Performance reviews end in the archive—without goals, process, and outcomes closed, appraisal becomes form-filling

At the end-of-year meeting, the goals column is still empty.

The HR director opened the evaluation page for this cycle and found that nearly one-third of the middle managers still had goals written as 'actively cooperate' or 'enhance communication.' The manager explained, 'We'll deal with it after things settle down, make up for it at the end of the year.' Employees clearly understood: the goals that are added later will surely match the results that have already occurred — appraisals no longer guide work, they just provide a rationale for the scores.

In manufacturing companies, this formalism rarely manifests as 'employees slacking off'; more often, it's the management chain breaking: organizational goals aren't broken down, processes lack review checkpoints, results neither affect compensation nor lead to improvements, and the optimal strategy for rational employees becomes 'don't offend anyone, don't spend time.' The more polished the forms, the thinner the trust.

There is also symmetrical harm: those who seriously write down their goals and keep evidence, if they suffer losses in the application of results, will have the entire team return to Excel in the next round. Whether the loop is closed is not judged by the completion rate of the forms, but by whether the 'serious people suffered.'

How to play the fill-in-the-blank game

Randomly select five employees and ask their supervisors: What are the three most important goals this cycle? What was discussed in the most recent review? How did the results affect salary or development? If they can't answer, performance is still just filling out forms—don't blame the employees for 'not cooperating with digitalization'.

If attitude-related items are retained, they must be linked to observable behavior anchors (for example, 'weekly meeting materials submitted on time' or 'customer complaints responded to within 24 hours'); otherwise, they should be deleted. When organizational responsibilities are unclear, objectives will inevitably be empty: if you don't know who you are accountable to, you can only write 'actively cooperate.' Therefore, a performance feedback loop usually requires a clear relationship between organizational roles and evaluations as a prerequisite—during the reassignment season, if evaluators are still based on the old managers, goals and evidence will be misaligned.

Four nodes: Goal, Review, Calibration, Application

A closed loop is not just adding another score sheet, but ensuring that all four nodes occur and are traceable:

Goal setting. Aligned with organizational goals, measurable, and time-bound. Setting them too late is equivalent to being ineffective; major environmental changes allow for adjustments, but they must be documented, and secretly changing the criteria at the end of the period is prohibited. Managerial goals should come from being broken down by superiors, and then passed down to the team — no matter how well individual goals are, if they are disconnected from the organization, they are still only locally optimal.

Process review. Manufacturing and project-based teams can have light monthly or bi-monthly reviews (progress, risks, and whether goals have changed); functional teams can review once a quarter. Reviews should leave brief records, rather than suddenly reckoning everything at the end. Evaluations without process records can only lead to arguments based on impressions at the calibration meetings.

Evaluation and calibration. Calibrating addresses the issue of 'inconsistent departmental standards': for the same performance, lenient departments give high scores to everyone, while strict departments suppress top performers. Calibration meetings are not about reassigning favor points, but about asking three questions based on evidence — how were the goals originally set, what factual events occurred during the process, and whether peer comparisons are obviously unbalanced. Any adjustments must be recorded: who proposed them, on what basis, and how the result version changed. Results that haven't been calibrated shall not enter the floating salary or promotion/demotion lists — this is a strict safeguard against form-filling games. Forced distribution can be used as a reference; without supporting evidence, the curve itself can destroy the trust of high performers.

Result application. Performance ratings are mapped to variable pay/year-end bonus coefficients, promotion eligibility, and the triggers for improvement plans; the mapping rules are announced in advance, so employees are willing to take goal-setting seriously. Low scores should have improvement plans and review checkpoints, not just punishment. If there is no application cycle, it is better to shorten the scope and conduct a deep pilot.

Project-based employees: Project managers input contribution evidence (delivery quality, milestones, collaboration), and line managers make a comprehensive assessment; the weights are configured in advance to avoid 'more hours = better performance' or 'the loudest person gets their way'.

