The analysis view: talent grid, placements and gaps
The analysis view is where performance data becomes a picture of your workforce rather than a list of ratings. It reads the score cards for a period and presents where people sit, which way they are moving, and where capability is thin.
Where: Performance › Analysis You need: performance rights covering the employees shown
What it shows
| View | Answers |
|---|---|
| Talent grid | Where does each person sit, and which way are they trending? |
| Employee placements | How is the workforce distributed across the grid? |
| Manager scorecard | How does each manager's team look, together? |
| Gap analysis | Where is capability missing? |
All of it is built from score cards, so the analysis is only as sound as the weights and cycles behind them — see Score cards.
The talent grid
Ratings become scores by multiplying by 20, and scores fall into three bands:
| Band | Score | Rating |
|---|---|---|
| Below | under 60 | below 3 |
| Meets | 60–79 | 3 to just under 4 |
| Exceeds | 80+ | 4 and above |
Trend compares the start of a window against its end: a move of 3 points or more is improving or declining, anything smaller is steady. Three points is well under one rating step, so trend catches genuine movement without treating ordinary variation as a signal.
See Performance scores and the talent grid for how to read it honestly.
Manager scorecard
The manager view aggregates a manager's team. Its real value is not judging the team — it is comparing managers, because the commonest integrity problem in performance data is that two managers rate the same quality of work differently.
If one manager's team is uniformly Exceeds and another's is uniformly Meets, the likeliest explanation is the managers, not the teams. That is worth knowing before any decision is made on these numbers.
Gap analysis
Gap analysis highlights where capability is thin — skills or performance areas with little depth behind them.
Read it as a prompt to investigate, not a conclusion. A gap can mean the capability is genuinely missing, or that nobody recorded the skill, or that the relevant people were not in the cycles the score cards selected.
Saved reports
An analysis can be saved, so a view you rely on is reproducible rather than rebuilt from memory each time.
What Happens Next
- The view reflects score cards as they stand; it changes when the underlying ratings or the score card definition change.
- It is a report, not a record — nothing in KAMI acts on a grid position.
- What follows is a management decision, made by people who know the context the data does not.
Tips
- Check the distribution before reading individuals. If nearly everyone is in one band, the inputs are not discriminating — look at the assessment design before drawing conclusions about people.
- Compare managers before comparing teams. Rating behaviour varies more than performance does.
- Treat a single cycle's trend cautiously. One movement is not a pattern.
- Check who is missing. People without ratings do not appear at all, and absence is not a performance signal.
- Never use this view as the sole input to a pay or promotion decision. It summarises ratings; it knows nothing about context, circumstances or potential.
Troubleshooting / FAQ
Q: Everyone is in Meets. Ratings are clustered at 3. Check whether the assessment discriminates and whether reviewers default to the middle.
Q: Someone is missing. They have no score card for the period — check they were in the selected cycles.
Q: Two teams look very different. Compare their managers' rating behaviour before concluding anything about the teams.
Q: The trend says steady but the score moved. It moved by less than 3 points.
Q: The grid disagrees with what I know about someone. The grid summarises ratings. If the ratings are wrong, fix the ratings — do not reinterpret the grid.