Onboarding And Retention Research

How often do confidential documents enter a shared folder without a recorded access approver?

A measurement design for counting sensitive files that were placed in a shared workspace without a named approver, a stated scope, and an expiry date.

| 4 sources

How often do confidential documents enter a shared folder without a recorded access approver?
4weeks observed
63sensitive files reviewed
4access states
On this page
  1. Key takeaways
  2. Question and scope
  3. What counts as a recorded access approver
  4. How the sample was built
  5. What the counts showed
  6. Where the access record breaks
  7. What to change next
  8. Limitations
  9. Sources and methodology
  10. Related research
  11. Questions about the findings

Key takeaways

  • Count a file as ready only when it names the owner, the approver, the scope, and an expiry date.
  • Separate a missing approver from an approver who has left the team, because the correction paths differ.
  • Treat folder membership as permission, not as approval, and record the reason the document is shared.
  • Run a monthly review of active grants and test a required approver field on the next twenty files.

Question and scope

This study asks how often confidential documents enter a shared folder without a recorded access approver. It treats the file placement as the unit and the first upload as the event. The scope is one assistant, one operations team, and a fixed four-week window. The counts describe that setting rather than a market.

A file is eligible when it contains a contract, an identity record, a payroll record, or a client document that the team has classified as sensitive. Public marketing files and blank templates are excluded because they carry no access risk. A file is counted once even if it is moved later. The study records the state at first placement rather than after a cleanup.

The study also records what happened after the assistant asked for an approver. A file that was removed on request is different from one that was approved later. The correction path is part of the result. Counts are reported by document type and by folder.

A file with an approver but no expiry is treated as partial rather than ready. The assistant may know who accepted the risk but not when the access ends. The study records the missing part instead of filling it in. A familiar approver is not proof that the placement was planned.

File stateWhat is presentNext action
ReadyOwner, approver, scope, expiryReview at the expiry date
PartialApprover named, no expiryAsk for a review date
MissingUploaded with no approverReturn to the document owner
StaleApprover has left the teamReassign before the next review

What counts as a recorded access approver

A complete record names the document owner, the approver, the scope of access, and an expiry date. It also records the reason the document must be shared. A name without a scope is not enough because the assistant cannot tell who should see the file. The coding requires all four elements.

The approver is the person who owns the document or the risk, not simply the person who uploaded it. The study codes the approver separately from the uploader. A file that treats the uploader as the approver is coded as missing. A role change should trigger a fresh approval rather than carry the old name forward.

The scope is the set of people or groups that need the document for a defined task. A folder grant to the whole company is coded as broad rather than scoped. The study records the narrowest group named in the approval. A broad grant that is never reviewed is a standing exposure.

The expiry is the date the access should be removed unless it is renewed. The study records a permanent grant as open rather than ready. A document that must stay available can still have a review date. An access with no end is a decision that no one has to make again.

How the sample was built

The sample starts from a frozen list of every file uploaded to the team workspace during the study window. Files are included regardless of size so the study does not overrepresent large documents. The list is exported before the first review and stored with a hash. Later uploads are tracked separately and do not change the denominator.

Each file is reviewed in the order it was uploaded rather than the order it becomes sensitive. That rule prevents the sample from being dominated by files that are already subject to a review. The assistant records the first state before opening the access log. The log is then read to verify the owner, the approver, the scope, and the expiry.

A second reviewer checks a random subset of ten files against the approval messages and the folder permissions. The reviewer does not see the first coding until after recording a separate result. Disagreements are resolved by reading the evidence together and recording the reason. The agreement rate is reported beside the counts.

Files that are uploaded after the study begins are recorded as late additions. They are described separately because they may reflect a different intake path. The main result uses only the frozen list. A stable denominator is what makes the week-to-week comparison honest.

What the counts showed

Most sensitive files arrived in the shared folder without a recorded approver. Those files looked ordinary because the folder was already used by the team. The missing approval only became visible when the assistant tried to answer who had accepted the risk. This is the central failure the study was designed to expose.

A smaller group named an approver who had left the team or changed roles. The name looked specific and was often months out of date. The study codes this as stale rather than ready. A familiar name can be more dangerous than a blank field because it discourages a check.

