research
Measuring the share of staff who can restore a file without calling IT
A test-based method for measuring whether staff can recover a deleted file from the company backup or file service without raising a ticket.
Use this benchmark to size repeatable IT work, set the review cadence, and decide what stays with the technical owner before assigning the workflow to an IT virtual assistant.
Research playbook
On this pageJump to section
Key stats
Key takeaways
Most file restores are small, urgent, and avoidable. A user who can recover a deleted document from the file service does not need to wait for the helpdesk. This study measures how many staff can complete that task without calling IT, using a short test rather than a survey. The result shows whether the self-service route is usable, not merely available.
Question and scope
The study asks what share of staff can restore a deleted file from the company file service without contacting IT. The unit is one participant. The population is the staff who store work files in the supported service and have access to its restore feature. Record the teams included and the test period. Keep the participant list with the result so the share can be checked.
Exclude staff who only use local storage, contractors without a company account, and users who are on leave for the whole test. Note the teams that use a different file service and set them aside for a separate test. A share that mixes several tools will not tell the service owner which instructions need work. Record the exclusion reason so the share can be explained.
Define a successful self-restore
Define success as the participant recovering the correct version of the test file, in the agreed location, without help from IT. The file must open and contain the expected content. A participant who finds the recycle bin but restores the wrong version has not succeeded. Write the definition before the test so every observer applies the same standard. Use the same standard for every participant in the round.
Decide whether a hint from the test script counts as help. A pointer to the menu is help, while the written task itself is not. Record the level of assistance used. The measure should distinguish a user who knows the route from one who can follow a prompt. Record the help level in the observer sheet for every participant attempt.
Build the participant frame
Draw the participants from the teams that use the file service every day. Include a mix of roles, tenure, and comfort with technology. Do not pick only the people who usually volunteer for pilots. Record the team and the role category without collecting unnecessary personal detail. A sample of confident users will make the guide look better than it is.
Choose a sample size that the team can run in one week and that is large enough to show a change. Note the people who declined and the reason when one is offered. A sample that only includes confident users will overstate the share. Keep the recruitment note with the test plan. Record the refusals so the report can describe the sample honestly.
Design the test task
Create a test file with a clear name and place it in a folder the participant can access. Delete it or move it to the service's recoverable location at a known time. Ask the participant to restore the latest version to the original folder. Give a time limit that is long enough to read the screen and short enough to reflect a real request.
Write the task in plain language and avoid naming the menu or the button. Tell the participant that the exercise is about the instructions, not about their ability. Ask them to say what they are looking for as they work. The observer should record the route, not coach the participant through it. Read the task exactly as written and do not add examples.
Run the test without coaching
Start the session by reading the task once and asking whether it is clear. Then stay quiet unless the participant asks a direct question. Record the time, the steps, and the point where they stopped. If they ask for help, note the question and answer only what was asked. Time each participant from the task read to the first click and record it.
Stop the test when the file is restored or the time limit ends. Ask the participant to show the restored file and confirm the content. Thank them and explain how the result will be used. Do not turn the session into training, because that would change the measure. Record the restored file path and the total time the attempt took.
Record the failure points
For each failed attempt, note whether the participant could not find the feature, did not recognize the version, or did not trust the restore. Record the screen or menu where they stopped. A failure that happens at the same step for several people is a design problem, not a user problem. A repeated failure at the same step is a design problem.
Note the questions people asked and the words they used. If several participants looked for a recycle bin that the service does not have, the instruction is using the wrong mental model. Keep the quotes anonymous. The failure points become the input for the next version of the guide. Use the participant's own words in the notes rather than a paraphrase.
Compare with the helpdesk queue
Count the file-restore tickets the helpdesk received during the same period. Group them by team and by file type. Compare the ticket rate with the self-restore share for the same teams. A team with many tickets and few successful restores needs a different intervention from a team that simply prefers to ask. A team with low tickets may simply call a different person.
Check the time the helpdesk spends on these tickets and the delay the user experiences. A self-service route that saves a ticket but takes longer than calling is not an improvement. Record both the ticket count and the user wait. The comparison keeps the measure tied to service, not only to adoption. Record both the ticket count and the wait the user sees.
Report by team and file type
Report the share of successful restores for the whole sample and for each team. Add the file types that failed most often, such as spreadsheets, presentations, or shared documents. Use counts as well as percentages so a small team is not overinterpreted. Keep personal results out of the team report. Use the same denominator for every team in the table.
Note the participants who succeeded with a hint and the ones who needed a second attempt. Show the median time to restore and the longest attempt. A report that only gives one percentage hides the difference between a quick success and a long struggle. Attach the test script and the observer notes. A hint that changes the result should be reported as help.
Improve the instructions
Rewrite the restore guide around the steps where participants stopped. Use the words they used and show one screenshot for each decision. Put the guide where the file service is opened, not only on the intranet. Ask two participants to test the new version before it is published. Ask two participants to read the draft aloud before it is published.
Add a short note to the file service welcome page and to the new-starter pack. Tell users what the restore can recover and how long the window lasts. Remove the old guide so two versions do not compete. A clear instruction is part of the service, not an extra document. Remove the old guide so users do not land on two versions.
