Engagement: employees with the lowest engagement (Diagnostic). Lowest participation first. The starting point for a re-engagement conversation.

Engagement: employees with the highest engagement (Diagnostic). The same index read the other way.

Archetypes: profile distribution (Diagnostic). The whole organisation distributed across the behavioural profiles, using HR names rather than codes. Each person counts once. A profile is a summary of behaviour over time, not a label attached to a person.

High performance with low platform involvement (Diagnostic). People with a formal review at the top and low platform activity. An observable crossing to explore, not a list of alerts, and it does not infer exit risk.

High performers (Factual). A direct ranking on the performance review score, highest first, restricted to people with a formal review. Not crossed with engagement or risk.

Wellbeing: most fragile teams (Factual). Wellbeing by team, worst first, based on mood and pulse survey responses.

eNPS: recommendation as an employer (Factual). Employee Net Promoter Score from -100 to +100: promoters minus detractors. You can compare it with an earlier snapshot.

Team Health Index: most fragile teams (Diagnostic). Overall health per team on a 0-100 scale, combining wellbeing, retention, learning, growth and social connection. Worst first, red below 40 and amber below 65.

Top 10: highest Manager Index (Diagnostic, marked Experimental). The managers who most develop, move and retain talent, on a 0-100 index combining promotion, internal mobility, retention and team engagement against the organisation's average. Organisations without a formal review see the retention component only.

Priority attention: 10 managers with the lowest Index (Diagnostic, marked Experimental). The same calculation, read from the bottom. This is a development instrument and a candidate list for coaching. It is not a performance assessment of a manager and must not be used as one.

Recognition network: main connectors (Diagnostic). Who connects the organisation most through recognitions given and received, with network totals and a table of the main connectors. A higher score means a more central person in the network, not a better employee.

Warning signs: teams with the most flagged people (Diagnostic). The number of flagged people per team, with the teams with the most signs first; it does not show every team. A person is flagged for low wellbeing, low mood or a strong inactivity signal on the platform. Only in organisations that have a validated exit-risk model, the people that model scores highest are flagged as well, and they are the ones shown in red (critical) or amber (high).

Turnover: exit rate (Factual). Exits in the chosen period divided by the average active headcount of that period, twelve months by default (3, 6, 12 or 24 months). An exit is a confirmed leaving date, never a prediction. Green below 10 per cent, yellow from 10 to 20, red above 20.

Exits per month (Factual). Real exits per month across the chosen period, counted on the leaving date.

Onboarding: new employees (Diagnostic). Employees in their first days on the platform (90 by default; 30, 60, 90 or 180), ordered by behavioural engagement with the lowest first. The first ninety days are the earliest point at which intervening is cheap.

Organisational health: behavioural dimensions (Diagnostic). The organisation across wellbeing, retention, learning, growth and social connection, showing only the dimensions available for your organisation. A switched-off data source does not appear as a zero. You can compare it with an earlier snapshot.

Learning activity: completed content (Factual). Contents completed per person in the chosen window, 30 days by default (7, 30 or 90 days). Content without a completion date was not completed and does not count. It measures observed completions, not speed, motivation or learning potential.

Mandatory training: completion status (Factual). Not started, in progress and completed, with the completion rate. A healthy benchmark is above 70 per cent completed.

Mandatory training: who is furthest behind (Factual). The people with the most mandatory training still outstanding. An action list for managers.

Mandatory training: completion by team (Factual). Worst completion first, with the largest gaps at the top. The bars stack the number of enrolments rather than a percentage, on purpose: a large team at zero per cent shows as a long bar, which is the real size of the gap, instead of a flat zero indistinguishable from a three-person team. This Insight has no filters.

Mandatory training: overdue (Factual). People past a mandatory training deadline. This can legitimately read zero when no deadline has passed yet, and becomes critical as deadlines go by.