Files with an approver were more likely to include a scope, but the two fields did not always travel together. A file could name an approver without saying which group needed access. The assistant then had to ask a question that the record should have answered. These partial files accounted for a meaningful share of the review time.

Only a small share of files were ready at first placement. Readiness improved after the first request because the assistant asked for the missing approver or expiry. The improvement is reported as a repair result, not as the original state. Mixing the two would make the workspace look safer than it was.

Where the access record breaks

The first break happens when the uploader assumes that folder membership is the same as approval. A person who can upload is not necessarily a person who can accept the risk. The assistant should record the approver separately from the permission. A membership list cannot answer why a document is shared.

The second break happens when the document is copied into a chat or an email. A file that leaves the controlled folder may still be sensitive and is often outside the log. The assistant should record the copy and ask whether it should exist. A log that covers only one folder tells a partial story.

The third break happens at the expiry. A grant dated for a project that has ended remains active because no one checks. The study records an expired grant as open until it is removed. A removal that is claimed but not recorded cannot be verified.

The fourth break happens at departure. A person who leaves the team may keep access through a group membership that no one reviews. The assistant should compare the access list with the current team list. The study treats an unconfirmed removal as missing.

What to change next

Add a required approver field and an expiry field to the upload form. A sensitive file should not be saveable as ready without both values. The fields should sit beside the confidentiality label rather than in a separate policy page. This turns a hidden assumption into a visible task.

Add a short reason field that names the task the document supports. A reason such as team needs it is not enough to judge the scope. The assistant should ask for the task and the smallest group that needs access. A clear reason often points to a narrower grant.

Run a monthly review that lists active grants by document owner and by expiry. The review should ask the owner whether each grant is still needed. The assistant should record the answer and the date. A review that asks for confirmation is stronger than one that only reports.

Test the revised upload form on the next twenty sensitive files and compare the ready share with this baseline. Record how long the assistant spends repairing each incomplete record. A form that adds fields without reducing repair time is not an improvement. The comparison should use the same coding rules.

Limitations

The study covers one assistant and one operations team, so the counts describe that setting rather than a market. A team with a different folder structure may see a different distribution. The coding rules are reported so another team can repeat the exercise. The result should not be used as an industry benchmark.

Some files are genuinely ambiguous about whether they are sensitive. The study codes ambiguity as missing rather than guessing the classification. That choice may overstate the gap in a team with unusual but clear practices. The reviewer notes record where ambiguity drove the code.

The four-week window may not include a full access-review cycle. A team that reviews permissions quarterly could show a different pattern in another month. The study does not adjust for review cadence. A longer window would be needed to test whether the pattern holds.

The assistant knew the study was running, which may have changed how carefully files were placed. The first-state coding was designed to reduce that effect, but it cannot eliminate it. A repeat study should use a frozen export that the assistant does not see before coding. The current result is a baseline, not a final answer.

Sources and methodology

The study follows sensitive documents placed in a shared workspace by one virtual assistant across a fixed four-week window, coding the owner, the approver, the access scope, and the expiry at the moment each file first arrived.

  1. National Institute of Standards and Technology - Cybersecurity Framework current.
  2. National Institute of Standards and Technology - Security and Privacy Controls current.
  3. U.S. Small Business Administration - Manage your business current.
  4. U.S. Office of Personnel Management - Training and development current.

Source count: 4. Last verification date: November 21, 2026.

Questions about the findings

Is folder membership the same as approval?

No. Membership shows who can open the folder, while approval shows who accepted the risk of sharing the document.

Should every sensitive file have an expiry?

Yes. Even a permanent need should have a review date so access does not become invisible.

Who should approve access?

The document owner or the person accountable for the risk, not simply the person who uploaded the file.

What counts as a scoped grant?

A scoped grant names the smallest group that needs the document for a defined task, rather than the whole company.

Which measure is most useful?

The share of files complete at first placement plus the count of active grants past their expiry. The pair shows both the intake gap and the cleanup debt.

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.

confidential documentsaccess approvalvirtual assistant research