Free Guide for CFOs & CHROs
The Headcount Efficiency Playbook
How to use performance data to make smarter headcount decisions
Most headcount decisions are made with bad data. Managers protect their favorites. Quiet contributors lose. Low performers stay. This playbook shows you how to do it differently.
The Headcount Efficiency Playbook
For CFOs & CHROs
Why headcount decisions go wrong
The post-layoff playbook is broken. Companies cut by number, not by performance data. The wrong people leave.
When companies need to cut headcount, they usually start with a number: 10%, 15%, 20%. Then they ask managers to decide who stays.
Managers work with what they have: visibility, recency, personal rapport. They protect the people they know best. They let go of the people who were quieter, more remote, or contributing to work the manager couldn't see.
The result: companies often cut their quiet contributors — the people doing foundational work across teams — and keep the people who are good at managing up. Within a year, productivity is lower than before the reduction. The wrong people left.
The problem isn't the decision to cut. The problem is making the cut without performance data.
What headcount decisions look like without data
Managers rate employees who are visible to them more highly. Remote workers and quiet contributors lose out — not because their performance is worse, but because they're harder to see.
Employees who communicate their work up are rated higher than employees who let their work speak for itself. Self-promotion skill and performance don't correlate well.
Employees with strong manager relationships are protected in reductions. Employees without that relationship aren't — regardless of their contribution to the organization.
The real cost of keeping low performers
Salary is the floor. The true cost includes everything their presence does to the people around them.
performance gap between top and bottom quartile. Low performers don't just underperform — they generate review work, corrections, and workarounds for everyone around them.
more manager time spent on low performers than high ones. In a company where manager time is expensive, this compounds fast.
of annual salary when you include salary, benefits, manager time, morale impact, and productivity drag. For a $120K employee, that's $180K–$300K per year.
of annual salary to replace a top performer who leaves. When top performers watch low performers get protected, they start looking elsewhere. Keep one low performer, lose one top performer: a $400K–600K mistake.
Why you don't know who your top performers are
Most companies think they know. They're wrong about at least 30% of the list.
of what employees actually do is visible to their direct manager. The rest — cross-functional contributions, mentoring, foundational work — is invisible to the person writing their review.
Who gets rated highly under traditional reviews
- Employees who are visible to their manager
- People who communicate their work up regularly
- Employees in high-visibility roles
- People whose work lands in front of the manager at review time
Who gets underrated
- Remote workers and employees on distributed teams
- Quiet contributors who let their work speak for itself
- People doing cross-functional work their manager doesn't see
- Employees working on foundational projects with delayed outputs
The quiet contributor
Does exceptional work without self-promotion. Frequently nominated by peers. Rarely visible to management. In a headcount reduction, they're at risk — even though they're exactly who you need to keep.
The organizational connector
Bridges teams. Unstalls cross-functional projects. Highly trusted by people across the organization. Invisible on the org chart. Their departure drops team performance for months — and you may not know why.
The visible low performer
Communicates well, presents confidently, manages up effectively. Gets strong manager ratings. The organization, if asked, would tell a different story. This is who gets protected in reductions that rely on manager judgment alone.
The disengaged employee
Peers rarely nominate them for anything. Works alone. May be disengaged, struggling, or stuck. Often invisible to management too — until the situation becomes a formal PIP or a surprise departure.
The performance data you actually need
Most companies have some of this. Few have all four. All four are necessary for good headcount decisions.
Organizational network data
Who does the rest of the organization trust, learn from, and identify as a contributor? ONA collects structured nominations from every employee — crossing org chart boundaries to surface who is actually driving outcomes. 90 seconds per person. The resulting network map reveals more about real organizational performance than most full review cycles.
Calibrated manager ratings
Raw manager ratings are noisy. Different managers have different standards. A "Meets Expectations" from a tough rater is not the same as a "Meets Expectations" from a lenient one. Before using any manager ratings for headcount decisions, they need to be calibrated: adjusted for each manager's rating distribution, compared against peers, and reviewed in group calibration sessions.
Multi-cycle trend data
A single rating cycle tells you performance in a window of time. Two or three cycles tell you trajectory. Who is growing? Who has plateaued? Who is declining? Trend data is often more predictive than any single-cycle rating — and matters especially for investment decisions: where do you put development resources?
Role fit data
Headcount decisions aren't always about absolute performance. As companies change, the roles that matter change. Someone well-suited to an earlier-stage company may not fit the role the company needs now — not because their performance declined, but because the job changed under them. Good headcount decisions distinguish between performance problems and role fit problems.
A performance-based headcount process
Five steps from data collection to defensible decisions.
Run ONA before assessments start
Collect organizational network data before managers write assessments — so ONA captures clean signals, not signals anchored to management opinion. Every employee answers three questions in 90 seconds. The network data is available for calibration.
Calibrate ratings before you use them
Bring department leaders together to review rating distributions across their groups. Surface lenient raters, tough raters, and systematic inconsistencies. Agree on a calibrated distribution before any rating is used for a headcount decision.
