Prepare facts and open questions for a performance conversation

Author: AILesson8 min setupTested with:ChatGPTReviewed: 2026-08-28

Quick answer

Organize observable work evidence, expectations, context, and fair questions without inferring motives or predetermining outcomes. Provide: Observed evidence, Role expectations and context, Conversation purpose and boundaries. Expected result: A conversation brief with evidence table, gaps, neutral opening, sequenced questions, support options, and follow-up record.

Add your context

Your text stays in this browser. AILesson Prompts does not send it to a model or server.

Observed evidenceRole expectations and contextConversation purpose and boundaries

Your prompt

Unfilled fields remain visible as placeholders, so you can still copy and edit the prompt

Prepare a fair performance-conversation brief from the supplied record.

Observed evidence:
[evidence]

Expectations and context:
[expectations]

Purpose and boundaries:
[conversation]

Separate direct observations, calculations, employee-provided explanations, manager interpretations, disputed items, and unknowns. Test whether each example is specific, relevant to a documented expectation, within the review period, and comparable. Include positive and mixed evidence. Do not infer attitude, intent, personality, health, family status, protected characteristics, or a diagnosis. Do not recommend discipline, ratings, legal conclusions, or accommodations beyond supplied authority. Create an evidence table, missing-context checklist, neutral opening, open questions that invite the employee's account before solutions, questions about obstacles and support, specific expectations that may be discussed, and a follow-up record. Flag procedural or fairness risks and any decision that must wait for HR or additional evidence.

Private by defaultPrompt assembly happens locally in your browser. Avoid placing confidential information into any AI service unless your organization allows it.

From input to outcome

A worked example

See how concrete context turns this recipe into a usable result

Actual input

Observed evidence: Review period 1 May–31 July. Analyst Noor delivered 17 of 20 scheduled weekly reports by the agreed Tuesday 12:00 cutoff; three arrived Wednesday at 10:15, Wednesday at 16:40, and Thursday at 09:20. Source: report archive timestamps. Two of the late reports depended on Finance extracts delivered after Tuesday noon; one had no recorded dependency. In June, Noor found and corrected a duplicate-revenue error before distribution, documented the check, and received written praise from the Sales lead. Four of 20 reports required later formatting corrections; team median was two, but assignments differed. On 18 July Noor wrote that ad-hoc requests were displacing scheduled work and asked for priority guidance; manager did not reply in the record. Stakeholder survey: 6 responses, five rated accuracy good, two mentioned timing; sample is small. Noor disputes that all three delays were under personal control.
Role expectations and context: Role guide: publish the weekly report by Tuesday 12:00, validate totals, document exceptions, and alert the manager as soon as a dependency threatens the deadline. Accuracy has priority over formatting. May–July priorities included weekly reporting plus migration support; no allocation percentage was documented. Finance extract service time is not documented. Team median is not an approved performance standard. Noor received informal feedback in June to use the formatting checklist. No formal rating decision has been made. Workload by week and whether the manager verbally answered the 18 July request are unknown.
Conversation purpose and boundaries: Purpose: understand the record, hear Noor's context, agree near-term working practices and support; not a disciplinary meeting and not a final rating. Participants: Noor and manager Eva. Eva may agree workflow changes but HR owns formal process questions. Available support: report checklist, earlier draft review, priority review, and dependency escalation. Do not ask about health or family circumstances. Keep personal information limited to need-to-know notes. Meet for 35 minutes; send a factual recap for Noor to correct within two working days.

Example output

Performance-conversation brief — May to July

Purpose: understand the evidence and Noor's account, then agree practical support and working expectations. This is not a disciplinary meeting or final rating.

Evidence table

Timeliness | archive timestamps | 17/20 by Tuesday noon; three late | two late reports followed Finance extracts arriving after the cutoff; control of the third is unknown | relevant to documented deadline.
Quality | report record | Noor caught a duplicate-revenue error before distribution; 5/6 survey respondents rated accuracy good | small survey | positive evidence aligned with the priority on accuracy.
Formatting | correction history | 4/20 needed correction | assignments differ; team median is not a standard | discuss checklist use, not relative ranking.
Escalation and priorities | 18 July message | Noor asked for guidance about ad-hoc work | no recorded manager reply; possible verbal reply unknown | shared context requiring clarification.

Missing context: workload by week; Finance service expectation; dates of dependency alerts; cause and workload for the third late report; whether priority guidance was given verbally; which formatting errors occurred after June feedback. These gaps prevent a complete conclusion.

Opening: “I want to review the May–July reporting record, hear what you saw, and agree how we protect accuracy and deadlines. I have not made a final rating decision. Some delays involved late Finance inputs, and I also want to recognize the revenue-error catch.”

Questions, in order: How do you read the 20-report record? What happened in each of the three late weeks? When did you know the deadline was at risk, and what escalation was possible? How did migration and ad-hoc work affect priorities? What response, if any, did you receive after 18 July? Which parts were within your control and which were not? What helped accuracy? What made the formatting checklist hard or easy to use? Which support—priority review, earlier draft check, dependency escalation, or another work adjustment—would best help?

Potential agreement, not a predetermined outcome: preserve Tuesday noon as the documented expectation; define an early warning trigger for late inputs; have Eva resolve priority conflicts; use the checklist before release. Owners, trigger times, and review date must be agreed in the meeting. Record facts, Noor's corrections, agreed actions, support, and unresolved items; send the recap within two working days. Consult HR before any formal-process decision.

Why this works

1
Evidence labels stop interpretations and disputed claims from being presented as settled facts
2
Hearing the employee's account before solutions makes missing context reviewable

Check the result

Is every concern tied to dated, role-relevant evidence and a documented expectation?
Does the brief include positive, disputed, missing, and dependency-related context?
Do questions invite explanation without implying motive, diagnosis, or a decided outcome?

Use it with confidence

Frequently asked questions

Practical answers about when to use this recipe, what to provide, and where human review still matters

What should I prepare before using “Prepare facts and open questions for a performance conversation”?

For “Prepare facts and open questions for a performance conversation,” prepare Observed evidence, Role expectations and context, and Conversation purpose and boundaries. Replace placeholders only with information you can verify. If a detail is unknown, preserve that uncertainty explicitly instead of asking the model to infer it.

When is the “Prepare facts and open questions for a performance conversation” result not ready to use?

The result is not ready if it does not yet deliver the stated outcome—A conversation brief with evidence table, gaps, neutral opening, sequenced questions, support options, and follow-up record—from the supplied evidence, or if it relies on unresolved assumptions, missing approvals, or invented details. Use the checks as release gates: revise the source inputs or assign a named, authorized reviewer instead of polishing an unsupported output.

Which AI tools have recorded tests for “Prepare facts and open questions for a performance conversation”?

The published test record for “Prepare facts and open questions for a performance conversation” lists ChatGPT as of 2026-08-28. This confirms recorded runs, not guaranteed compatibility or identical results in later product versions. For another tool or version, keep every constraint visible and repeat the result checks before use.

More ways to explore