oldest open item. The average hides the cases that block a close.
Pair the numbers with one or two examples of items that waited longest and why. The examples make the report useful to an owner who does not read distributions.
Limitations
The protocol treats every decision as equally important. It does not weight by amount, risk, or regulatory deadline.
It cannot prove that faster approval improves accuracy. A rushed approval may create rework that the latency metric does not see.
It also depends on honest recording. If decisions are backdated to the request date, the metric will flatter the process.
Sources
- Committee of Sponsoring Organizations of the Treadway Commission, Internal Control over Financial Reporting guidance: https://www.coso.org/guidance-on-ic
- U.S. Government Accountability Office, Standards for Internal Control in the Federal Government: https://www.gao.gov/green-book
- PCAOB Auditing Standard 1105, Audit Evidence: https://pcaobus.org/oversight/standards/auditing-standards/details/AS1105
- Internal Revenue Service, How long should I keep records: https://www.irs.gov/businesses/small-businesses-self-employed/how-long-should-i-keep-records