Cross-reference manager ratings with ONA
Compare what managers say with what the organization says. Flag two patterns: employees the organization rates highly that managers don't (possible quiet contributors being undervalued) and employees managers rate highly that the organization doesn't (possible visibility-performance gap worth examining).
Apply trend data to prioritize
Segment your workforce by performance and trajectory. Strong performance plus positive trend: strong retention priority. Average performance plus declining trend: structured support or transition plan. Low performance plus flat trend: assess role fit first, then decide.
Document the decision framework before you decide
Write down your criteria before you apply them. What data are you using? How are you weighting it? Who signs off? Documentation builds discipline into the process and protects you legally. Frameworks built before decisions are more defensible than frameworks built to justify decisions already made.
The data layer for headcount decisions
Confirm is performance management software built to surface the four types of data headcount decisions require.
ONA in every review cycle
Confirm runs ONA as part of every performance review — not as a separate tool or process. Network data is available for every calibration session. No integration required. No separate survey to run.
Calibration built in
Confirm's calibration module surfaces rating distributions across managers, flags systematic differences, and maintains the calibrated record. HR teams see where manager ratings cluster — and can run calibration sessions inside the platform.
Multi-cycle trend tracking
Confirm tracks performance data across review cycles. HR and executive teams can see trajectory for any employee — who is growing, who has plateaued, who has changed over the past two to three cycles.
ONA-manager rating comparison
Confirm shows where manager ratings and ONA signals diverge significantly — and flags those gaps for calibration review. Quiet contributors get surfaced before they become attrition statistics.
"With ONA data, not only did we identify our top performers, we retained 100% of them."Julia Psitos, Head of People, Thoropass
"Hands down, the organizational network analysis approach is the best I've seen for identifying who is actually making an impact."Joe Bast, VP People
"Confirm is the first tool that lets me see the behavioral side with holistic evidence — not just what managers report."Joanna Yeoh, VP People
What's inside the playbook
12 pages. No filler. Free with a demo request.
Why headcount decisions go wrong
The structural flaws in how most companies make headcount decisions — and why the wrong people consistently end up leaving.
The full math on low performers
Salary, manager time, morale drag, and productivity cost — with a worked example for a mid-market organization.
The visibility problem
Why 70-80% of employee contribution is invisible to direct managers in modern organizations — and which archetypes are most at risk.
The four data types you need
Organizational network data, calibrated ratings, trend data, and role fit — what each tells you and where to get it.
The five-step headcount process
A practical process for using performance data in headcount decisions — from ONA collection to documented decision criteria.
30-day action plan
Week-by-week steps to build a performance-based headcount process before your next review cycle.
Key metrics to track
Headcount efficiency and process health metrics — what to measure, what benchmarks to use, and what to do when metrics move.
Legal and documentation requirements
How to document headcount decisions to protect your company — and why performance-based criteria matter for legal defensibility.
Free with a Confirm demo request. No spam. Instant access.
Frequently asked questions
What is headcount efficiency?
Headcount efficiency is how much value each employee generates relative to their cost. A headcount-efficient organization retains its highest contributors, addresses low performance systematically, and makes headcount decisions based on data rather than visibility or politics. Improving headcount efficiency means getting more output per dollar of salary — without necessarily cutting headcount.
How do you identify low performers before a reduction in force?
Identifying low performers before a RIF requires more than manager ratings. The most reliable process combines calibrated manager assessments (adjusted for each manager's rating tendencies), organizational network data (who peers identify as contributors), and trend data across multiple review cycles. Single-cycle manager ratings are too noisy for defensible headcount decisions — they reflect visibility as much as performance.
What is the real cost of keeping a low performer?
The full cost typically runs 1.5–2.5x annual salary when you account for salary and benefits, manager time spent on remediation, team morale impact, and productivity drag. Research from McKinsey puts the performance gap between top and bottom quartile performers at 4x in complex roles and up to 8x in creative roles. The cost of keeping a low performer in a key role is substantial and often invisible to finance.
How does organizational network analysis help headcount decisions?
ONA collects structured data about who employees trust, learn from, and identify as contributors. When you cross-reference ONA signals with manager ratings, you surface two critical patterns: employees the organization trusts but managers underrate (quiet contributors at risk in a headcount reduction) and employees managers rate highly but peers rarely nominate (visible to management but less impactful organizationally). These gaps often reveal where the real performance risk is.
What performance data do you need for headcount decisions?
Smart headcount decisions require four types of data: calibrated manager ratings (not raw ratings, adjusted for each manager's distribution tendency), organizational network data (who peers across the company identify as contributors), trend data across two to three review cycles (who is growing, plateauing, or declining), and role fit data (whether the employee's skills match the role the company needs). Most companies have some of this data; few have all four.
See how Confirm surfaces headcount performance data
Book a 30-minute demo. We'll show you exactly what ONA data looks like in your organization, how calibration works, and what headcount decisions look like with the full data set.
Used by CHROs and CFOs at high-growth companies. No commitment required.