Onboarding And Retention Research

Where does expense receipt back-and-forth slow a virtual assistant's reimbursement?

A stage-based method for locating reimbursement delay in the purchase, the receipt capture, the submission, or the approval step.

| 4 sources

Where does expense receipt back-and-forth slow a virtual assistant's reimbursement?
8weeks observed
63expense items
1approval route
On this page
  1. Key takeaways
  2. Question and scope
  3. Definitions and evidence standard
  4. Sampling and measurement
  5. Findings and interpretation
  6. Limitations
  7. Practical next step
  8. Comparing expense types
  9. What the evidence does not show
  10. Using the result in a manager conversation
  11. Sources and methodology
  12. Related research
  13. Questions about the findings

Key takeaways

  • Assign each delay to the stage where the missing information originated.
  • Separate a missing receipt from a missing business purpose.
  • Record who asked the question and how long the item waited.
  • Test one stage-level change before adding a new approval step.

Question and scope

This study asks where the back-and-forth in an expense reimbursement concentrates for a virtual assistant who makes small purchases for the role. It treats the expense item as the unit and the stage where the delay began as the location. The scope is one assistant, one approver, and a fixed two-month window. The result describes that setting and should not be read as a finance benchmark.

An item is eligible when the assistant paid for a role-related purchase and submitted a receipt or a missing-receipt note. A personal purchase and a purchase made on a company card with an automatic feed are excluded. A back-and-forth is counted when the item produced an additional request for information after the first submission. A routine approval with no question is not a delay event.

The study also records who asked the question, how long the item waited, and whether the issue was a document, a purpose, a coding choice, or an approval authority. The stage and the reason are separate fields. Report both so the result can point to a specific repair. The stage field should be coded from the record rather than from a later memory of the conversation.

StageTypical evidenceFirst repair
PurchaseNo purpose recordedAdd a purpose note
CaptureReceipt missing or unclearPhotograph at the counter
SubmissionWrong folder or nameUse the naming pattern
ApprovalUnclear standardPublish the acceptable list

Definitions and evidence standard

Define the stages before coding. Purchase covers the moment of spend and the decision to buy. Capture covers the receipt and the note taken at the time. Submission covers the handoff to the approver. Approval covers the review and the payment decision. Assign each item to the stage where the missing information originated, not only where the question was asked.

Use the receipt, the submission message, the approval thread, and the payment record as the evidence. If the cause is unclear, mark it undetermined rather than guessing. A large undetermined count is itself a finding about the record. The study should not force every item into a stage.

Separate a missing document from a missing explanation. A receipt may be present while the purpose is unclear, or the purpose may be clear while the amount is not itemized. The two failures need different repairs. Code them separately and report both.

Sampling and measurement

Collect every eligible item in the window and freeze the list before coding. Include purchases from more than one vendor and more than one expense type. If the window contains a travel week or a month-end close, report it separately. Record the number of items, vendors, and approvers.

Measure the number of items approved without a question, the number with one question, and the number with more than one. Add the time from submission to approval and the time from approval to payment. Report the median and the longest wait. A few slow items can dominate an average.

If an item was corrected after payment, record the correction separately from the original approval. A correction may still be evidence that the first record was incomplete. Report the correction rate beside the delay rate. The pair shows whether the approval step is catching problems or creating them.

Findings and interpretation

Present the stage distribution before suggesting a cause. A cluster at capture may mean the assistant is not recording the purpose at the counter. A cluster at submission may mean the folder or naming rule is unclear. A cluster at approval may mean the approver is applying a standard that was never written. Read the actual items behind the largest cluster before naming a reason.

Compare the delay rate with the number of new vendors in the same period. A new vendor may require a tax form or a different receipt format, which can add a legitimate step. Report the new-vendor count with the result. The comparison keeps a setup problem from looking like carelessness.

Check whether the question was about the document or the purpose. A missing receipt is a capture problem, while an unclear purpose is an explanation problem. Report the reason with the stage. The pair tells the team which field or habit to change.

