How a system of intelligence HR transforms modern teams

HR teams translate workforce intelligence into board-level narratives by starting from a risk the board already tracks, attaching a small number of people measures to it, showing the trend rather than a snapshot, and closing with the decision they are asking the board to make. Workforce intelligence is the layer that sits above your system of record and explains why the numbers moved. The board narrative is the short version of that explanation, expressed in money, risk or capacity.

Last reviewed September 2026

How HR teams translate workforce intelligence into board-level narratives

Most people packs fail in the first slide, because they open with HR measures and ask the board to care. Boards care about delivery, cost, risk and reputation. The translation job is to tie a people number to one of those four, then keep the chain of reasoning short enough to survive a hostile question.

Four moves do most of the work.

• Start from the risk the board already has on its register. If the register says "delivery of the FY27 program depends on hiring 40 engineers", your people data has a home. If you arrive with an engagement score and no anchor, you are competing with the finance pack for attention and you will lose.
• Pick few measures and defend them. Two measures a director can remember beat nine they cannot. Every measure you include should survive the question "and what would we do differently if this moved".
• Show the trend and the comparison. A single figure invites argument about methodology. The same figure over six quarters, split by the division that matters, invites a decision.
• Name the decision. Close with what you want: budget, a policy change, a pause on a restructure, or simply an agreement to look again in a quarter. A pack with no ask is filed as reporting.

Directors tend to interrogate the source before the conclusion, so keep the provenance of each number one sentence away. There is a fuller walkthrough of the mechanics in our guide on how to present HR data to a board.

What belongs in the board pack

The quickest way to build a pack that gets read is to write the board's questions first and work backwards to the measure. The table below maps the questions that come up most often to the measure that answers them and where the number is usually sourced.

Leave out anything that cannot be tied to one of those questions. Training hours delivered, survey response rates and headcount by office are operational measures. They belong in your own dashboard, not in front of directors, and including them dilutes the two numbers you actually want remembered.

If the ask involves money, build the case in the same language finance uses. The HR business case builder is a reasonable starting structure, and revenue per employee is often the measure that gets a people conversation onto the same page as a commercial one.

Task intelligence in HR: what the work is actually made of

Task intelligence is the part of workforce intelligence that looks at the work itself rather than the person doing it. Instead of asking who sits in a role, it asks what the role is actually made of: which tasks repeat, which require judgement, which depend on someone else finishing first, and which have quietly moved to a different team without anyone updating the position description.

This has become a live question for two reasons. Automation changes the composition of a job long before it changes the headcount, so a role can lose half its routine tasks while the org chart looks identical. And hiring against a position description written four years ago produces confident, well-run processes aimed at the wrong target.

Task-level data usually comes from a mix of sources: job analysis or role profiling, workflow and ticketing systems, and capability frameworks that describe work in terms of what gets done rather than what the title says. Joined up, it answers questions a headcount report cannot, like whether a team is short of people or short of one specific kind of work.

Compono's work personality framework approaches the same problem from the behavioural side. It describes eight ways of working that every team needs covered to some degree: Doer, Auditor, Helper, Advisor, Pioneer, Campaigner, Evaluator and Coordinator. The framework describes work, not fixed personality labels for people, and a team can be well staffed on headcount while one of those ways of working is barely represented. Reading task data and behavioural data together is how you tell the difference between a team that needs another pair of hands and a team that needs a different kind of contribution.

HR intelligence versus a system of record

A system of record answers who, what and when. It pays people correctly, tracks leave, stores contracts and produces the compliance reports you are obliged to produce. Those are not small things, and any organisation without one has a bigger problem than board narratives.

What a system of record cannot do is explain why. It knows that fourteen people left the operations division last year. It does not know that eleven of them reported to managers appointed in the same restructure, because that connection lives across three systems that were never joined. HR intelligence is the joining work plus the interpretation, and most of the value sits in the joins rather than in any single dataset.

There are two different risks here, and most HR technology only addresses one of them. Process risk is admin, payroll accuracy, compliance and onboarding errors, and the market handles it well. People insight risk is the wrong hire, the top performer nobody noticed disengaging, training that does not build the capability the plan needs. That second risk is the one boards ask about and the one most reporting cannot answer. Compono's HR software was built around the second, which also means it is a poor fit if what you need right now is a payroll engine.

One caution on measurement. A single engagement score is a thin input for a board conversation, because it compresses culture, climate and management quality into one number that moves for reasons nobody can name. It is worth understanding how engagement surveys and culture measurement differ before you build a narrative on top of either.

A worked example of a board narrative

What follows is an illustrative example with made-up numbers, not client data. Take a 400-person business with 60 exits last year. HR classifies 22 of those as regretted. Fourteen of the 22 came from one division of 90 people, which is a regretted turnover rate of 15.6 per cent in that division against 2.6 per cent across the remaining 310 staff.

The narrative writes itself from there. One division is losing people the business wanted to keep at roughly six times the rate of everywhere else. The delivery risk is specific: that division owns two of the four commitments in the operating plan. The ask is specific too, whether that is funding for a management capability program, a pause on the next restructure phase, or an independent read of that division's culture before any of it.

Notice what the narrative does not do. It does not claim to know the cause, because turnover data alone cannot establish one. It says where to look and what decision is needed now, which is what a board can actually act on.

Where this usually goes wrong

The most common failure is volume. A thirty-page pack signals effort and produces no decision, while three pages with one clear ask produces a minute in the board papers. The second is precision theatre, where a number is quoted to one decimal place from a survey with a 41 per cent response rate. State the confidence you actually have.

The third is a missing counterfactual. If attrition fell, say whether the market also softened, because a director will ask and the answer needs to already be in the pack. And the fourth is the narrative that never returns. Bring the same measure back next quarter, whether or not it moved in your favour, or the board learns that your reporting is selective.

Frequently asked questions

How do HR teams translate workforce intelligence into board-level narratives?

Anchor the people data to a risk the board already tracks, choose a small number of measures you can defend, show the trend and the comparison rather than a single figure, and close by naming the decision you want the board to make. A pack with no ask gets read as reporting.

What is task intelligence in HR?

Task intelligence looks at the work rather than the worker. It asks what a role is actually made of: which tasks repeat, which need judgement, which have shifted to another team, and which are being automated. It tells you whether a team is short of people or short of a particular kind of work.

What is the difference between HR intelligence and an HRIS?

An HRIS is a system of record. It answers who works here, what they are paid and when they took leave. HR intelligence joins that data to hiring, culture and capability information and interprets it, which is what lets you answer why something happened rather than only what happened.

Which people measures should go in a board pack?

Only measures that answer a question the board is already asking: capacity against the plan, regretted turnover and its cost, loss of high performers, culture trends by team, the defensibility of hiring decisions, and capability gaps against future need. Operational measures such as training hours belong in your own dashboard.

Is workforce intelligence only worth it for large enterprises?

No. Mid-sized organisations often get more out of it, because they have passed the point where a leader can know everyone personally but have not yet built the reporting layer that larger businesses take for granted.

How quickly will a board notice the difference?

The format change lands immediately, since a short pack with a named ask gets a different reception at the next meeting. The measures themselves need at least two or three reporting cycles before a trend is worth arguing from.