TL;DR

HR analytics is the practice of collecting, organising, and interpreting workforce data to make better decisions about hiring, retention, and people strategy. It ranges from basic operational reporting (how many open roles, what is the average time-to-fill) to predictive modelling (which employees are likely to leave in the next 90 days). Recruiters encounter it both as a discipline: analytics embedded in their own workflows: and as a client capability that shapes how staffing partners are evaluated.

What HR Analytics Actually Measures

The field covers three layers of increasing complexity, and most organisations operate primarily at the first. Descriptive analytics: the most common layer: answers "what happened": headcount trends, cost-per-hire by department, attrition rates by tenure band. This is standard reporting found in any ATS or HCM dashboard. It is useful for operational reviews but does not explain causes or predict outcomes.

Diagnostic analytics asks "why did this happen" and requires connecting datasets that often live in different systems. Why did the engineering team's 90-day attrition spike from 12% to 28% between Q2 and Q3? Answering that question means joining exit interview data, manager tenure records, compensation bands, and project assignment history. Most organisations lack the data infrastructure or analyst capacity to do this routinely, which is why diagnostic analytics is underused relative to descriptive.

Predictive analytics, the third layer, models the probability of future events. Flight-risk scoring, time-to-fill forecasting, and candidate quality prediction all fall here. These models require historical data volume, clean data hygiene, and often dedicated data science capacity. Organisations running Workday People Analytics, Visier, or IBM Watson Talent are working at this layer. The outputs matter for recruitment because a flight-risk model that flags 18 at-risk employees in a critical function generates a sourcing mandate before resignation letters arrive. Recruiters who can act on predictive signals fill roles 40 to 50 days faster than those waiting for formal requisitions.

Why It Matters for Recruitment

Staffing agencies and in-house recruiters are evaluated on metrics that are products of HR analytics: time-to-fill, cost-per-hire, quality-of-hire, and [offer acceptance rate](/glossary/offer-acceptance-rate). Understanding how those metrics are calculated: and what data feeds them: is necessary to defend performance, challenge unfair comparisons, and identify where process changes will move the numbers.

Time-to-fill, for example, is sometimes measured from requisition open date to offer acceptance and sometimes from offer acceptance to start date. An agency being benchmarked against an internal team using different definitions will consistently look worse without any underlying performance difference. Asking how a metric is defined before agreeing to SLAs is a basic analytics fluency skill that protects margins.

HR analytics also determines how agency performance is scored against preferred supplier agreements. Clients running analytics platforms generate quarterly scorecards showing fill rate, time-to-fill, retention at 6 months, and manager satisfaction by supplier. Agencies that proactively share their own data: candidate source breakdown, interview conversion rates, placement retention: contribute to a shared analytics picture and are harder to dislodge in supplier reviews. Agencies that treat placement data as confidential generate suspicion and often lose panel positions at annual review.

In Practice

A technology company carries out a quarterly talent analytics review. Their Visier dashboard shows that software engineers hired through job board sourcing have 14-month median tenure, while those sourced through specialist staffing agencies have 23-month median tenure. The cost-per-hire via agency (including fees) is £18,500 versus £9,200 via job boards. However, the fully-loaded cost of a replacement hire: including 4 months of reduced team productivity, re-onboarding, and recruiter time: is calculated at £41,000 per departure. The analytics team presents a business case: increasing agency-sourced hires from 20% to 40% of tech hiring reduces annualised replacement costs by £380,000 at current headcount, more than covering the incremental agency fees. The CPO approves a shift in sourcing strategy. Without the analytics model, the decision would have defaulted to "reduce agency spend" based on headline cost-per-hire alone.

Key Facts