Business
How to prepare for a performance review with evidence

Stefan Weiss
To prepare for a performance review, confirm the review criteria, gather examples from the whole period, check what your work actually achieved, and choose the questions you want to discuss. Bring a concise account of your contribution, development needs and proposed next steps, with supporting evidence available when needed.
This guide is for an employee preparing for their own appraisal. It is not a manager’s rating procedure. Follow your organisation’s form, deadlines and review process where these are specified.
Preparation should make the conversation more accurate and useful. It cannot guarantee a particular rating, promotion or pay decision.
1. Confirm what the review will assess
Start with the review period, your job description, agreed goals and any competency or role-level expectations. Find the previous review and notes from interim conversations. Ask for clarification if you do not know the submission deadline or how the meeting will work.
MIT’s performance-review guidance recommends reviewing previous appraisals, job descriptions, self-assessments and notes accumulated during the year. Its own process is an example, not a universal policy for every employer.
Write down what you believe you are being assessed against. For example:
Delivering the agreed project or service responsibilities.
How you collaborated and communicated.
Progress on a development goal from the last review.
Additional responsibilities taken on during the period.
Separate a formal expectation from a personal aspiration. “I would like to manage a team” may be a useful development topic; it does not mean people management was part of your current role.
If priorities changed, gather the conversation or written agreement that explains the change. Do not silently assess yourself against the original goal while your manager is using a later one.
Expected result: a clear scope for your preparation, including any unresolved question about the criteria.
2. Reconstruct the whole review period
Work through the period chronologically rather than starting with your most recent success. Useful material can include project plans, completed work, customer or colleague feedback, delivery records and notes from one-to-ones.
The University of Colorado Boulder’s employee preparation guide encourages reviewing the full cycle and keeping a performance journal to reduce dependence on recent events.
A practical way to start is to scan your calendar by month and note the work that deserves closer inspection. A calendar entry is a reminder of an event, not proof that the intended result happened.
For each possible example, collect the date, relevant goal, your role and where the supporting record lives. Include difficulties and changes of direction as well as positive feedback.
Avoid treating volume of activity as the achievement. Attending ten workshops describes activity. Clarifying a decision that allowed a team to proceed describes a possible contribution, which still needs a specific example and an appropriate account of the result.

If useful evidence is in saved conversations
Recorded one-to-ones or project discussions may help you recover an earlier goal, feedback or a commitment, where those recordings were made and retained appropriately. They are one source alongside the work itself.
In Weeve, chat with saved recordings can help you explore that material. Limit the review to the relevant period and recordings, then ask for an evidence-led account. For example:
Review the selected project conversations from this review period. Identify goals I agreed to, specific feedback about my contribution and unresolved questions. Keep a stated intention separate from a reported result. Cite the source passages and explain what material was reviewed.
This is a suggested prompt, not a tested result. Check the returned sources and coverage before using the account. A saved conversation saying “I will finish the guide” establishes an intention; it does not establish that the guide was finished or used.
The product image shows the actual local chat composer with synthetic example text. It illustrates asking across recorded material, not automatic employee assessment. Local chat runs locally; optional external chat providers have different processing boundaries and may receive relevant text.
Expected result: a broad evidence list covering the period, with gaps visible rather than filled by guesses.