9-box - talent matrix (Factual). Performance against potential, using your own position names. Where both axes have enough coverage it is a 3x3 grid; where performance cannot serve as an axis it shows positions ordered by potential instead of forcing a grid that would leave people out, and the Insight tells you which of the two you are looking at. Latest cycle only. It can be filtered by the team recorded in the talent review.

High potential (HiPos): future leaders (Diagnostic). The people in the top cells of the 9-box, ordered by potential score. The source is the manager's potential review.

High potential at risk of leaving (Diagnostic). High-potential employees the manager flagged with a high exit risk in the talent review. This is the manager's assessment, not a model output.

Self-assessment vs manager's assessment (Diagnostic). The gap between the two ratings on the same scale, largest over-rating first. The gap is more informative than either absolute value.

Critical roles and succession (Factual). People in roles marked critical and how many have an identified successor. No successor means succession risk, shown in red.

Review status: which stage the population is at (Factual). Where each person stands in the current cycle: self-assessment pending, manager's review pending, or finished. This is the only Insight that shows a cycle in progress. Where an organisation runs several periods in one cycle it never aggregates them, and always names the one it used. People who have already left still count in the population, because the cohort is who was enrolled; the number who left is reported separately.

Calibration: headline indicators (Factual). Eligible population, reviews completed, eligible for calibration, and critical roles at risk with no successor. The four numbers come from three different sources and degrade independently, so you may see two real numbers and two explained absences rather than a band that looks complete when it is not.

Talent pipeline health (Diagnostic). The shape of the talent matrix as configured in your organisation, with the number of high potentials and the submission rate of potential reviews. Aggregate, no names.

Potential review coverage (Factual). The percentage of potential reviews submitted, at organisation level only. It is deliberately not broken down by team: in some organisations that source's team column holds managers' names, which is personal data, so only the aggregate rate is shown.

Potential review: distribution (Factual). How potential was classified across the reviewed population. Blank answers appear as their own category rather than being folded into "no", because a person nobody reviewed is not the same as a person reviewed and found not to have potential. It is a manager's judgement in one cycle, not a permanent attribute of a person.

Calibration eligibility funnel (Factual). From the reviewed universe to who can actually enter calibration, step by step, showing where the process loses population.

Critical role: assessment (Factual). How many roles managers marked critical, keeping non-critical and unanswered separate. This is the base population for the Insights below.

Potential x critical role (Diagnostic). The overlap that concentrates risk, and specifically the group with potential and a critical role but no performance rating, who therefore fall outside calibration and out of sight.

Exit risk: critical roles (Diagnostic). Exit risk assigned by the manager, restricted to critical roles, most severe first. This is a manager's perception recorded in the talent review, not a model output. Aggregate, no names.

Succession readiness: critical roles (Diagnostic). Successor coverage: ready now, ready in one to two years, or nobody identified. Highlights the priority zone, a critical role with high exit risk and no successor. It uses the structured coverage field and never free-text successor names, which are personal data.

Employees: detail by manager and director (Factual). The nominative talent-review table, one row per active employee, filterable by director and manager. People with no talent-review record still appear, with the review columns empty, because the table is the active headcount and hiding them would overstate the coverage of the process. The director is derived from the manager's manager.

Data sources switched off: when a data source has been switched off for your organisation's People Intelligence, the Insights that depend on it are not offered. Wellbeing needs Mood or Pulse Surveys; eNPS needs Pulse Surveys; Recognition needs Recognition; Learning and Training need Learning; everything under Performance and Succession, plus High performance with low platform involvement and High performers, needs Evaluation.

Fields your data must carry: some Insights are offered only when your evaluation data has what they need. High potential, High potential at risk of leaving, Self-assessment vs manager's assessment and Critical roles and succession need those specific talent-review fields. The Succession Insights (except the employee table) need a potential review or a critical-role form. Review status needs an open review cycle, and Calibration: headline indicators needs at least one of these three.

Offered even without data: every other Insight is offered, and if your organisation has no data for it yet, the card says so instead of disappearing.