The quiet version of a missed deadline rarely starts with a speech. It starts with a sentence that sounds helpful. "While you're in there, can you also…?" The date on the wall stays put. The work behind the date gains a room. Weeks later the apology email talks about effort. The missing object is a decision that the new room was allowed, and that the date moved with it.
The definition is shorter than the argument
The Project Management Institute's Lexicon of Project Management Terms, version 5.0, updated January 2026, defines scope creep in one sentence: the uncontrolled expansion to product or project scope without adjustments to time, cost, and resources. Project scope, in the same lexicon, is the work performed to deliver a product, service, or result with the specified features and functions. Product scope is the features. Project scope is the work. Creep can hit either one. The word "uncontrolled" is doing the legal-style work. A change that is reviewed, priced, and given a new date is not creep. It is a change. Creep is the expansion that never goes through that door, so the old date and the old budget keep pretending they still describe the job.
Richard Larson and Elizabeth Larson made the same cut at the PMI Global Congress in Orlando in October 2009. Their working definition was adding features, functions, requirements, or work that is not authorized, beyond the agreed scope. They quote the fourth edition of the PMBOK Guide, which described scope creep as adding features and functionality without addressing the effects on time, costs, and resources, or without customer approval. They are explicit that change itself is ordinary. If the expansion is approved, they say, it is not scope creep.
That distinction is the whole deadline problem. Teams argue about whether the new request is "small" or "valuable." The lexicon does not ask whether it is charming. It asks whether time, cost, or resources moved with it. If they did not, the due date is now a fiction wearing the old number.
Five ways the date stays frozen
Larson and Larson list five causes from their practice and from a case narrative they walked through at that congress. Treat the narrative as their teaching story, not as a news report you can audit from this page. The causes themselves are the part worth stealing:
- Ambiguous scope. A sponsor has been thinking about the business need for months and assumes the team shares the picture. A high-level charter without a decomposed breakdown leaves a gap. People fill the gap by building what they guess is meant, or by collecting wishes directly from anyone nearby.
- No formal handling of new requests. Sponsors often do not want every small call. The team then decides. Larson and Larson call out gold-plating: someone adds a feature because it seems useful or interesting, without authority to add it. They also name the opposite failure, "scope kill," where a rigid process rejects a change that should have been taken to the sponsor.
- Inconsistent requirements collection. If one stakeholder can reach a builder directly, the backlog grows in side conversations. The Larsons cite colleagues who described unmanaged contact between client and team as a creep path.
- A sponsor who leaves after the signature. Approval of a plan is not the same as staying available for the next tradeoff. When the sponsor vanishes, the date they insisted on becomes a wall, and every new fact has to crash into it.
- A long timeline. Length does not cause creep by magic. It multiplies the number of afternoons on which a small yes can happen. A short project can creep too. It just has fewer afternoons.
Their recommendation is not "say no forever." It is a change path with a named decision maker, a scope statement that lists what is out as well as what is in, and a willingness to move the date or cut something else when the date is fixed. If the date cannot move and the new work is real, something already promised has to leave. Pretending both can stay is how the apology email gets drafted.
A change line you can say out loud
You do not need a PMO to use the lexicon. You need one sentence that separates a clarification from an addition. Clarification: "Does the table include last year's figures, or only this year's?" Addition: "Can the table also include the three regions we did not mention?" The first question belongs in the work you already estimated. The second question belongs in a decision.
Write the decision in a place the date can see. A shared note is enough. Columns that matter: what was asked, who asked, whether it is in the current promise, and what happens to the date if yes. If you cannot fill the last column, the answer is not yes yet. It is "not in this delivery."
Here is an illustrative example, invented for the shape of the choice. A freelancer agreed to a four-page brief by Friday. On Wednesday the client asks for a one-page appendix of competitor prices. The appendix is useful. It is also not the four pages. An honest reply names the trade: the appendix on Friday and the brief on Monday, or the brief on Friday and the appendix next week, or a shorter brief that makes room. A quiet reply says "sure" and discovers at 11 p.m. that prices are not in the source folder. The deadline did not kill the work. The unpriced yes did.
Larson and Larson warn that tiny requests are exactly how this happens. The request looks smaller than a meeting. The lexicon still counts it, because the adjustment to time never occurred. Size is not the test. Adjustment is the test.
Creep versus learning
Software and research work discover requirements. Larson and Larson say sponsors often experience that discovery as a broken contract, while the team experiences it as finding out what the job actually is. Both descriptions can be true. Discovery becomes creep when the new fact is absorbed into the old promise without a cut or a date change. Discovery stays honest when someone with authority picks which fact the date still covers.
If you are the person with authority, do the unfashionable thing and decide in writing. If you are not, your job is to make the unpriced yes visible before you start the appendix. Starting is the uncontrolled part. The lexicon's word is expansion without adjustment. Typing the appendix is the expansion. Silence about Friday is the missing adjustment.
Related pages on this site, if the date is already lying for other reasons: buffer days are not laziness covers reserved time that was never meant to be free work, and a dashboard that can show red covers how to see the slip before the apology. Scope is the specific case where the work grew and the date did not.
Boundaries of this advice
The PMI lexicon is a glossary, not a statute, and it does not settle a contract dispute. If money, employment, or a signed statement of work is in play, read the document or ask someone qualified to read it. Larson and Larson's five causes come from a conference paper and a teaching case, not from a controlled trial with a success rate you can quote. "Most projects seem to suffer from scope creep," their wording, is their impression, not a census. Do not turn it into a percentage.
You can still use the test this week. Look at one thing you agreed to add since the date was set. Did time, cost, or people move with it? If the answer is no, the deadline is carrying weight it was not given.
Documents behind the definition
- Project Management Institute. (2026). PMI Lexicon of Project Management Terms (Version 5.0). Definitions of scope creep and project scope. PMI lexicon PDF
- Larson, R., & Larson, E. (2009, October 13). Top five causes of scope creep … and what to do about them. Paper presented at PMI Global Congress 2009—North America, Orlando, FL. pmi.org library