3. Turn activities into accurate achievement examples
Choose a small number of examples that address the review criteria. Each should explain the situation, your contribution, the result you can support and what remains uncertain.
Use four questions when editing an example:
What needed to change? State the problem or agreed objective.
What did I contribute? Name your work without claiming the whole team’s effort.
What happened? Use a verified result or a precisely attributed report.
What proves that account? Keep the source or work record available.
You do not need a numerical metric for every contribution. A reviewed deliverable, an agreed decision or specific feedback can be useful evidence. Do not invent a percentage improvement to make an example sound more impressive.
If a numerical claim is available, check its period, definition and relationship to your work. A team metric moving after a project is not, by itself, proof that your individual work caused the change.
Worked example: a clearer account without invented impact
The following is fictional. Noor is a project coordinator preparing for a six-month review.
Initial note: “Improved onboarding and saved the team lots of time.”
What the records support: Noor reorganised the onboarding instructions, incorporated feedback from two colleagues and obtained approval for the new version. A project conversation records the intention to reduce repeated questions. There is no before-and-after measure of support time.
Review-ready example: “I reorganised the onboarding guide around the questions new starters were asking, incorporated colleague feedback and delivered the approved version. That addressed our goal of making the instructions easier to follow. We have not yet measured whether it reduced support time; I would like to agree a way to assess that next.”
This version names a delivered contribution and its intended purpose. It does not turn an expected benefit into a measured result. Noor should bring the approved guide and the relevant feedback, rather than a transcript of every conversation about it.
Expected result: a few concise examples you can explain and substantiate, without exaggerating your role or impact.
4. Prepare a balanced account of challenges and development
Identify a difficulty that matters to your role. Explain what happened, what you could influence, what you tried and what support would help. Avoid both blaming everyone else and claiming responsibility for conditions outside your control.
For instance, “I need to communicate better” is too broad to guide a useful discussion. A more workable account might be:
When the project deadline changed, I updated the delivery team but did not confirm that the support team had received the new date. I have started checking the affected groups explicitly. I would like feedback on whether my updates now reach the right people.
That example describes an observable behaviour, a change being attempted and a way to review it. It does not claim that the problem is permanently solved.
Development can also build on a strength. You might want experience facilitating a planning discussion or taking responsibility for a larger piece of work. Connect the request to the role and propose a practical opportunity to learn.
Keep the evidence specific and proportionate. You usually do not need to reproduce private information about a colleague or client to explain your own development need.
Expected result: development priorities framed as behaviours, opportunities and support needs, rather than vague personal labels.
5. Decide what you want to ask
A review is a conversation, so prepare questions alongside your account of past work. Choose the ones that resolve uncertainty or help you act after the meeting.
Useful questions include:
Which parts of my contribution were most useful to the team?
Where does your assessment differ from my self-assessment, and which examples explain the difference?
What would stronger performance in my current role look like?
Which goal should take priority if these commitments compete?
What support or experience would help me develop the skill we discussed?
How and when will we check progress on the next steps?
If you want to discuss progression or pay, confirm whether that decision belongs in this meeting and what process applies. Prepare a specific request and relevant evidence, but keep it distinct from the claim that a good review guarantees a particular outcome.
Avoid spending all your preparation rehearsing a speech. Leave room to hear information you did not expect and ask for concrete examples before responding.
Expected result: a short agenda that includes what you want to understand, not only what you want to say.
6. Assemble a concise review pack
Use your organisation’s required form first. If you have flexibility, prepare a short summary with supporting records linked or available separately.
The summary should cover your agreed goals, selected achievements, relevant challenges, development priorities and questions. Keep the detail needed to make the account credible, then remove repetitive background.
Bring the original evidence when it is useful, but select what to share. A private recording library is not a presentation deck. Verify quotations and speaker attribution, and avoid distributing whole conversations when a concise factual account will do.
Before submitting, read every achievement sentence as a claim. Can you distinguish “we discussed”, “I proposed”, “I delivered” and “the result was verified”? If the evidence supports only one of those statements, use that one.
You can rehearse aloud to find unclear wording. Aim for an explanation you can discuss naturally, rather than memorising a polished paragraph that you cannot adapt.
Expected result: a usable preparation document with a clear route back to the evidence.
If the review is tomorrow
Use this order when time is short. It is a proposed triage sequence, not a researched time-saving benchmark.
Confirm the criteria, form and deadline.
Find the previous review and current goals.
Choose a few examples from different points in the period.
Check your contribution and result against the available records.
Prepare one meaningful development topic and the questions that matter most.
Submit the required material and keep supporting evidence available.
If you cannot verify a result before the meeting, describe the contribution you can support and mark the outcome for follow-up. A candid gap is more useful than a confident number with no source.
Do not spend the remaining time building a perfect archive. Prepare a clear account for this review, then start a lightweight ongoing record for the next one.
During and after the conversation
Listen to the assessment, ask for examples where it is unclear, and note where you agree or see the evidence differently. If you need time to check a fact, agree how you will return to it.
Close by confirming the next steps, who owns each one and when progress will be reviewed. MIT’s guidance includes recording notes and planning follow-up after the conversation.
Afterwards, write a concise account of the agreed goals and unresolved points using the organisation’s normal process. Distinguish your proposal from an agreement, and an agreement from completed work.
Keep that record current as the work develops. The next review becomes easier to prepare when the evidence includes the whole period and the difference between what was intended, attempted and achieved remains clear.