Retest after the change
Run the same task with a new sample after the guide is published. Keep the definition, the file type, and the time limit the same. Compare the share and the failure points with the first test. A rise in success at the same step shows that the instruction, not the sample, changed. Do not change the test file or the time limit between rounds.
Ask the observers whether the new guide was easier to follow. Check the helpdesk queue for the same period to see whether the tickets fell. If the share rose but the tickets did not, the people who need help may still be calling. Report both results before declaring the change a success. If tickets did not fall, the people who need help may still call.
Keep the evidence
Store the test plan, the participant list, the observer notes, and the guide version in one folder. Name the files by date and test round. Remove names from the shared report and keep the consent records separately. Give the service owner read access to the evidence. Give the service owner read access to the evidence folder before the report is shared.
Write a one-page method note that explains how to create the test file and run the session. Include the definition of success and the level of help allowed. Record the date of the last test and the next planned round. The note should let a new reviewer repeat the test without asking the original author. Name the person who will run the test next time.
Limits of the study
The test measures one file type and one restore route. A participant who succeeds in the exercise may still struggle with a folder restore or a version from last month. The setting is observed, so some participants may try harder than they would alone. State these limits when the share is compared with real ticket data. A folder restore is a different task from a single file.
The sample is small and drawn from teams that agreed to take part. A team that declined may have a different experience. The measure also cannot show whether the restored file was the version the user needed for the business task. Use the helpdesk comparison and the failure notes to interpret the result. A declined team may have a very different experience.
Sources and next step
Use the NIST Cybersecurity Framework 2.0 and the CISA Cross-Sector Cybersecurity Performance Goals as primary references for the reopen, laptop order, self-restore, alert acknowledgement, and shared mailbox measures discussed here. The NIST SP 800-53 Rev. 5 supports the access and audit records, the NIST SP 800-34 contingency planning guidance supports the restore measure, and the NIST SP 800-61 incident handling guidance supports the alert review. Apply your own policies, contracts, and system documentation when adapting the method.
Compare this study with the ticket reopen measure and the alert acknowledgement study before combining them into one review cycle. If the measurement work is recurring and your technical owner can define the scope, review the service areas and contact ITVirtualAssistant to discuss a limited pilot.
Benchmark brief
What this research page must produce
A practical estimate for volume, review time, escalation rate, and assistant capacity.
A clear split between routine support, preparation work, and technical ownership.
What the measuring the share of staff who can restore a file without calling it data shows
Treat this as a planning benchmark, not a universal number. Compare the benchmark against your ticket volume, SaaS stack, documentation backlog, and support risk before assigning recurring work.
The useful output is a decision about capacity, not a static statistic. If the workflow is high volume and low judgment, an IT virtual assistant can absorb coordination and upkeep. If the workflow is low volume but high risk, keep it with the technical owner and use the assistant only for preparation, reminders, and documentation.
Workflow
Recommended operating workflow
Collect a baseline
Pull the last 30 to 90 days of examples related to measuring the share of staff who can restore a file without calling it, including completed work and unresolved exceptions.
Classify the work
Tag each item by routine admin, manager approval, technical decision, security risk, or vendor dependency.
Set the operating number
Use the median weekly volume and review time to decide how many assistant hours the workflow deserves.
Refresh the benchmark
Recheck the numbers quarterly so tool growth, new systems, and security requirements do not silently change the scope.
Decision rules
| Metric | Use it to decide | Manager action |
|---|---|---|
| Weekly volume | Whether the workflow is worth assigning as recurring assistant work. | Approve a weekly capacity target and backlog threshold. |
| Access sensitivity | Whether the assistant can work directly or only prepare review notes. | Set least-privilege permissions and removal dates. |
| Escalation rate | Whether the workflow is stable enough to delegate. | Rewrite the SOP when exceptions exceed the agreed threshold. |
Consolidated statistics
| Statistic | Figure | Source |
|---|---|---|
| Unit | One participant | Study protocol |
| Task limit | 10 minutes | Study protocol |
| Retest | New sample | Study protocol |
Sources
- NIST SP 800-34 Rev. 1, Contingency Planning GuideRecovery and testing language for file restore exercises.
- CISA Cross-Sector Cybersecurity Performance GoalsData recovery baseline for small organisations.
- Restore deleted files or folders in OneDriveConsumer restore route used to design the test task.
- NIST Cybersecurity Framework 2.0Recovery outcomes used to frame the self-service review.
Measurement checklist
| Field | What to capture | Owner |
|---|---|---|
| Volume | Weekly request count, backlog age, and repeat issue patterns | Assistant prepares, manager reviews |
| Risk | Access level, customer impact, security sensitivity, and approval needs | Technical owner |
| Cadence | Daily, weekly, monthly, or quarterly review rhythm | Manager |
| Evidence | Sample tickets, logs, screenshots, and before-after examples | Assistant collects, owner validates |
| Escalation | Triggers, approval path, response time, and stop-work rules | Technical owner |
How to read the result
A good research page should leave the manager with a working number and a clear boundary: what the assistant can do every week, what the assistant can prepare for review, and what must never move without the accountable technical owner.
Source and refresh note
This planning page is dated for 2026 and should be refreshed quarterly as tool stacks, ticket patterns, and security expectations change.
How should teams use this benchmark?
Use it to define task volume, access limits, review cadence, and escalation rules before assigning work.
Get free benchmark review