Short answer
A continuous performance review is a managed cycle of expectations, recent evidence, feedback, support, and follow-through. It does not mean watching employees continuously or repeating a full appraisal every month. Set a sustainable baseline cadence, add clear event triggers, define the purpose of each discussion, and keep formal decisions in their governed process.
Define continuous as a cycle, not surveillance
Continuous performance management should make relevant conversations timely. It should not create an always-on stream of sentiment, activity, or risk scores about individuals.
The CIPD performance management factsheet describes performance management as a group of practices rather than a single activity, with regular, timely discussion and objectives that can evolve with business priorities. The US Office of Personnel Management cycle separates planning, monitoring, developing, rating, and rewarding. Its federal context is not a universal employment model, but the separation is a useful design check.
For each part of your cycle, state its job:
- Planning: confirm current expectations, evidence, dependencies, and decision rights.
- Work check-in: examine progress, blockers, support, and changed priorities.
- Feedback: discuss a recent observable example while it can still inform the work.
- Development: agree an opportunity, resource, practice, or instruction.
- Periodic review: summarise a defined period and support any formal record or decision.
- Recognition: acknowledge a specific contribution close to the work.
A check-in should not silently turn into a rating meeting. Tell the employee whether a discussion is operational, developmental, or part of a formal process.
Build a cadence from fixed moments and triggers
Start with the work rather than a universal weekly or monthly rule.
| Moment | Trigger | Purpose | Minimum record | Owner |
|---|---|---|---|---|
| Role and priority reset | Start of role, project, or material scope change | Confirm what good work means now | Expectations, evidence, dependencies, effective date | Manager and employee |
| Regular work check-in | Predictable interval suited to the role | Remove a blocker and review prior actions | One agreed action, owner, return date | Manager |
| Recent-work feedback | Milestone, handover, customer case, or observed event | Discuss a specific example while context is available | Observation, employee context, next step if needed | Person giving feedback |
| Development discussion | Employee request, role need, or planned cycle | Match a development need with a real opportunity and support | Action, resource, sponsor, review date | Manager and employee |
| Formal review | Policy, contract, agreement, or decision window | Summarise the period and support a governed decision | Required assessment, employee response, approvals | Formal decision owner |
| Follow-up | Agreed date or unresolved action | Check whether the action happened and what remains | Result, evidence, revised owner or escalation | Action owner |
The fixed baseline creates predictability. Triggers prevent a team from waiting for the calendar when priorities change, a project ends, support fails to arrive, or an expectation needs correction. Neither requires automatic monitoring.
The CIPD performance review factsheet recognises the move towards regular conversations while retaining structured reviews as part of the wider cycle. Acas guidance similarly places informal feedback, coaching, and check-ins alongside more formal reviews and recommends a written record shared with the employee.
Give every check-in a small contract
A sustainable check-in can use one page:
- Purpose: What work or support question are we discussing?
- Current expectation: What outcome, behaviour, or responsibility was communicated?
- Recent evidence: Which specific example can both people examine?
- Employee context: What changed, helped, or got in the way?
- Manager responsibility: Which decision, resource, clarification, or escalation is needed?
- Employee action: What is reasonably within the employee's control?
- Follow-through: Who will do what, by when, and when will both people return to it?
- Boundary: Does any issue need a separate HR, employee-relations, safeguarding, or formal process?
Use the effective performance conversation guide for the meeting itself. This guide addresses the cadence and accountability around those conversations.
Keep evidence current and correctable
Ongoing notes can reduce year-end reconstruction, but more notes do not automatically create fair evidence. Records need purpose and discipline.
- Record observable work and dated examples rather than personality labels.
- State which expectation applied at the time and when it changed.
- Let the employee add context and correct an inaccurate record.
- Keep disagreement visible instead of forcing false consensus.
- Separate a work blocker from an individual capability conclusion.
- Define who can access each record and how long it is needed.
- Do not reuse developmental or confidential material for an undisclosed decision.
