Question and scope
This protocol measures the share of accepted cross-team requests that arrive without a usable effort estimate. An effort estimate is a stated range of work, expressed in hours, days, or a comparable unit, that the receiving team can use to decide whether the request fits its capacity. A request without one forces the receiver to guess before accepting, and that guess becomes the plan. The result is a percentage, a breakdown by request type, and a count of estimates that exist but are too vague to schedule.
The question matters because intake reviews often check for an owner, a due date, and a description, then declare the request complete. None of those fields tells the receiver how much work the request implies. A clear deliverable with no size estimate can still consume a week that was reserved for something else. Measuring estimate presence moves the capacity conversation earlier, to the point where the receiver can still decline or renegotiate.
Scope is one coordination team and the requests it accepted during a fixed window. Include requests from any channel, because informal requests are often the least estimated at all. Exclude ideas that were discussed but never accepted, and exclude requests rejected before an owner was assigned. Record the window, source systems, and reviewer names so the count can be rebuilt.
Why estimates disappear from intake
Estimates are hard to produce when the requester does not know the receiving team's work. A marketing lead asking for a data pull may have no basis for guessing how long the query takes. Rather than appear uninformed, the requester leaves the field blank and lets the receiver fill it in. The omission is rational at the individual level and expensive at the system level, because the receiver now owns a estimate it did not choose.
A second reason is that intake forms reward speed. A short form with required title, owner, and date gets used, while a form that asks for a sizing range gets abandoned or filled with a placeholder. Teams then interpret the placeholder as an estimate even though it carries no information. The protocol has to distinguish a number that was chosen from a number that was typed to satisfy a required field.
A third reason is that some work genuinely resists early sizing. Discovery, incident response, and legal review can vary widely depending on what the first hour reveals. The measure should not punish an honest range or a stated unknown. It should distinguish a range with a named assumption from a blank field, because the assumption is what makes the range useful to a capacity decision.
Defining the coding unit
The unit is one accepted request with one owner and one expected deliverable. Split the request when each part could be estimated, scheduled, and delivered independently. Keep it as one unit when the parts share a single approval or must ship together. Write the split rule into the codebook before sampling, since bundled requests are the most common source of disagreement.
Use the version of the request that existed when the owner accepted it. Later estimates produced during delivery are useful, but counting them measures how quickly the team learned rather than how well it scoped the request. If the acceptance moment is unclear, use the first status change that names an owner. Record the source of that timestamp so another reviewer can reproduce the choice.
What counts as a usable estimate
A usable estimate states a unit and a range, such as three to five business days for a report refresh. It also names the assumption that drives the range, such as the number of source files or the need for legal review. A single number with no range can be usable when the work is genuinely routine, but the protocol should flag it for a second look. The test is whether a receiver could compare the estimate with available capacity without asking a follow-up question.
An estimate in the wrong unit is not usable for capacity planning. A statement that a request is "high effort" or "a big lift" carries no schedulable quantity, even if it conveys the requester's anxiety. Similarly, a story-point value that the receiving team does not use for planning cannot be compared with hours. Code these as present but vague rather than absent, because the request did try to size the work.
A stated unknown can be usable when it includes a discovery step and a decision point. For example, a request may say the work is unknown until a two-hour sample is reviewed, with the estimate to be set after that review. Code the discovery step as the estimate and note the follow-up date. That pattern is honest and schedulable, and it should not be scored the same as a blank field.
Sampling requests
Use a census when the team accepted fewer than roughly one hundred requests in the window. A census avoids sampling error and is practical for a single coordinator. Above that threshold, draw a random sample and publish the method. Stratify by receiving team and request type so a large volume of small, routine requests does not hide a small set of complex ones.
Freeze the sample before reading the request bodies. Reading first and selecting later invites the reviewer to choose the most interesting cases. Store the random seed or the selection procedure with the sample file, and record any substitutions. If a selected request is out of scope, replace it with the next item on the randomized list rather than a hand-picked similar case.
Reading the artifacts
Read the request in a fixed order: the original ask, the acceptance comment, the linked specification, and any sizing note in an attachment. Stop once the estimate and its assumption are clear. Continue to later channels only when the estimate is still unresolved, and record which channel supplied the answer. That record shows where estimates usually live and where they tend to hide.
Quote the smallest phrase that supports each score, because a quote anchors the judgment and makes a second review possible. When the evidence is a link, record the link and the section heading rather than a summary. Do not paraphrase a vague statement into a specific range, since that would credit the request with precision it did not have. Keep the quote short enough to fit in the coding row.
