The assignment went to whoever promised the earliest date. That’s an auction. And you just won it.
A director needs a system migration done. Three project managers are in the room. “When can you have it live?”
The first PM says, “Let me come back with an estimate by Friday.” The second says, “Realistically, end of Q3.” The third says, “I can get it done by end of Q2.”
Guess who runs the project. Now guess who’s in a status meeting in month five explaining why Q2 became Q3.
This is a composite, but you’ve been in this room. Parts 1 and 2 of this series looked at how the winner’s curse selects the most wrong bid and the most optimistic business case. This is the version that happens to you personally.
- Why the Lowest Bid Costs the Most?
- Optimism Bias in Project Management: Why Your Best Business Case Is the Most Dangerous One
Why does the fastest yes win the room?
Because the assignment is an auction, the currency is confidence, and three people just estimated the same unknown. The PM who won is the one whose estimate was furthest below the true duration. Not the best PM. The most wrong one.
The PM who asked for time to estimate looked less committed. The PM who said Q3 looked slower. Nobody rewarded accuracy, because accuracy wasn’t visible yet. At the moment of commitment it never is. All anyone can see is who sounds sure.
What does the curse cost you?
The date you said becomes the baseline. Every fact that surfaces afterwards, the vendor delay, the data quality problem, the stakeholder who wasn’t in the room, gets read against your promise. So each one lands as your failure rather than the plan’s.
You start trading things nobody can see to protect a date everyone can: test coverage, team hours, code quality, the scope that quietly slides to “phase two.” And the reputational math is asymmetric. Nobody remembers that you said yes when they needed a yes. They remember the month you missed.
| Confidence is not an estimating technique. |
How do you push back on an unrealistic deadline without saying no?
- Answer with a range, not a point. “Between April and June, and here’s what moves it toward April.” A range is honest and negotiable. A point is a promise.
- Ask what’s driving the date. Many deadlines are anchors, not constraints. Someone said a quarter once and it stuck. A real constraint, regulatory, contractual, an event, gets planned around. An anchor gets challenged.
- Convert the ask into a trade. Date, scope, resources: pick two. Put the tradeoff on the table so the decision belongs to the sponsor, not to your heroics.
- Commit to the estimate before you commit to the date. “I’ll have a defensible estimate in five working days” is a yes. It’s just a yes to the right thing.
- Bring a three-point estimate when you come back. Optimistic, most likely, pessimistic, and the assumptions behind each. It moves the conversation from confidence to evidence. Part 4 shows how to build one that isn’t cursed itself.
- Write down the conditions. If you’re held to the aggressive date anyway, document the assumptions it depends on. When one breaks, the conversation is about the assumption, not about you.
- Offer the phased version. A credible date for a smaller first release beats an incredible date for everything.
The script
Director: Can you have it live by the end of Q2?
PM: Possibly. Two things would have to be true: the vendor delivers the API on their current schedule, and the legacy data is clean enough to migrate without manual fixes. If both hold, Q2 is realistic. If either slips, we’re into Q3. What’s driving Q2?
Director: The board update is in July.
PM: Then let’s plan a Q2 release of the customer-facing module for the board, with the back-office migration following. I’ll have a full estimate to you by Friday.
That isn’t a no. It’s a yes to something you can actually deliver, with the conditions on record and the sponsor’s real need met.
What if you already said yes?
Then the cheapest moment to fix it is now, not month five. Go back with the range and the conditions this week. Frame it as new information, because it is: “Now that I’ve scoped it, here’s what Q2 depends on.” A sponsor can work with a revised estimate in week two. In month five they can only work with an excuse.
Frequently asked questions
How do you tell a stakeholder a deadline is unrealistic?
Don’t tell them. Show them. Bring the range, the assumptions the earliest date depends on, and the tradeoff between date, scope, and resources. Then ask what’s driving the date. Most stakeholders will trade scope for a date they can trust.
What if my manager won’t accept a range?
Give the point they need with the conditions attached in writing: “End of Q2, provided the vendor API lands on schedule and the data needs no manual cleanup.” That’s still a commitment. But when a condition breaks, the conversation is about the condition.
Isn’t “let me come back with an estimate” a sign of weakness?
It reads that way in the moment, which is exactly why the fastest yes wins. Counter it by naming the day you’ll return and what you’ll bring. A specific commitment to an estimate is stronger than a vague commitment to a date.
| WHERE THE NEW PMP® EXAM TESTS THIS
Estimating and schedule management are Process domain content (41% of the exam), but since July 9, 2026 the exam tests them as negotiation and stakeholder judgment across the People (33%) and Business Environment (26%) domains as well. Expect scenarios where a sponsor imposes a date and the correct answer is a tradeoff conversation, not compliance and not refusal. The exam is testing the same thing the director was: what you do when you’re asked for a number you don’t have yet. |
| This is the kind of judgment call our instructors work through live.
The PMP® Online Class runs four days and 35 contact hours, mapped to the PMP® exam in effect since July 9, 2026 and to the PMBOK® Guide 8th Edition. Both plans include lifetime access to the 35-hour self-paced training. The Pro plan adds a one-hour one-on-one PMP® coaching session after the class ends. |
NEXT IN THE SERIES
Part 4 — Reference Class Forecasting: How to Estimate Without Cursing Yourself
PMI, PMP, PMBOK and PMI-ACP are registered marks of the Project Management Institute, Inc.