People Intelligence: the Insight dashboard for your organisation
A closed catalogue of Insights, authored by GFoundry, that you assemble into your own dashboard. Every Insight carries a versioned metric contract, an honesty classification and server-side privacy protection. Your organisation's data stays in your organisation.
What it is
People Intelligence is the analytics surface of the GFoundry backoffice. Instead of a fixed dashboard that everyone gets, or a blank report builder where you assemble metrics yourself, it gives you a catalogue of ready-made Insights. You pick the ones that answer the questions you actually have, and they become your dashboard.
Each Insight is one question with one answer: which teams are most fragile, who is furthest behind on mandatory training, where the succession pipeline thins out, how the last review cycle is progressing. It is not a chart you have to interpret from scratch. The Insight states what it shows, what it does not show, and how confident you should be in it.
Where to find it
People Intelligence lives in the GFoundry backoffice, alongside the Gi assistant. Open the Dashboards section, then Add Insight to browse the catalogue. There is a second surface, on an individual employee's profile page, covered in Talent Intelligence: Insights on an employee's profile.
Why the Insight catalogue is closed
You cannot author an Insight, and that is deliberate.
A self-service metric builder lets two people in the same organisation produce two different turnover rates and argue about which is right. Worse, it lets someone build a number that looks authoritative and is quietly wrong, because nobody wrote down which population it covers or what it excludes. In people analytics that is not a cosmetic problem: the numbers end up in conversations about individuals.
So GFoundry authors every Insight, and every Insight has to pass a fixed checklist before it can be offered to you. If you need an Insight the catalogue does not have, that is a conversation with us, not a field you fill in. See the co-creation note at the end of this article.
Two kinds of tile on one dashboard
That does not mean the dashboard is read-only. It holds tiles from two different origins, and the difference between them matters more than anything else on this page.
Insights from the catalogue, authored by GFoundry. Each one carries a metric contract, an honesty classification badge and a freshness date, and each one can be recomputed against the latest data.
Charts you composed yourself, pinned from the Data Analyst Agent. You ask a question in natural language, choose the fields and the cut you want, and pin the resulting chart onto your dashboard. It sits alongside the Insights.
A pinned chart is your artefact, not ours. It has no metric contract, no honesty classification and no version, because nobody certified it: you composed it and you know what you asked for. It is also a frozen snapshot rather than a live tile, so a chart pinned in March still shows March's numbers until you pin a fresh one.
The line between the two is worth stating plainly: you can pin facts you composed; intelligence is authored by us. A crossing such as exit risk against manager effectiveness is not self-service by design. It carries assumptions that need writing down and validating, so it enters through the catalogue with a contract attached, or it does not enter at all.
How to pin one is in Building your People Intelligence dashboard.
What every Insight carries
A metric contract: the decision it answers, the formula, the population it covers, what it excludes, the time grain and lookback window, the data sources it reads, its thresholds and the reasoning behind them, its known limitations, and the interpretations that are explicitly forbidden. The contract is versioned, so a change in meaning is a new version rather than a silent edit.
An honesty classification: Factual, Diagnostic, Predictive or Experimental. This is the badge you see on the card. It tells you how much weight the Insight can carry. See How to read a People Intelligence Insight.
A freshness stamp: each card shows the date of the data it is built on, taken from the source snapshot rather than from the clock. A card is never presented as current when its source is a week old.
Server-side privacy protection: the rules about what may be shown, and to whom, are enforced where the data is computed, not in the browser.
Your catalogue is not the whole catalogue
Two organisations open the same catalogue and see a different number of Insights. That is expected, and there are two reasons for it.
Active modules. An Insight that reads pulse surveys cannot be offered if Pulse Surveys is not switched on. The same applies to Learning, Recognition, Mood and the evaluation cycle. Switch the module on, use it for a while, and the Insights that depend on it appear.
What your data actually carries. Some Insights need specific fields to exist in your evaluation extract, for example a potential review, a critical-role marker, or an open review cycle. If those columns are not there, the Insight is withheld rather than offered and then rendered empty. An Insight that would show you nothing is worse than an Insight that is not there.
At the time of writing, the catalogue holds 37 organisation-level Insights across eleven categories, and 13 person-level Insights on the employee profile. A given organisation typically sees between twenty and thirty-five of the first group.
The categories
Insights are grouped by the decision they support: Engagement, Talent, Wellbeing, Leadership, Recognition, Retention, Overview, Learning, Training, Performance and Succession. The full list, with what each Insight answers and what it deliberately does not answer, is in The People Intelligence Insight catalogue.
Who it is for
HR directors and CHROs: a defensible read of the organisation, with the caveats attached rather than discovered later.
People analytics teams: a set of certified starting points, each with its population and exclusions written down.
HR business partners and managers: the operational lists that turn a pattern into a next conversation.
Customer success teams: a consistent view to bring to a client sponsor without rebuilding the analysis each time.
What People Intelligence does not do
It does not decide anything about a person. No Insight produces an outcome for an employee. Promotion, exit, pay and development decisions stay with the people accountable for them, informed by evidence that goes well beyond a dashboard.
It does not compare your organisation with other organisations. Behavioural measures are normalised inside your own population. A score of 70 in your organisation and a 70 in another are not the same number, and we do not present them as one.
It does not let you author an Insight. You can compose and pin your own chart, as described above. What you cannot do is have your own measure carry a metric contract and an honesty classification, because those mean something only when somebody certified them.
It does not show a control that does nothing. If an Insight has no real time window, it has no period selector. A filter you can move that changes nothing is a lie told slowly.
If the catalogue is missing what you need
New Insights are co-created with clients, and they go through the same route every time: the decision it would support, the metric contract, a prototype on sandbox or synthetic data, a shadow validation against an independent calculation that you check, then certification and a controlled launch. An Insight that cannot survive that route does not ship. Talk to your customer success contact to start one.
A good way to make that conversation concrete is to build the thing in the Data Analyst Agent first and pin it. A chart you already rely on every month is the strongest possible case for turning it into a certified Insight.
Where to go next
How to read a People Intelligence Insight. The badges, the freshness stamp, the filters, and what a suppressed or unavailable card means.
Building your People Intelligence dashboard. Adding Insights, pinning your own charts, filtering, refreshing and removing.
The People Intelligence Insight catalogue. Every organisation-level Insight, what it answers, and what it refuses to claim.
Talent Intelligence: Insights on an employee's profile. The thirteen cards on an individual profile.
People Intelligence: privacy, limits and responsible use. What is protected, what is withheld, and the questions a works council or DPO will ask.
Data Analyst Agent. Asking questions in natural language, building charts and Excel exports, and pinning a chart to your dashboard.