September 8, 2026 · 7 min read
TL;DR: Surveys and questionnaires are keyed 6 times and offered as distractors 20 times in this bank. The keyed cases share one shape: many respondents, spread out, where speed of collection is what you need. Almost every distractor puts a survey where a conversation belongs, and several put one where its real cost is that it exposes an individual or delays a decision.
Surveys look diligent. They read as evidence-gathering, which is why the exam uses them so heavily as the responsible-sounding wrong answer.
Techniques that do not scale.
The bank's clearest keyed case is procurement. A project manager needs to gather data quickly from dozens of geographically scattered candidate vendors, and the question asks for the best method. Keyed is questionnaires and surveys, and the explanation names the property that decides it: they are designed to collect information rapidly from many respondents spread across different locations, while workshops and live interviews do not scale as quickly to a large, dispersed group.
Every element of that stem is doing work. Dozens of respondents rules out interviews. Geographically scattered rules out workshops. Quickly rules out anything sequential. Change any one of those and the answer moves.
The second keyed use is about whose judgement counts. An agile innovation team keeps turning out novelties with very little business value in them. Keyed, alongside tightening the value criteria with the product owners, is surveying end users and ordering the product backlog by the value the survey reveals. The explanation gives the logic: the team is building things nobody values, so letting real users rank the backlog forces the judgement to be made before the work starts. Here the survey is not gathering requirements, it is gathering a ranking from the people whose opinion of value is the one that matters.
The bank also confirms the technique's home. Asked what a project manager fleshing out requirements would not be engaged in, keyed is inspection, and the explanation places questionnaires and surveys alongside facilitation and prototypes as tools for collecting requirements, with inspection belonging to validating scope.
Wherever the scenario contains a person you could simply talk to.
One person is underperforming. On a retail loyalty programme, every team member is meeting their agreed indicators except one, who has now missed for three consecutive reporting periods. Surveying the whole team to find out what is causing the underperformance is offered. Keyed is discussing the shortfall privately with the team member. The explanation names two separate failures in the distractor: a team-wide survey investigates a problem the team does not have, and it quietly exposes the individual to their colleagues in the process.
You need to map resistance, not collect opinions. A project manager a few weeks into a high-profile initiative picks up that not everyone welcomes it. Distributing surveys to collect feedback is offered. Keyed is the stakeholder engagement assessment matrix. The explanation draws the distinction precisely: a survey collects opinions but does not map them to a desired position, whereas the matrix plots each stakeholder as unaware, resistant, neutral, supportive or leading, so engagement effort can be aimed at the gaps.
Two experts disagree and both are partly right. A landscape architect and an ecology consultant dispute plantings for a riverside greenway, each on ecological grounds. Surveying the local community to determine their preference is offered. Keyed is a collaborative session between the two parties. A survey here would substitute a popularity count for a technical reconciliation that the stem says has to satisfy both goals.
The requirements need depth, not breadth. A customer says earlier projects never delivered what they actually needed. Keyed is running a focus group with the customer's experts and brainstorming requirements with them, because the failure was in how elicitation was done. A survey would collect more of the same shallow input that produced the original miss.
You need historical schedule data. A project manager wants to build a schedule from similar past ones. Surveys and questionnaires are offered and lose to benchmarking, which compares current data against historical data. Surveys ask people what they think; they do not surface comparable schedule records.
Common trap: using a survey when the real requirement is agreement. Two records in this bank make that failure explicit. In the vendor-claim scenario, a provider insists its security software is plug-and-play and needs no validation, and surveying department heads to collect initial impressions and concerns is offered. It loses to convening a briefing with key stakeholders to consolidate standards and set unified expectations, because security, compliance, IT operations and the affected departments each hold requirements the solution must satisfy, and consolidating them produces a single agreed position. A survey would produce a list of separate opinions and nothing that binds anyone. The bank flags this shape by name rather than just marking the option wrong, and it is the most reliable tell in the slice: if what the scenario needs at the end is one agreed answer, a survey has not produced it.
| The stem describes | Keyed | Why not a survey |
|---|---|---|
| Dozens of scattered vendors, data needed fast | Questionnaires and surveys | This is the case surveys are for |
| Agile team building low-value features | Survey end users, order backlog by value | Also keyed; the ranking must come from users |
| One team member missing targets | Private conversation | Investigates a problem the team does not have |
| Stakeholder resistance, positions unknown | Engagement assessment matrix | Collects opinions without mapping them |
| Two experts disagreeing on technical grounds | Collaborative working session | Replaces reconciliation with a vote |
| Past projects delivered the wrong requirements | Focus group with the customer's experts | Repeats the shallow input that caused the miss |
| Vendor claims no validation is needed | Stakeholder briefing to consolidate standards | Produces opinions, not an agreed position |
Watch for the tally sheet.
Beta testers of a mobile banking app report crashes one to three times a day, and the project manager asks them to complete a tally form each time it happens. Survey is offered. Keyed is a check sheet, and the explanation defines it usefully: a form for recording occurrences as they happen, with categories set out in advance, so what accumulates is frequency data, how often, under what conditions, at what time of day. It then gives the reason for asking at all, which is the sentence worth keeping: "one to three times a day" is an impression, whereas a tally produces data.
That is a good summary of the whole slice. A survey collects impressions at scale. When the scenario needs something other than impressions, something else is keyed.
When is a survey the keyed answer on the PMP exam? When you need information from a large number of respondents who are geographically scattered, and speed of collection matters more than depth. The bank keys questionnaires and surveys for gathering data quickly from dozens of scattered candidate vendors, and for ordering a product backlog by the value end users report.
Why do surveys lose so often on the exam? Because most scenarios describe a small, identifiable group or a single person, where a conversation gets better information faster. A survey also loses whenever the underlying need is to reach agreement or make a decision, since collecting opinions is not the same as resolving a disagreement.
Is a survey ever the right response to one person underperforming? No. The bank keys a private conversation. Surveying the whole team investigates a problem the team does not have, and it quietly exposes the individual to their colleagues in the process.
What is the difference between a survey and a focus group? A survey collects many shallow responses in parallel, without interaction between respondents. A focus group is a facilitated conversation with a selected group, which produces depth and lets people build on each other's answers. The bank keys focus groups when the problem is that previous requirements missed what the customer actually needed.
PMP Practice's 2,141 questions are certified against PMBOK 8 and the July 2026 ECO, with every wrong answer explained rather than just marked wrong. Start the free 20-question sample — no card, no signup required to try it.
Related reading: Requirements Elicitation: Which Technique the Scenario Actually Wants and The Private Conversation: 21 Keyed, 19 Distractors. Communication tasks are mapped on the People domain study page.
When is a survey the keyed answer on the PMP exam?
When you need information from a large number of respondents who are geographically scattered, and speed of collection matters more than depth. The bank keys questionnaires and surveys for gathering data quickly from dozens of scattered candidate vendors, and for ordering a product backlog by the value end users report.
Why do surveys lose so often on the exam?
Because most scenarios describe a small, identifiable group or a single person, where a conversation gets better information faster. A survey also loses whenever the underlying need is to reach agreement or make a decision, since collecting opinions is not the same as resolving a disagreement.
Is a survey ever the right response to one person underperforming?
No. The bank keys a private conversation. Surveying the whole team investigates a problem the team does not have, and it quietly exposes the individual to their colleagues in the process.
What is the difference between a survey and a focus group?
A survey collects many shallow responses in parallel, without interaction between respondents. A focus group is a facilitated conversation with a selected group, which produces depth and lets people build on each other's answers. The bank keys focus groups when the problem is that previous requirements missed what the customer actually needed.