Small-business outsourcing research
What share of small-business calendar requests arrive with missing details?
A protocol for measuring how often scheduling requests lack the date, the duration, the attendee, or the purpose needed to book them safely.

Research question and scope
This protocol asks what share of calendar requests arrive without the details a small-business assistant needs to book them. The unit is one request for a new meeting, a change, or a cancellation, identified by the message or form that started it. The study measures the missing fields, the time to complete them, and the number of messages spent on clarification. It does not judge whether the meeting was necessary or whether the requested time was convenient.
Set the scope to one business, one scheduling channel, and one calendar. Record the form, the inbox, and the chat channel through which requests arrive. If the business adds or removes a channel during the window, analyze the cohorts separately. A new form can reduce missing details without any change in the assistant's skill.
Define the population as every request recorded during the window, including requests that were withdrawn or rescheduled. The unit is the request, not the meeting, because one meeting can generate several changes. Test the definition on two recent requests before collection begins. Record the test result so a later reader can see how the boundary was applied.
Definitions that make the measure reproducible
Define a complete request as one that names the date, the start time, the duration, the time zone, the attendees, and the purpose. A request with a date but no duration is not complete, because the assistant cannot safely book it. A request that names a person but not an email address is also incomplete if the invitation needs an address. Write one positive and one boundary example for each field.
Define the request time as the moment the message or form was recorded, not the moment the assistant read it. Define clarified as the point when all required fields are present and the assistant has enough to book. Define booked as the point when the invitation is sent and the calendar shows the event. Define confirmed as the point when the key attendee accepts or the owner approves.
Record the channel, the requester, the meeting type, and whether the request was internal or external. A recurring internal stand-up and a customer kickoff have different evidence needs. A cancellation can be complete with only the event reference, so use a separate field list for it. Do not score a cancellation as incomplete because it lacks a purpose.
Evidence standard and data dictionary
For each request, capture the raw message, the fields present, the fields missing, and the time each missing field arrived. Record the number of messages sent to clarify and the person who supplied the answer. Add whether the assistant booked from the first message or had to wait. Keep the message thread with the record so the classification can be checked.
Record the meeting type, the number of attendees, and whether the request involved a customer. A large external meeting may need a room, a link, and an agenda, while a one-to-one call may need only a link. The required fields should follow the meeting type, not a single universal list. State the required list in the scope so the reader can reproduce the count.
A second reviewer should be able to rebuild the request and the booking from the record. If they need to read every message to find the date, the data dictionary is incomplete. Test the record on three requests before scaling the collection. Record the result of that test.
Sampling and collection
Draw the sample from the scheduling log and the inbox, not from a list of meetings that were booked successfully. Include requests that were abandoned, because they may be the most incomplete. Record the export date, the filters, and the number of records excluded. If the population is large, use a fixed interval across the window.
Collect the request as it existed at the time, not after a later cleanup. If a missing field was added days later, record the addition date and analyze the original delay separately. Do not infer a request from a calendar event that has no matching message. Note the collection date separately from the request date.
Record the assistant's shift and the owner's availability during the request. A request that arrives when nobody can approve a room may wait for reasons unrelated to the requester. Add the office holiday calendar and the time zone difference. These fields help separate a missing-detail problem from a coverage problem.
Analysis and reporting
Report the number of requests, the share missing each field, and the median time to a complete request. Show the counts by channel, meeting type, and internal or external requester. Present the share booked from the first message and the share that needed clarification. Use the same denominator for every rate and show requests with incomplete times.
Inspect the most common missing field and describe the context without naming the requester. A missing duration may point to a form that does not ask for it, while a missing attendee may point to a habit. Compare the result with the channel and the meeting type. Report abandoned requests separately so their timelines do not distort the booked cases.
A useful additional view is the time from the complete request to the sent invitation. A complete request that sits for a day is a different problem from an incomplete one. Show the clarification time and the booking time side by side. That view tells the owner whether to fix the request form or the assistant's queue.
Limits of the evidence
The record cannot show whether the meeting was useful or whether the time suited everyone. It can show only the fields, the messages, and the booking times. A fast booking is not proof of good coordination, and a slow one is not proof of neglect. The protocol also cannot capture a meeting arranged in a hallway conversation that was never logged.
A small business may have too few requests in a month to support a stable share. Describe the counts and the cases instead of ranking a channel. If one requester dominates the sample, report that concentration. Do not present a general rate when one person supplied half the requests.
A request that is complete but ambiguous can still produce a bad booking. The field count does not measure clarity, so add a short note when the assistant had to interpret the purpose. Record the interpretation and whether the requester corrected it. That note is qualitative evidence that the field list alone cannot provide.
Give the result to the owner and the assistant and ask them to choose one observable change. The change might be a required duration field, a default meeting length, a booking link, or a short request template. Record the owner, the effective date, and the measure that will show whether the change worked. Repeat the protocol with the same definitions after the change.
If missing details fell but booking time did not, the queue may have become the bottleneck. If booking time fell but reschedules rose, the first booking may have been too quick. Report the two movements together before declaring a success. Close the review when the owner accepts the finding and the next cohort is scheduled.
Record whether the request arrived through a template, a free-text message, or a booking link. A template can require fields that a free-text message never asks for. Report the share by input method so the owner can see whether the form is doing its job. A low missing rate on a form may simply mean the form is longer.
Ask the assistant how often they guessed a duration or a time zone to keep the booking moving. A guess that is not recorded is invisible in the missing-field count. Record the guess and whether the requester corrected it. That evidence shows whether speed is being bought with silent assumptions.
Ask the last five requesters whether the booking process asked for the right information. A requester who had to repeat a detail is a direct signal. Record the answer and the point in the process where the repetition occurred. Use the finding to simplify the form rather than add another field.
Sources
- U.S. Small Business Administration, manage your business, U.S. Small Business Administration; checked 2026-11-09
- NIST Cybersecurity Framework, National Institute of Standards and Technology; checked 2026-11-09
- Federal Trade Commission, data security guidance, Federal Trade Commission; checked 2026-11-09
- U.S. Government Accountability Office, internal control standards, U.S. Government Accountability Office; checked 2026-11-09
Frequently asked questions
Does a favorable sample prove the routine caused the result?
No. The design is observational and supports a bounded operating decision, not a universal causal claim.
What should an owner review first?
Start with high-consequence defects, missing denominators, exceptions, and cases that required the owner to reconstruct the evidence.
Related research
How long does a new virtual assistant take to clear a first support queue?
A protocol for measuring the time from a new assistant's first login to a cleared support queue, with stage times and the owner's review load.
How often do assistant-run shops miss a supplier delivery-date update?
A protocol for measuring delivery-date changes that never reach the order record, the customer message, or the inventory plan.
How long does a customer onboarding handoff wait for the owner?
A protocol for measuring the delay between a completed sale and the owner-approved onboarding start, with the customer waiting experience.