Expectations & Performance Canvas

Write down what success means in one role before the period starts, collect what you actually see during it, and turn the review into a conversation about examples and next goals instead of a verdict.

Goal-setting is strong research (Locke & Latham, 2002). SMART is a practitioner mnemonic with weak or contested support (Doran, 1981). Rater bias is long established (Thorndike, 1920), but calibration and rating distributions have mixed evidence (DeNisi & Murphy, 2017). The weights and score bands on this page are rules of thumb.

Use it when:

• You are setting expectations for a new role, a new report or a new period.
• A quarterly or annual review is coming and you want ratings you can back with examples.
• A performance concern is building and you want it written down fairly before it turns into a formal process.

Skip it when:

• You want a regular development conversation without ratings: use 4.9 Development & Coaching Notes.
• The issue is one incident or a conflict, not a pattern: use 4.6 Conflict & Feedback Guide.

Setting expectations works because of one of the sturdiest findings in work motivation: specific goals tend to produce higher performance than vague ones, a result with a long research record (Locke & Latham, 2002). The same literature also records what goes wrong when goals are overused, such as tunnel vision on whatever gets measured. SMART is a practitioner mnemonic from a management magazine (Doran, 1981), with weak and contested support of its own, so here it serves only as a writing checklist; the evidence lives in goal-setting theory. On the judging side, rater bias is old and replicated news: Thorndike (1920) showed that one general impression leaks into every specific rating, the halo effect. The usual remedies, calibration meetings and target rating distributions, are only weakly tested, so use them as a bias check and never as a quota.

That is why every rating on this page sits next to an observation, since a rating without evidence will not survive the conversation, and why every goal needs a metric, a date and the support it requires.

Write this with the person, not for them. Describe outcomes someone else could observe.

Strong answer: "Every bus due for morning traffic leaves the depot on time, and planned work is closed in the week it was scheduled." Not: "Show ownership and be proactive."

Role:

Review period:

What excellent looks like in this role, in two or three sentences:

The two results that matter most this period:

Illustrative example: the roles, the organization and all numbers are made up.

The workshop manager at a regional bus operator in Norway reviews a maintenance planner after twelve months.

Section 1: "Every bus due for morning traffic leaves on time; planned work closes in its scheduled week." The two results: fleet availability and on-time planned work.

Section 2: Result 1, planned work: 41 of 46 orders closed in week in March, similar most months; rated 4. Behaviors: trained two new mechanics on the planning system, but skipped two shift handovers in May; rated 3. Impact: rebuilt the night schedule after a depot flood; first rated 5.

Section 3: the bias check flagged halo: the flood night stood out, but it was the only Impact evidence. Rated 4 instead. Weighted score 3.8, Solid.

Section 4: the planner rated her own Behaviors at 2, citing the handovers. The manager added the training evidence; they agreed on 3 and on what a good handover looks like.

Section 5: move brake inspections into the planned schedule, measured by emergency work orders per month, by end of Q3. Development goal: lead the winter tire changeover plan with the manager as reviewer. Check-in booked for six weeks later.