Onboarding And Retention Research

How often does a new virtual assistant's question already have a written answer?

A bounded method for coding first-month questions against the existing reference material and turning repeated gaps into a documentation repair.

| 4 sources

How often does a new virtual assistant's question already have a written answer?
4weeks observed
5question groups
1library version
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 weeks
  9. What the evidence does not show
  10. Using the result in a client conversation
  11. Sources and methodology
  12. Related research
  13. Questions about the findings

Key takeaways

  • Match each question to the library version that existed on the day it was asked.
  • Separate a missing entry from an entry the assistant could not find.
  • Treat a verbal answer as a future gap until it is written down.
  • Choose one repair and retest with the same answer categories.

Question and scope

This study asks what share of a new virtual assistant's first-month questions already have an answer in the written reference material. It treats the question as the unit and the existing library as the standard. The scope is one role, one client, and a fixed onboarding window. The result describes that setting and should not be read as a benchmark for other roles.

The unit is one question asked by the assistant during the window. A question is eligible when it concerns a task the role performs and can be matched to a topic in the library. Personal questions and questions about the assistant's own schedule are excluded. Record the inclusion rule before reviewing so the denominator is stable.

The study also records whether the question was asked once or repeatedly. A single question may be a normal learning moment, while a repeated question is stronger evidence of a gap. Report the repeat count separately. That distinction keeps the result from treating every first-week question as a documentation failure.

Question typeWritten answerPartial answerNo answer
Routine taskCount each caseRecord the missing decisionName the missing topic
Client preferenceCite the confirmed sourceNote the open choiceRoute to the owner
ExceptionQuote the ruleRecord the judgment left openAdd the example

Definitions and evidence standard

Define a written answer before looking at the questions. An answer exists when the library contains a passage that addresses the question and is current at the time it was asked. A partial answer exists when the library addresses part of the question but leaves a decision open. No answer exists when the topic is absent or the entry is out of date.

Use the question log, the library version, and the date of the question as the evidence. Do not use the current library to judge a question asked before an update unless the update date is recorded. A later addition does not prove that the answer existed when the assistant needed it. That timing distinction is central to the result.

Sampling and measurement

Collect the questions from the assistant's own log or from the messages sent to the manager. Freeze the list before coding. If the role generated many questions, sample across the first, second, and fourth weeks rather than only the first few days. A first-week sample may overstate the gaps because the assistant has not yet learned the library.

Measure the number of eligible questions, the number with a full answer, the number with a partial answer, and the number with no answer. Report the counts by week and by topic. Add the number of questions that were answered verbally and never written down. A verbal answer that is not recorded is a future gap.

If the role has more than one client, report the result by client rather than pooling them. A library may be strong for one client and thin for another. Pooling can hide a gap that matters to a single account. Use client codes in the working file and keep names out of the report.

Findings and interpretation

Present the distribution before suggesting a cause. A cluster of no-answer questions in one topic may mean the library was built for a different task mix. A cluster of partial answers may mean the entries describe the process but not the decision. A cluster of verbal answers may mean the manager is the real library. Read the actual updates behind the largest cluster before naming a reason.

Read the questions behind the largest gaps and compare them with the entries. Record whether the entry was missing, hard to find, or out of date. A question that the assistant could not find may still have a written answer. Distinguish a coverage problem from a search problem before recommending a fix.

Compare the no-answer rate with the age of the library entry. An old entry may be out of date even when the topic exists. A new entry may be accurate but hard to find. Report the age and the search result together. The combination points to a different repair than either measure alone.

Limitations

The study measures whether an answer exists, not whether the assistant could understand it. A written answer can be present and still be wrong, ambiguous, or too long to use under pressure. It also cannot capture the value of a question that prevented a mistake. The count is a documentation signal, not a performance measure.

The library may change during the window, and the question log may miss questions asked in conversation. If the role changed scope or the client changed a preference, the questions may not be comparable across weeks. Record these limits with the result. A short methods note is more useful than a precise-looking rate with hidden gaps.

Report the number of questions that were answered by another assistant rather than by the library. Peer help is useful, but it can hide a documentation gap. A question answered from memory is a candidate entry. Keep the peer answer and the date so the owner can confirm it.

Practical next step

Choose one repair that matches the most common gap, such as a missing entry, a clearer title, or a decision example. Write the change, the owner, and the date it starts. Define the measure that will show whether it helped, using the same answer categories. Test it on the next set of questions.

Keep the old and new library versions separate so the comparison is honest. Record the questions that were excluded and why. Review the result with the assistant and the manager. If the change does not help, record that result too. The purpose is to learn which documentation change reduces repeated questions.

Assign the library repair to the person who owns the topic, not to the assistant who asked the question. The assistant can draft the entry, but the owner must confirm the answer. Record the reviewer and the date. A repair without an owner becomes another unverified note.

Comparing weeks

Questions in the first week often reflect a lack of familiarity, while later questions may reflect a gap in the work itself. Report the result by week so the reader can see the difference. A high first-week rate may be normal, while a high fourth-week rate is more likely to indicate a library problem. Ask the assistant directly before treating a quiet week as proof of coverage.

If the fourth week still shows many no-answer questions, check whether the role took on new tasks that the library never covered. If the rate falls, check whether the assistant simply stopped asking. A fall in questions is not proof of coverage. Ask the assistant what they now do when an answer is unclear.

Keep the original question wording in the record. A paraphrase can make a question look more general or more specific than it was. The wording helps the owner decide whether the entry should match the exact phrase or the underlying task. Store the wording without client identifiers.

What the evidence does not show

The study does not establish that the assistant's work was correct or that the client was satisfied. It also does not measure whether the written answer was followed. Those questions need a task-quality review. A covered question is only evidence that the library addressed the topic.

Do not use the result to promise a client that every question will have an answer or that the assistant will never need help. Do not publish client names or message content. Keep the sample size, the answer definitions, and the review decisions with the result. If the firm needs a performance conclusion, it should use a different method.

Keep the question log in a restricted workspace because it may contain client details. Publish only the categories and the aggregate counts. If a client asks for the result, describe the method and the repair rather than sharing raw messages. The study is an internal improvement tool, not a client-facing audit.

Using the result in a client conversation

A client may ask how the team is documenting the role, and the study supports a careful answer. Describe the window, the number of questions, the share with a written answer, and the repair that was made. Do not say that every question is covered or that the library is complete. Explain that the measure covers first-month questions and that the manager reviews open items.

If the client asks for a commitment, offer a process commitment such as a weekly question review, a named library owner, and a dated update. Record the commitment and the measure that will show whether it was met. Keep client-specific details out of any public write-up. The study is an internal onboarding tool, not a service guarantee.

Sources and methodology

The study codes first-month questions against the reference library in use on the date of the question. It reports full, partial, and missing answers with the library version and the review decisions.

  1. U.S. Small Business Administration — Hire and manage employees current.
  2. U.S. Office of Personnel Management — Training and development current.
  3. U.S. Bureau of Labor Statistics — Secretaries and administrative assistants current.
  4. CISA — Secure Our World current.

Source count: 4. Last verification date: November 8, 2026.

Questions about the findings

Does this measure prove that the assistant's work was correct?

No. It describes a documented pattern in one role and one window. A separate quality review is needed to judge the work itself.

Can the result be compared across clients?

Only when the definitions, the task mix, and the review method stay the same. Small groups should be described rather than ranked.

What should the team do first?

Pick one repair that matches the most common gap, record the change, and retest with the same definitions before claiming improvement.

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.

reference coveragevirtual assistant researchonboarding documentation