“How long should a performance review cycle take?” has two answers, and people often mix them up. One is elapsed time: the calendar weeks from launching the cycle to the last feedback conversation. The other is effort: the hours managers, employees and calibrators spend inside that window. A good cycle is short on both. This guide gives a practical target for each, explains what drives them, and shows how to measure your own.
To put numbers on your current cycle, use the performance review cost and cycle-time calculator. It takes about a minute and uses only your own inputs.
The short answer
Following the quarterly process in the performance management framework, a well-run cycle takes about three to four weeks from launch to the last feedback conversation:
| Stage | Playbook guidance |
|---|---|
| Manager reviews (with self-assessments running alongside) | About 1 to 2 weeks |
| Grade calculation, calibration and leadership review | About 1 week at most |
| Announce results and deliver feedback in 1:1s | Within about a week of sign-off |
Add about two weeks of set-up beforehand (scope, dates, scorecards, manager briefing). These are the playbook’s own timings and practical starting points, not an industry benchmark. There is no reliable public benchmark for cycle length at mid-sized companies, so the most useful comparison is with your own previous cycles.
Why cycle length matters
Long cycles cost more than time. While a cycle is open, decisions on promotion, pay and underperformance wait. Evidence gets older. Managers context-switch between review writing and their actual jobs for weeks. Employees who are told results are “coming soon” for two months start to see the process as a formality.
The effort side can be very large. Buckingham and Goodall’s “Reinventing Performance Management” (HBR, 2015) describes how Deloitte found its process consumed close to 2 million hours a year across forms, meetings and ratings. McKinsey’s 2024 article “In the spotlight: Performance management that puts people first” notes that annual reviews can create a bottleneck on managers and the C-suite, and that more regular performance conversations can supplement formal reviews.
How often should you run a cycle?
Frequency and length are linked. If you run one cycle a year, each cycle has to cover twelve months, so it tends to be heavy and slow. The playbook recommends quarterly reviews, each kept light. Many companies have moved in this direction: Cappelli and Tavis’s “The Performance Management Revolution” (HBR, 2016) describes firms including Adobe, Deloitte and General Electric moving away from traditional annual appraisals towards more frequent conversations.
| Cadence | Strengths | Risks |
|---|---|---|
| Annual | Low frequency of admin | Recency bias; slow decisions; heavy cycles |
| Twice a year | A common compromise for growing companies | Still relies on six months of memory |
| Quarterly | Fresh evidence; aligns with quarterly goals; faster decisions | Only works if each cycle is light |
A quarterly cycle at four weeks means roughly four months of the year have a cycle in progress. That is sustainable only if the effort inside each cycle is small: short structured self-assessments, one scorecard per person and focused calibration.
For how weekly check-ins, quarterly scorecards, formal reviews and pay decisions fit together, see how often performance reviews should be conducted.
What makes cycles slow
| Driver | Effect | Fix |
|---|---|---|
| Evidence scattered across tools | Each review starts from scratch | Keep goals, feedback and 1:1 notes in one place through the period |
| Long, open-ended forms | Self-assessments become essays | Three or four structured prompts tied to goals and scorecards |
| No firm deadlines | Missing scorecards delay calibration | Fixed deadline several working days before calibration; daily chasing |
| Calibrating everyone | Sessions overrun and get rescheduled | Focus on top, bottom and outliers with a template |
| Leadership sign-off booked late | Results wait for a calendar slot | Book the sign-off when you set the cycle dates |
| No written standard | Calibration becomes debate | Publish the talent bar before the cycle |
How to measure your own cycle
- Find last cycle’s actual dates. Launch, scorecard deadline, calibration sessions, sign-off, last feedback conversation.
- Ask three managers for real hours. Time per review, including evidence gathering.
- Count calibration time. Sessions × length × attendees.
- Run the calculator. Enter these into the cost and cycle-time calculator to see total hours, hours per manager and what elapsed time would look like with published customer results applied. For example, Sertis cut its self-review cycle from a month to a week after consolidating its HR systems.