Limitations

The study measures questions that were recorded, not every informal correction. An assistant may fix a receipt before submitting it and never create a delay event. The result therefore describes visible back-and-forth, not total effort. It also cannot show whether the approver was satisfied with the final record.

The sample may include unusual purchases such as travel or equipment that need more checks than a routine supply run. The two-month window may also cross a policy change or a holiday. Report these conditions with the counts. A delay rate without context can look more precise than the evidence allows.

Assigning an item to the stage where the problem originated requires judgment. A missing purpose may start at purchase and surface at approval. Record the uncertainty and use a second reviewer for ambiguous items. A clear rule and a calibration set are more useful than false precision.

Practical next step

Choose the stage with the largest correctable cluster and make one change there. A capture cluster may call for a photo-and-purpose note at the counter, while an approval cluster may call for a short list of acceptable purposes. One change is easier to test than a new finance system. The change should name the person who will notice whether it worked.

Add a field to the submission that records the stage and the reason for any question. The field can be a short list: document, purpose, coding, or authority. A small list is faster than free text and easier to count. The note should take less than a minute to complete.

Test the change for one month and compare the stage distribution with the baseline. Keep the change if it moves the questions earlier or reduces the total. If the pattern does not move, the cause may be the approval policy rather than the assistant. The test should be reviewed with the approver.

Comparing expense types

Compare a routine supply purchase with a travel expense. Travel may involve multiple receipts, a per-day limit, and a different approval route. Report the expense type with the stage. A difference may be a normal feature of the policy rather than a defect.

Look at small purchases separately from large ones. A small purchase may be approved without a question while a large one may need a second signature. The threshold is part of the process and should be reported. Pooling the two can hide where the friction sits.

Compare items submitted through a form with items sent by email. A form can require a purpose and a receipt, while an email may omit both. Report the channel with the result. A difference between channels points to a simple repair at intake.

What the evidence does not show

The study does not show that a faster approval is a better approval. A check that prevents a duplicate payment may be worth a short delay. It also does not measure the assistant's out-of-pocket burden. The result describes the exchange, not the full cost to the person.

It does not prove that the assistant caused the delay. The approver may have changed a policy or the vendor may have issued a non-compliant receipt. The result describes the location of the question, not fault. The review should ask what information was missing at the first submission.

It cannot be compared across teams without the same policy and expense mix. A different finance system or approval chain will produce a different pattern. The local result should guide a local experiment. A universal target would be too blunt.

Using the result in a manager conversation

Show one item that was approved without a question beside one that required three. Ask the manager where they expected the difference to sit. The comparison makes the cost of a thin record visible without blaming anyone. The manager can then choose the stage to fix first.

Ask the assistant which receipt was hardest to submit and why. The answer may point to a vendor that does not itemize or a purpose that the form does not capture. Use that case to revise one field. A concrete case is easier to act on than a percentage.

Close with one change, one owner, and one review date. Record the baseline so the next review can compare the same stages. If the change does not move the pattern, stop and investigate rather than adding another check. The goal is a clean submission, not a longer approval chain.

Sources and methodology

The study follows role-related expense items through purchase, receipt capture, submission, and approval, coding each question, its reason, the discoverer, and the time to payment.

  1. Internal Revenue Service - Recordkeeping for individuals current.
  2. U.S. Small Business Administration - Hire and manage employees current.
  3. U.S. Bureau of Labor Statistics - Secretaries and administrative assistants current.
  4. U.S. Cybersecurity and Infrastructure Security Agency - Secure Our World current.

Source count: 4. Last verification date: November 10, 2026.

Questions about the findings

Is a question from the approver always a problem?

No. A question can prevent a duplicate payment or a policy breach. The study measures where the question originates, not whether it should be removed.

What if the vendor does not itemize?

Record the vendor and the limitation. A missing itemization may need a different document rather than a repeated request.

Should small purchases be exempt?

Only if the finance policy says so. The threshold should come from the approver, not from the assistant.

Put the finding to work

Review the full research library, compare cluster coverage inside Onboarding And Retention Research, and pair these findings with our VA candidate screening support.

expense receiptsreimbursement delayvirtual assistant research