Indicator database: multiple plans can branch out, language must be unified

Sales season awards, completion of probation, annual comprehensive review, factory team assessments—it's normal to have multiple schemes coexisting. Increased formalization often happens because each scheme is managed separately in Excel, resulting in an inability to integrate them into the EHR payroll system. The correct approach is to share a common indicator database (definitions, data extraction instructions, behavioral anchors), with process templates branching where needed; calibration rules should be shared. If a department privately maintains its own scheme that cannot be integrated into payroll, it is equivalent to encouraging a dual-track reality.

How to read the indicator and whether a closed loop has occurred:

The system can summarize business data and reminder points, but evaluations and interviews cannot be omitted — automation is suitable for data retrieval, not for replacing judgment. When business fluctuations are large, use interval goals and key result indicators, with process indicators as support; a single external shock that ruins the annual KPI will only force end-of-year backdating.

The position of JEDP-EHR performance in the link

JEDP-EHR performance evaluation emphasizes: indicator pool, goal alignment, process review, calibration, and applying results to variable pay/development; uncalibrated results do not affect salary to prevent evaluations from becoming end-of-year form-filling and results that cannot be explained. HR provides evaluation relationships (inherited with position); salary calculations only use the calibrated official results version. If inventories and talent maps are enabled, results do not automatically change salary or permissions—the position, pay, and permissions still follow organizational and EHR processes.

Boundaries with organization, attendance, and payroll: The scope of evaluating people and data changes with the position; transfer of position switches according to the effective date to avoid old managers entering new positions; if attendance discipline affects performance, the data must come from confirmed attendance results, and no separate table should be made; the floating salary input should only read the official performance results, and changing scores through WeChat to affect salary is prohibited. Phase one recommends piloting with a small group (such as middle management plus one factory) to run through 'Target—Review—Calibration—Application'; phase two will involve all employees and multiple template plans; phase three will address a more detailed integration with talent review.

With an extremely small team and almost no variable incentives, the benefits of this special topic are limited; but once the performance coefficient is included in year-end bonuses or promotion thresholds, the closed loop becomes a must-answer question. When managers have not received interview training, the system will only push them to complete forms—training and feedback first, then roll out to all employees.

Factory Pilot: First Prove 'Serious People Don't Suffer Losses'

Choose a factory or a functional line, don't have the whole company fill out forms on the first day:

1. Clean up attitude words, establish an indicator library v1 (measurable results, a few behavioral anchors).
2. Alignment will complete goal setting, and review days should be written into the calendar; if expired and not locked, entry into the evaluation is not allowed.
3. Run through at least one round of process review and develop the habit of attaching evidence (such as project evaluations or business data references).
4. Hold a calibration meeting after evaluation to adjust the records; batches that have not been calibrated shall not have compensation fluctuation items pushed.
5. Interviews and improvement plans are closed, and the trial-side verification of the 'level → coefficient' mapping is readable.

The target achievement rate within two weeks is still very low, so first stop expanding the scope. What is being checked is whether the organizational goals have been broken down and whether managers can write measurement methods—not just sending out score sheet templates again. If the calibration meeting turns into allocating quotas and refuses to leave a trace, first rebuild the evidence discipline, then open up results for salary adjustment; otherwise, the system will only perpetuate unfairness. When the compensation linkage policy has not yet been approved, you can first close the loop on goals and calibration, but it must be clearly communicated that it will not lead to salary adjustments for now, to avoid managing expectations poorly.

Before a full rollout, use pilot projects to prove: those who set goals carefully and keep evidence will not be disadvantaged when results are applied. Without achieving this, any EHR module is just a more expensive form.

Focus on three targets this week first

There's no need to wait for the new module to go live. Pick a department and require them to complete the 'Three Goals Measurement Method Next Review Date' within this week. Do not use unobservable words like 'proactive'—this is the smallest viable step to close the loop. Putting the review date in the calendar is more useful than sending an extra set of score sheets.

For calibration results without traceability, do not push the compensation fluctuation items at all. If not applied, shorten the cycle and go deeper to avoid starting with a blanket pseudo-closed-loop for everyone. When organizational appointments are still unclear, first supplement the evaluation relationships and effective dates of job transfers before discussing performance digitization; if the evaluator is chosen incorrectly, all subsequent evidence will be attached to the wrong person.

• EHR HR solution
• Compensation data is reliable
• Organizational Permission Inheritance
• Attendance anomaly