Score cards: how the overall rating is built

Score cards: how the overall rating is built

A score card is what turns everything else in Performance into a single number. It takes assessments, goals and feedback over a period, weights each of them the way your company has decided, and produces one overall rating per person.

Where: Performance › Score cards · the definition: Performance › Settings You need: performance rights covering the employees scored

The four components

Every score card blends exactly four inputs:

ComponentComes from
Self assessment scoreThe employee's own answers
Reviewer assessment scoreTheir reviewer's answers
Reviewer goal ratingRatings given against their goals
Feedback ratingRatings from feedback given about them

Each carries its own weight, set on the score card definition. Those four weights are the most consequential configuration in the module: they decide, numerically, what your company means by performance.

💡 The weights are where your review philosophy becomes explicit. Weighting self-assessment heavily says you trust people's own judgement; weighting goals heavily says you care about delivery over behaviour; weighting feedback heavily says you value how people work with others. Most companies have never written this down — the score card forces the decision.

What the score card definition holds

SettingDecides
NameWhat this score card is called
Start and end dateThe period covered
Selected goal cyclesWhich goal cycles count
Selected assessment cyclesWhich assessment cycles count
Four weightsHow much each component contributes

Cycles are selected explicitly, not inferred from the dates. That is deliberate and worth knowing: a cycle that ran inside the period but was not selected does not contribute. It also means you can build a score card from exactly the cycles you consider representative.

What ends up on a person's score card

FieldIs
Self assessment scoreTheir self-assessment result
Reviewer assessment scoreTheir reviewer's result
Reviewer goal ratingTheir goal ratings
Feedback ratingTheir feedback ratings
Overall ratingThe weighted combination

The overall rating is what flows into the talent grid, where 60 is the bottom of Meets and 80 the bottom of Exceeds — see Performance scores and the talent grid.

Before You Start

Decide the weights before you build the card, with whoever owns the review policy. Changing them afterwards changes everybody's rating at once, and people will notice.

Check the underlying cycles are sound. A score card is a weighted average of its inputs, so it inherits their problems:

  • Multi-select assessment questions award full marks to everyone and flatten the assessment component entirely — see How assessment scores are calculated.
  • Goal priority weights the goal component. If every goal is Critical, that weighting does nothing.
  • Feedback ratings only exist where feedback was given. A person nobody was asked about has a thin feedback component.

How to create a score card

  1. Open Performance › Settings and create a score card definition.
  2. Name it and set the period.
  3. Select the goal cycles and assessment cycles that should count.
  4. Set the four weights.
  5. Generate the score cards and review the distribution before showing anyone.

What Happens Next

  • Each person in scope gets a score card with the four component scores and an overall rating.
  • The overall rating feeds the talent grid and the analysis view.
  • Changing the definition changes the resulting scores — it is a live calculation over the selected cycles, not a frozen snapshot.

Tips

  • Look at the distribution before the individuals. If everyone lands in a narrow band, the inputs are not discriminating and no amount of weighting will fix it.
  • Sanity-check one person by hand. Take their four component scores, apply the weights, and confirm you get the overall rating. If you cannot reproduce it, you cannot defend it to them.
  • Do not weight feedback heavily unless feedback is being given consistently. A component that is empty for half your people makes their overall rating incomparable.
  • Write down why the weights are what they are. In a year, someone will ask, and "that is how it was set up" is not an answer anyone accepts about their own rating.
  • Keep the weights stable between periods. Changing them makes year-on-year comparison meaningless, which is most of the value of scoring at all.

Troubleshooting / FAQ

Q: An employee has no score. Check they were in the selected cycles, and that those cycles completed.

Q: A cycle ran in the period but is not counted. Cycles are selected explicitly on the score card definition, not picked up by date.

Q: The overall rating does not look like the average of the components. It is a weighted combination, not an average. Check the four weights.

Q: Everyone scored similarly. Look at the assessment component — multi-select questions award maximum to everyone and flatten it.

Q: Someone's feedback component is empty. No feedback was given about them in the period. Consider whether the feedback weight is fair to people nobody was asked about.

Q: We changed the weights — did old score cards change? The score card is calculated from its definition, so changing the definition changes the result.

Related articles