For US employers, the EEOC's small-business evaluation guidance recommends communicating standards, applying them consistently, explaining whether they were met, using relevant facts, and informing employees when standards change. Organisations in the US and UK still need to apply their own policies, agreements, and relevant legal advice.
Make follow-through part of the cadence
A meeting is not complete because it happened. At the return date, check the action itself.
| Prior agreement | Follow-up question | Possible outcome |
|---|---|---|
| Manager will clarify a priority | Was the new expectation shared and understood? | Close, correct, or escalate the clarification |
| Operations will remove a blocker | Was the dependency changed, and what evidence shows it? | Close or reassign to the accountable owner |
| Employee will practise a task | Was time, instruction, and a suitable opportunity provided? | Review the work example, not course attendance alone |
| Team will test a process | What did the test show, including counterexamples? | Adopt, revise, or stop the test |
| Evidence remains disputed | What additional source can reasonably check it? | Record both views and the next decision route |
Do not carry an overdue management action forward as if the employee failed to complete it. Name the owner and make escalation possible.
A fictional cadence for a US and UK services team
Northline Advisory is a fictional services business with project teams in the US and UK. Its quarterly form asks managers to rate delivery, collaboration, and growth, but actions are rarely revisited. Some teams already meet weekly; others work in six-week project cycles.
HR does not impose another weekly meeting. It sets a common operating rule and lets teams fit it to the work:
- Expectations are confirmed at project start and whenever scope materially changes.
- A short support check-in occurs at least once during each six-week project cycle.
- Recent-work feedback is triggered at a major handover or client milestone.
- Development has a separate discussion every quarter because available assignments require planning.
- The annual review remains the documented summary used in the existing pay and progression process.
- Any action from a check-in receives an owner and return date.
On one pilot project, an employee raises repeated late changes to the client brief. The manager records the changed requirement and agrees to ask the project lead for a freeze point. At the next check-in, the team checks whether the freeze point was used. It does not label the employee's slower delivery as weak performance before examining the dependency.
After two project cycles, HR reviews whether conversations occurred at the intended moments, actions were revisited, employees could correct records, and managers could sustain the effort. It does not claim that completion or a changed delivery measure proves the cadence caused better performance.
Test whether the cadence is sustainable
Before scaling, inspect one complete cycle:
| Test | Evidence |
|---|---|
| Purpose clarity | Employees can distinguish a support check-in from a formal review |
| Coverage | Roles, shifts, locations, and leave cases have a workable path |
| Manager capacity | The tested effort fits workload without unpaid or hidden work |
| Current expectations | Material changes are dated and visible |
| Employee voice | People can prepare, add context, and record disagreement |
| Follow-through | Actions have owners and return dates, including manager actions |
| Formal boundary | Pay, promotion, capability, and other decisions stay in their approved processes |
| Data access | Employee, manager, HR, reviewer, and administrator views match the stated purpose |
| Quality | A sample contains specific evidence and usable actions, not copied text |
Revise the cadence if routine meetings are empty, managers duplicate records, employees cannot tell how notes will be used, or actions disappear between cycles. More frequent activity is not the goal.
If a defined work issue needs employee context before a manager discussion, explore Lontra's employee conversations and manager briefs. Lontra can help prepare that conversation; managers receive a brief rather than raw employee conversations. It is not a system for continuous employee monitoring, goal ownership, ratings, or automated employment decisions.
Frequently asked questions
What is a continuous performance review?
A continuous performance review is an ongoing management cycle that keeps expectations, recent evidence, support, feedback and follow-through current. It can include regular check-ins and event-triggered discussions alongside a formal periodic review.
Does continuous performance management remove annual reviews?
No. An annual or other periodic review can remain useful for a documented summary, employee response, calibration or a formal decision. Ongoing check-ins should prevent that meeting from becoming the first discussion of an issue.
How often should performance check-ins happen?
Choose a cadence that fits the work, decision cycle and manager capacity. Combine a predictable baseline with clear triggers for changed priorities, completed work, blocked actions or support needs, then test the workload in a pilot.