On this page7 sections
Growth Operations ResearchA bounded record review, not an industry benchmark.

Measuring the Gap Between Forecast Commit and Actual Close

A bounded measurement design for comparing monthly forecast commitments with actual closed revenue.

Published ••3 min read
Measuring the Gap Between Forecast Commit and Actual Close research illustration

Published November 6, 2026.

Question and scope

How far does the monthly committed forecast differ from the revenue that actually closes, and where does the difference come from? The scope is the forecast record in one growth team during a defined period, not an industry benchmark. The measure matters because a repeated gap can hide a stage-definition problem. This note does not judge whether a representative was optimistic or whether a deal should have closed. The narrow purpose is to describe the difference and the deals that explain it.

Population and unit

Start with a dated export of monthly commits and the deals that supported them. Record the commit date, amount, owner, stage, expected close date, and the final outcome. Include deals that closed, slipped, or were removed. The unit is one deal in one monthly commit. A deal that appears in three monthly commits creates three units, because each commit made a different claim. Keep the original commit version rather than replacing it with the final state. Exclude deals that were never included in a commit and deals that were cancelled before the forecast window. Record the exclusions so the denominator can be reproduced.

Definitions and measures

A commit begins when the deal is included in the monthly forecast with an expected close date. A close is the recorded revenue event, not a verbal promise. A slip is a deal that moves to a later period without closing. Compute the gap as committed amount minus closed amount for each period. Report the absolute gap, the percentage of commit, and the number of deals that slipped or were removed. A single percentage without the deal list is not enough to explain the result. Split the gap by stage at commit, owner, and source. The split shows whether the miss is concentrated in one stage or spread across the pipeline. Keep the categories stable across periods.

Evidence and data limits

Each deal needs the commit reference, stage, amount, expected close date, last customer contact, and outcome. If the close date was changed after the commit, record both dates and the reason. A later date can be legitimate, but it should be visible. Preserve the original commit export. Do not rebuild the commit from the final closed list, because that removes the deals that missed. If the system cannot reproduce the commit, note the limitation and use the archived export. Keep customer and pricing details in the approved system. The analysis file should use a deal reference rather than copying commercial terms.

Interpretation limits

A small gap does not prove that the forecast process is strong. The team may be committing only deals that are already closed. Pair the gap measure with a review of how many deals were in the commit and how early they were added. A large gap does not prove that the team was careless. A few large deals can move the number, and a single customer delay can change the month. Report the concentration and the reasons rather than a single judgment. Avoid ranking representatives by gap. Deal mix, deal size, and customer timing all affect the outcome. The useful question is whether the stage definitions and the commit criteria are clear.

Review cadence

Run the review after each monthly close, then summarize the pattern quarterly. End with one change to the commit criteria, one owner, and one recheck date. Archive the commit exports, the deal list, and the notes so the measurement can be repeated. Keep the observation window beside the finding so the next review compares like with like.

Sources

Sources

  1. https://www.ftc.gov/business-guidance/advertising-marketing
  2. https://www.nist.gov/cyberframework
  3. https://www.sba.gov/business-guide/grow-your-business
  4. https://www.dol.gov/
  5. https://www.census.gov/