Open Question, Listen-For, and Parry
Ask openly, avoid cues and score only the predictive pattern you defined.
- Difficulty
- Moderate
- Time to result
- ~weeks to results
- Steps
- 6
- Confidence
- 98%
The method combines openness with disciplined scoring. First define the trait or repeated behaviour you want to predict. Test many candidate questions against a study group of measurably strong performers and a contrast group of average performers. Retain a question only when the strong group produces a recurring response pattern that meaningfully differs from the contrast group; that pattern becomes the listen-for. During the real interview, ask the question without a yes/no frame, then remain silent. If the interviewee asks what you mean, use a neutral parry that returns interpretation to them rather than narrowing the question. Score the validated pattern when it appears spontaneously and treat other responses as non-predictive for this trait, not as universally wrong. This preserves the richness of an open answer while avoiding improvised, biased interpretation.
Origin
Buckingham explains the method from his Gallup psychometrics work during The Diary of a CEO.
Core principles
- 01Open questions reveal top-of-mind patterns without prescribing an answer.
- 02Repeated past behaviour is more useful than a single past event.
- 03A predictive listen-for must be validated against contrasting groups.
- 04Clarifying for the candidate can accidentally cue the desired answer.
- 05Non-predictive answers are not necessarily bad answers.
How to run it
- 1
Define the target
Specify the trait, behaviour or outcome the interview should help predict. Decide how strong and average performance will be distinguished independently of the answers.
Watch out Do not label a question predictive before defining an external performance criterion.
- 2
Generate candidate questions
Create a broad set of open questions that could surface the target pattern. Avoid wording that advertises the preferred response.
Pro tip Ask about concrete experience, such as what someone enjoyed most in previous work.
- 3
Run a contrast study
Ask the same questions of measurably strong and average performers. Look for response patterns shared by the strong group and uncommon in the contrast group.
Pro tip Discard attractive questions that do not separate the groups.
Watch out A pattern for one role may reverse or disappear in another role.
- 4
Define the listen-for
Write the precise idea that counts as the predictive response, allowing different wording that expresses the same meaning. Specify how it will be coded.
Watch out Do not broaden the listen-for after hearing a candidate you happen to like.
- 5
Prepare a neutral parry
Choose a phrase that declines to clarify the question’s intended scope. Buckingham’s example returns the question by saying he is interested in what the interviewee means.
Pro tip Practise the phrase until it feels natural enough not to introduce extra cues.
- 6
Ask, wait, and code
Ask the open question, stay quiet and let the person interpret it. Code the listen-for only if it appears without prompting.
Pro tip Monitor your own share of the conversation so the interviewee does most of the talking.
Watch out Do not confuse a non-predictive answer with a generally poor answer.
In the wild
Buckingham describes asking, ‘How do you know if you’re doing a good job of listening?’ In his account, highly empathetic respondents repeatedly indicated that the other person keeps talking. That idea became the listen-for; repeating words, nodding or mirroring were treated as non-predictive for that particular test rather than automatically bad.
→ An open question can be coded without telling the interviewee the preferred answer.
Buckingham says successful salespeople and successful teachers reacted differently when asked how they feel when someone doubts them. He presents this as evidence that a useful listen-for is role-specific rather than a universal personality rule.
→ The same question is validated separately for each role instead of transferred blindly.
Common mistakes
Cueing through clarification
Narrowing the setting or offering examples helps the interviewee reverse-engineer the desired answer. Use a neutral parry instead.
Talking through the evidence
The interviewer cannot observe a spontaneous, top-of-mind pattern while dominating the conversation. Ask and wait.
Transferring a listen-for across roles
A response associated with success in sales may not predict success in teaching or nursing. Validate within the target role.
Is it for you?
Best for
Hiring teams or researchers who can validate questions against known performance groups before using them for decisions.
Not ideal for
Casual conversations where no validated outcome, comparison group or scoring purpose exists.
From the transcript
“you should be asking open-ended questions”
“well i know what i mean by that but i'm interested in what you mean by that”
“repeated past behavior is the best predictor of repeated future behavior”
From the episode
How To Find Ultimate Fulfilment At Work: Marcus Buckingham