Handling partial and disputed cases
Some requests include an estimate from the requester and a different one from the receiving team. Code both, and use the receiving team's acceptance estimate for the headline because that is the number the schedule was built on. Note the requester's figure separately so the gap between them remains visible. A large gap is a finding about intake quality, not a reason to average the two.
When two reviewers disagree, return to the quoted evidence rather than debating the request's intent. If the quote supports both readings, the rubric is ambiguous and should be tightened with a new example. Recode the disputed case and every similar case after the revision, then record the change in a codebook log. The log should name the rule that changed and the date, not just the case.
Exclude a request only when the record is genuinely unavailable, such as a deleted ticket with no export. Mark it unverifiable and report the count separately. Do not guess an estimate from memory or from the final effort. A visible gap in the records is itself a finding about intake hygiene, and it should not be hidden inside a completeness percentage.
Calculating the share
Report two numbers rather than one. The first is the share of requests with an estimate that names a unit, a range, and an assumption. The second is the share with any estimate at all, including vague or unitless statements. The gap between the numbers shows how many requests look estimated at a glance but cannot be used for scheduling. Include raw counts beside both percentages.
Break the result down by request type and receiving team. A team that receives many small, repeatable requests may score well without a strong intake process, while a team that receives discovery work may score poorly even with good documentation. Suppress groups with fewer than five requests so one unusual case does not look like a pattern. Pair each group result with the type mix that produced it.
Report the most common missing element separately: missing unit, missing range, missing assumption, or no estimate at all. Each points to a different fix. A missing unit may need a form change, while a missing assumption may need an example of a good sizing note. Pair each finding with the specific intake change it suggests, and avoid a general instruction to estimate more carefully.
Reporting without overclaiming
Present the first result as a baseline before presenting any target. A first measurement tells the team where it starts, not how far it has to go. Show the distribution across request types so readers can see whether the problem is concentrated or spread out. A single percentage hides the difference between one team missing estimates everywhere and another missing them occasionally.
Keep the reporting period consistent. A window that includes a holiday, a system migration, or a leadership change may not be comparable to a normal month. Note those events next to the result so the coordinator can explain a movement without inventing a cause. Write the note in the same row as the result so the context travels with the number.
Do not turn the measure into a judgment about individual requesters. The useful question is which intake fields and examples make estimates easier to produce. Share the coding rubric with requesters so they can see what a usable estimate looks like. The request acceptance checklist is a natural place to add the sizing field, and the two-week lookahead is where the estimate meets available capacity.
Limits and sources of error
The sources cited here describe cost estimating, scheduling, and project scope in general terms; they do not provide a benchmark for estimate presence. The protocol therefore produces a local baseline, not an industry comparison. A team should not present its percentage as evidence that it is better or worse than another organization. Readers should treat the result as a starting point for a local intake conversation.
Estimate quality is not the same as estimate presence. A request can carry a confident range that is badly wrong, and this protocol would score it as usable. That is intentional, because presence is cheap to measure and quality requires outcome data the team may not have. Report the limitation plainly and pair the measure with a later estimate-versus-actual review if the team wants to judge accuracy.
Reviewer judgment remains a source of error. Two careful reviewers can disagree about whether "about a week" names a usable range or a vague guess. The quote requirement and the second-review sample reduce that risk but do not remove it. Report inter-rater agreement so readers can judge how much confidence the number deserves. A single reviewer should not be the only check on a published result.
Review cadence
Review the measure after two full cycles before setting a target. One cycle is a baseline, and a target set from a single period is likely to reflect that period rather than the process. At the end of the review, choose one intake change, one owner, and one check date. A practical first change is to make the sizing unit and assumption required fields that cannot save empty.
Retire the measure when most accepted requests carry a usable estimate and the schedule no longer needs emergency renegotiation. Until then, report usability beside vagueness so a filled-in form is not mistaken for a sized request. Link the result to the pre-handoff capacity check so intake sizing and capacity planning use the same unit. The purpose is to make the first capacity decision honest and cheap to reshape.
Sources
- U.S. Government Accountability Office, GAO Cost Estimating and Assessment Guide, reviewed November 19, 2026.
- U.S. Government Accountability Office, GAO Schedule Assessment Guide, reviewed November 19, 2026.
- Project Management Institute, PMBOK Guide, reviewed November 19, 2026.
- U.S. Bureau of Labor Statistics, Occupational Outlook Handbook: Project Management Specialists, reviewed November 19, 2026.
