A people analytics dashboard should answer questions, not display metrics. The 12 metrics worth showing, how to lay them out by audience, and what to leave off.
A people analytics dashboard should show the smallest set of metrics that lets its audience make a decision — typically 6 to 12 for an executive view, each segmented by team and shown as a trend rather than a point value. The most common failure is the opposite: a forty-tile dashboard that displays everything the HRIS can export and answers nothing.
The test for including a metric is simple. Name the decision it changes. If nobody can name one, the tile is decoration, and every decorative tile makes the useful ones harder to find.
Three audiences need three dashboards. Trying to serve all of them with one view produces something that serves none.
| Audience | Question they are asking | Horizon | Detail level | |---|---|---|---| | Executive / board | Is the workforce healthy and is it changing? | Quarterly trend | Company and function | | HR business partner | Which team needs attention this month? | Monthly | Team and manager | | Line manager | What is happening in my team? | Monthly | Own team only |
The executive dashboard should fit on one screen and be readable in ninety seconds. The HRBP dashboard is a triage tool and can be denser. The manager dashboard should show almost nothing except their own team, because anything else invites comparison rather than action.
These eight cover retention, experience, hiring and capability without overlap. Every one of them changes a decision.
1. Voluntary turnover rate, rolling 12 months Not total turnover — voluntary, because that is the part you can influence. Show it as a rolling twelve-month line with the prior year behind it.
2. Regretted attrition as a share of voluntary exits Regretted attrition is the metric that distinguishes a healthy 15% from a dangerous 15%. If you show only one retention number to a board, show this one alongside the headline.
3. Engagement score with the trend A single engagement figure with three prior periods. Add the between-team variance as a second number — a flat average with a widening spread is a very different situation from a flat average with tight distribution.
4. Headcount vs plan Actual against planned headcount by function. This is the one tile finance will look at first, and including it buys credibility for the rest.
5. Time to fill for critical roles Not all roles — time to fill for the role families on the critical list. An average across all requisitions hides the two positions that are actually blocking delivery.
6. Internal mobility rate Internal mobility is the cheapest lever on retention and the best available proxy for whether people can see a future. It is also the metric executives most consistently underestimate.
7. Manager effectiveness distribution Not the average — the distribution. Show the proportion of managers in the top, middle and bottom quartile of manager effectiveness. The bottom quartile is where most of your engagement variance lives.
8. Bench strength for critical roles What proportion of critical roles have a ready-now internal successor. Bench strength is the metric that turns succession planning from a document into a risk number.
The HRBP view keeps the eight above and adds the diagnostic layer.
The organising principle here is exception reporting. An HRBP does not need to see every team every month; they need to see the four teams that changed.
Manager dashboards fail when they are built by analytics people who find the data interesting. Managers do not want a dashboard. They want to know whether something needs their attention.
That is the entire useful manager dashboard. Everything beyond it reduces the chance the first three get looked at.
Headcount as a vanity tile. Total headcount is not a metric, it is a fact. It belongs on the executive view only as headcount-versus-plan, where it carries information.
Anything you cannot segment. A company-wide number with no breakdown cannot be acted on. If the data does not support segmentation by team, the tile is a report, not a dashboard.
Cost per hire on an executive dashboard. It matters, but it is a recruiting operations metric. Executives cannot act on it, and it displaces something they can act on. Keep it on the TA dashboard where it belongs.
Survey participation as a headline. Participation rate is essential context for the engagement number and misleading as a metric in its own right — it goes up when managers apply pressure, which makes the data worse.
Point-in-time values with no trend. A number without history is unreadable. Is 14% turnover good? Nobody knows without last year's figure next to it.
Diversity data at a granularity that identifies individuals. Beyond the ethical problem, it destroys response honesty for every future survey.
One screen, no scrolling, for the executive view. If it does not fit, it is not prioritised.
Trend over value. Every tile should show direction. A sparkline behind a number does more work than a second tile.
Segment on the same page, not behind a click. The insight is almost always in the breakdown. If the breakdown requires navigation, it will not be seen.
Annotate the anomalies. A spike with a one-line note explaining the restructure is useful. A spike with no note generates three meetings.
Define every metric on the page. A hover definition with the formula prevents the recurring argument about whether turnover includes contractors. Most dashboard disputes are definitional, not analytical.
Monthly for most metrics, with a rolling twelve-month window. Weekly refresh on people data creates noise that looks like signal — a three-person team with one departure shows 33% turnover, and someone will escalate it.
The exceptions are open requisitions and requisition ageing, which are genuinely operational and benefit from weekly or live data.
How many metrics should a people analytics dashboard have? Six to twelve for an executive view. The constraint is not screen space but attention: every additional tile reduces the probability that the important ones get read. If you cannot name the decision a tile informs, remove it.
What is the difference between an HR dashboard and a people analytics dashboard? An HR dashboard typically reports operational activity — headcount, absence, open roles, training completions. A people analytics dashboard connects those measures to outcomes and shows relationships, such as engagement against subsequent turnover. The distinction is whether the dashboard describes what happened or supports a decision about what to do.
Should managers see their own team's turnover data? Yes, with a minimum team size — usually five — to protect anonymity, and always alongside the organisational comparison. Managers who cannot see their team's data cannot own the outcome, and ownership is the entire point of pushing the data down.
How do I stop a dashboard from becoming a forty-tile report? Require a named decision for every tile at the point it is added, and review the dashboard annually to remove tiles that no decision has ever referenced. Dashboards grow by accretion unless something actively prunes them.
Build this without the spreadsheet layer. PeoplePilot Analytics generates executive, HRBP and manager views from your existing HR data, with every metric segmented and defined in place. Start a free trial, or look up the formula behind any tile above in the HR Metrics Library.