Start with three questions, not a dashboard
People analytics fails in Nigeria for the same reason it fails everywhere: teams start with a tool instead of a decision. Before you touch data, get your leadership to name three decisions they would make differently with better information. In our client work those are almost always the same three — why are we losing good people, what does our headcount really cost, and who is ready for the next level.
Those three questions become your scope. Anything that does not serve them is a chart nobody will open twice.
The eight metrics worth your time
You do not need forty KPIs. These eight cover retention, cost, capability and structure, and every one can be computed from data you already have.
Voluntary attrition rate
Resignations ÷ average headcount × 100
The single clearest signal of whether pay, managers, or careers are broken. Track it monthly by department and by manager, not just company-wide.
Regretted vs non-regretted exits
Split attrition by whether you wanted to keep the person
A 20% attrition rate is healthy if it is mostly non-regretted, and alarming if your top quartile is walking. Most Nigerian dashboards never make this split.
Time to fill and time to productivity
Days from requisition to offer accepted; days to full output
Time to fill measures recruiting. Time to productivity measures onboarding — and it is where most of the real cost sits.
Cost per hire, in naira, fully loaded
(Agency + advert + assessment + internal hours) ÷ hires
Include recruiter and hiring manager hours. Once leaders see the loaded number, referral and internal mobility programmes get funded.
Internal mobility rate
Internal moves ÷ total roles filled × 100
In a market where salary inflation makes external hiring expensive, mobility is the cheapest retention lever you own.
Manager span and layer depth
Direct reports per manager; levels from CEO to front line
Spans below four usually mean you are paying for supervision instead of work. This is the most common finding in Nigerian organisation reviews.
Absence and overtime concentration
Share of total absence or overtime held by the top 10% of staff
Averages hide the problem. Concentration tells you whether you have a policy issue or a handful of units under strain.
Pay equity gap
Median pay by gender and level, same job family
Cheap to compute from payroll, and increasingly asked for by international partners and lenders.
Where the data actually lives
Four sources carry most of the value: payroll (headcount, cost, pay equity), the leave and attendance register (absence, overtime, strain), the recruitment tracker (time to fill, cost per hire, source quality), and exit interviews (why people leave, in their words). The work is rarely analysis — it is reconciliation.
Give every employee one ID that appears in all four systems, standardise job titles into a short job-family list, make department a dropdown rather than free text, and record real hire and exit dates. Do that and 80% of your reporting problems disappear before you build a single chart.
A 90-day rollout
Weeks 1–2
Agree three questions, not thirty metrics
Sit with the MD and two line leaders and write down the three decisions they would make differently with better people data — usually about attrition, headcount cost, or where to promote. Everything downstream serves those three questions.
Weeks 3–4
Clean the data you already own
Payroll, the leave register, the recruitment tracker, and exit interviews are enough to start. Fix the basics: one employee ID everywhere, consistent job titles, a real hire date and exit date, and a department field nobody free-types.
Weeks 5–8
Build one dashboard and one narrative
One page: headcount and cost, attrition split by regretted, hiring pipeline, and manager spans. Pair every chart with a sentence in plain English. A dashboard nobody can explain in a board meeting is a dashboard nobody uses.
Weeks 9–12
Run one intervention and measure it
Pick the worst-performing unit, agree one change with its manager, and re-measure after a quarter. Proving one naira-value result buys you the budget and the credibility for everything after.
Five mistakes we see most often
- Buying a global HR analytics platform before your employee master data is clean — the tool will faithfully report the mess.
- Reporting company-wide averages that hide the two departments causing the problem.
- Measuring engagement with an annual survey and nothing between them.
- Treating headcount cost as basic salary only, when allowances, pension, and NHF often add 30–40%.
- Building dashboards for HR instead of for the people who approve budgets.
Doing this without building it yourself
We built HR Metrics Studio to do exactly what this guide describes — attrition, cost, spans and hiring pipeline on one page, from data you upload. Talento HRIS covers the employee record side so the numbers stay clean at source. Both have a free 24-hour guest login if you want to see your own numbers before committing to anything.
If you would rather have it done with you, our people analytics and organisation design work takes teams through this 90-day plan end to end.