- Find the biggest line. Fix that first.
Worked example
An illustrative 300-person company ran its last half-year cycle over nine weeks: three weeks for manager reviews, two weeks chasing missing scorecards, two weeks of calibration spread across executive calendars, and two weeks of feedback. The changes it makes for the next cycle are modest: a firm scorecard deadline three working days before calibration, calibration sessions booked at set-up, a one-page pre-read per department, and a manager briefing before results go out. The next cycle runs in a little over four weeks. Nothing about the review form changed; the time came out of waiting.
Does company size change the answer?
Less than you might expect. Larger companies need more calibration sessions, but they can run in parallel. What grows with size is coordination: more managers to chase, more sessions to schedule, more people to brief. That is why a fixed timeline with named owners matters more at 200 people than at 50. See the week-by-week timeline for a 200-person company and calibration at 500 people.
How much effort should a cycle take?
Elapsed time is easy to see; effort is easy to miss. Rather than chase a universal number, set an effort target for each group and check it after every cycle:
| Who | What drives their time | What a light cycle looks like |
|---|---|---|
| Employees | Self-assessment length; hunting for examples | A short, structured self-assessment drawing on goals and notes already recorded |
| Managers | Evidence gathering and writing, per direct report | Reading evidence that already exists, then filling in one scorecard with a note per area |
| Calibrators | Number of sessions × length × attendees | Focused sessions on top, bottom and outliers, with a pre-read |
| People team | Chasing, merging, building reports | Chasing by exception; pre-reads and reports produced from one system |
The calculator turns these into a single number of hours a year and hours per manager. If moving from two cycles a year to four would double that number, the cycles are too heavy to run quarterly yet; lighten them first.
Week-by-week timeline: a 200-person company
To make the guidance above concrete, here is a full cycle for an illustrative company of about 200 people, following the same quarterly process in the framework.
Assumptions
- About 200 employees reviewed, 20 to 30 people managers, one People partner or People lead running the cycle.
- Written scorecards and a talent bar by seniority already exist for most roles.
- Calibration in three or four sessions (for example, Product and Engineering, Commercial, Operations and G&A), plus a leadership sign-off.
- The cycle covers the previous quarter or half-year.
The timeline, week by week
| When | Step | Owner | Output |
|---|---|---|---|
| Week −2 | Set-up: confirm scope, dates, scorecards and talent bar; name calibrators; brief managers in a 30-minute session | People lead | Published schedule and scope list |
| Week −1 | Managers check goals are up to date; employees reminded to note examples | Managers | Current goals for everyone in scope |
| Week 1, days 1–3 | Self-assessments (short, structured, linked to goals) | Employees | Self-assessments submitted |
| Week 1, day 1 – week 2, day 5 | Manager reviews: scorecards with one-line evidence notes; input from other managers | Managers | Scorecards submitted by deadline |
| Week 2, days 3–5 | Chase missing scorecards; build calibration pre-reads | People lead | Pre-read per session |
| Week 3, days 1–3 | Calibration sessions (90 minutes to 2 hours each) | Department heads, People lead | Calibrated grades and change log |
| Week 3, days 4–5 | Leadership review and sign-off; promotions approved | Executive team | Final grades |
| Week 4, days 1–2 | Manager briefing on results and changed ratings | People lead | Managers ready to deliver |
| Week 4, days 2–5 | Results discussed in 1:1s; next steps recorded | Managers | Every employee has heard their result |
| Week 5 | Retrospective; book next cycle | People lead | Lessons and next dates |
From launch to the last feedback conversation, that is about four weeks, with two weeks of preparation before and a short retrospective after. If your last cycle took much longer, the cost and cycle-time calculator will show what the extra weeks cost in hours.
Who owns what
- People lead: the schedule, communications, chasing, pre-reads, facilitation and the change log.
- Managers: current goals, scorecards with evidence, attending calibration, delivering results.
