---
title: Read a team's health trends over time
description: How the trends grid encodes state and movement, when a drop alerts facilitators, and which dimension the needs-attention strip features
order: 3
---

Trends place a team's closed health checks side by side, one column per run and one
row per dimension. Cell color and the Movement column encode the same two signals for
every model and scale.

## Where it lives

Open **Team health → Health trends** in the sidebar - the leaf renders for the manager
and admin personas. Pick a team in the context bar; the **Health model** select defaults
to the latest closed run's model. The **Dimension history** grid lists that model's last
12 closed runs as columns, newest on the right, one row per dimension.

## How to read a cell

The grid encodes two signals:

- **Color is state**: where the score sits in its own scale, in thirds - the legend
  above the grid names them **Low** (red), **Mid** (amber) and **High** (green). Because
  the bands come from the scale's own range, a 1–5 Pulse dimension, a traffic-light
  rating and an eNPS row all read the same way. (eNPS rows band on the −100…100 score and
  print a signed value like +27; every other row prints 3.4 / 5.)
- **The Movement column is run-to-run change**: every row carries **↑ Improving**, **↓ Review**
  or **→ Stable**, set by the last change of at least **15% of the scale range** - 0.6
  points on a 1–5 scale, 1.5 on 0–10.

A dimension that drifts across bands over several runs, with no single step reaching the
threshold, carries **→ Stable** throughout.

Each column header shows the run's close date, and its full title on hover. Each row
header carries the dimension name and its response count in the latest run; the summary
strip above reports that run's participation as submitted of invited.

A cell is a dash when the run recorded no answer for that dimension, or when its answers
stayed under the run's reveal threshold - the grid never substitutes an overall average to
fill the gap. To keep runs comparable, a model's dimension set locks as soon as any health
check uses it, draft included.

## Needs attention

The featured dimension is the declining one, or failing that the one sitting lowest in
its own scale; the line chart draws it across the window. Carrying the ↓ Review verdict
or sitting in the lowest band also raises a **Needs attention** strip above the chart,
linking to **Review open actions** and **Open latest run**.

## Deltas and alerts

The same 15%-of-range rule that sets a row to **↓ Review** fires the "Team health score
dropped" notification to every facilitator when a run closes. A rise raises no
notification.

## Worked example

Platform's [first Pulse 5 check](/docs/health/first-run/), which spawned Marta's
on-call rebalancing action, is the baseline: Workload averaged 2.2, a red cell, with no
earlier column, so no movement and no notification. The next run climbs to 3.0 once the
rotation changes - the row turns **↑ Improving**, into amber. The run after that slides
to 2.4: a 0.6-point drop, exactly at the threshold - **↓ Review**, the cell still amber,
Workload featured in **Needs attention**, and every facilitator gets the "score dropped"
notification.

> [!AGENT]
> Pull Platform's health trends for the Pulse 5 model and tell me which dimensions moved last run and whether anything red has no open action.

## Related

- [Run your first team health check](/docs/health/first-run/)
- [Turn results into improvement actions](/docs/health/actions/)
- [The eNPS model](/docs/health/enps/)