- Department heads: decisions in their calibration session.
- Executive team: cross-department consistency, promotions and final sign-off.
- Employees: a short self-assessment and examples from the period.
What to communicate, and when
A clear timeline only helps if people know it. At 200 people, five short messages usually cover the cycle:
- Two weeks before launch, to managers: dates, scope, links to the scorecards and talent bar, and the invitation to the 30-minute briefing.
- One week before launch, to everyone: what the cycle covers, when self-assessments open, how long they should take, and when results will be shared.
- Launch day, to everyone: self-assessment link and deadline; to managers, the scorecard deadline and how to request input from other managers.
- Mid-cycle, to managers with gaps: a short, named reminder about missing scorecards, sent daily in the last three days before the deadline.
- After sign-off, to managers: final grades, any changed ratings with reasons, and the date by which every employee should have had their conversation.
Keep each message short and repeat the key dates every time. Most slippage at this size comes from people not knowing a deadline, not from refusing to meet it.
Keeping self-assessments short
Self-assessments are where a 200-person cycle quietly loses the most hours, because the time is spread across everyone. Limit them to three or four prompts tied to goals and the scorecard areas: what you delivered against your goals, one example of each value in action, one area you want to develop, and anything your manager might not know about. Published results such as Cleo’s show self-assessments can take minutes rather than hours when the prompts are structured and the evidence is already to hand.
Where 200-person cycles usually slip
| Slip point | Why it happens | Prevention |
|---|---|---|
| Manager reviews overrun | Evidence is scattered, so each review starts from scratch | Keep goals, feedback and 1:1 notes in one place during the period |
| Missing scorecards at calibration | No firm deadline, or deadline too close to calibration | Deadline at least three working days before the first session; chase daily |
| Calibration runs long | Discussing everyone instead of top, bottom and outliers | Use the calibration template and a pre-read |
| Sign-off waits for the executive calendar | Leadership session booked late | Book it at set-up, two weeks ahead |
| Feedback delayed | Managers unsure how to explain changed ratings | Brief managers before results go out |
Sources: McKinsey, In the spotlight: Performance management that puts people first (2024); CIPD, Performance management factsheet. Timeline and durations are illustrative playbook guidance.
Quick answers
Is two weeks enough for manager reviews? Usually, if evidence is collected during the period. If managers say they need longer, the problem is normally where the evidence lives, not the deadline.
Should calibration ever take longer than a week? In the playbook’s process, calculation, calibration and leadership review together take about a week at most. If yours takes longer, check whether sessions are discussing everyone or waiting for executive calendars.
When should employees hear their results? As soon as possible after sign-off, in their next 1:1, once managers have been briefed on any changed ratings.
Common mistakes
- Shortening the window without lightening the work. Managers will just work evenings.
- Going quarterly with an annual-sized form. Frequency multiplies effort.
- Measuring only elapsed time. A fast cycle that consumes every manager’s week is not a win.
- Comparing with another company’s number. Compare with your own last cycle.
For every step of the cycle in one list, use the performance review checklist for HR managers.
Sources: Buckingham & Goodall, Reinventing Performance Management, HBR (2015); Cappelli & Tavis, The Performance Management Revolution, HBR (2016); McKinsey, In the spotlight (2024). Timings are playbook guidance; examples are illustrative.
- Free tool: Performance review cost and cycle-time calculator
- Method: Performance management framework
- All tools: free performance management tools
More guides
- Year-end performance review: HR run sheet and template
- How to run a calibration meeting for a 500-person startup
- Calibration meeting template for mid-market tech companies
- Performance review cycle timeline for a 200-person company
- How to reduce manager rating bias in performance reviews
- Switching from spreadsheets to performance management software
- OKR and performance review alignment for engineering teams
- Performance review checklist for HR managers
- Goal-setting guide: set, cascade and review goals
- Manager training for performance management
- Employee development plans and career conversations
- All guides