.png)
Run review cycles where the training completions, certificates and checklist on-time rates are already on the card, pulled automatically rather than remembered.
Define your departments, the rating scale and the KPIs, then cascade a target downward.
Cards are issued with the reviewer and scale frozen on, and the evidence already assembled.
The employee self-assesses, the manager rates against the same evidence, then decides.
When a cycle is issued, Wikidoc builds an evidence pack for each person out of data it already holds. Nobody compiles it, and nobody can quietly leave a bit out.
Training: assigned modules complete against total, and the average score
Certificates held, and the ones expiring inside the review window
Checklist compliance: the on-time rate for the cycle period, per person
The system will not turn “94% checklist compliance” into a score, because that is an automated employment decision. The number sits beside the rating, and a human chooses the rating.
The scale is always labelled: “4” means nothing on its own, “Exceeds” does
Autosave on every field, because a review is long and losing it is unforgivable
AI drafts the employee's own summary and SMART-checks a KPI.
The same run engine that drives compliance checklists, with a rating scale in place of a tick. The states, the locks and the audit trail are ones your team already trusts.
The employee rates first, and the manager sees what they said while rating
Changes requested hands it back with a reason and unlocks the fields
Nothing is ever deleted: a correction is a new event plus a note
Set a target at org level and push it down. Every KPI names a level, an owner and a measurable target, and a child always sits strictly below its parent.
Four levels: org, department, team and individual
Five metric types, in both directions, since lower-is-better is a real goal
Carry-forward brings a missed goal into the next cycle with its last attempt
Performance data is the most sensitive thing in the product. It is scoped against a real department registry, not a free-text field an employee can type on themselves.
One scope resolver, used by every endpoint and every chart
The reviewer, department, job title and scale are frozen on the card at issue
Nobody decides their own card, and small teams are suppressed on aggregates
Instructor certifications and their expiry dates land on the review card as context, so renewal is a conversation in the cycle rather than a scramble.
Kitchen, bar and front of house each carry their own KPIs under one venue target, so a shift lead is reviewed against the thing they actually control.
Individual KPIs carry forward with last cycle's attempt shown inline, so a goal in its second year is visibly a different conversation.
Protocol compliance per practitioner arrives as an on-time rate for the period, so the review discusses the pattern rather than the last thing that went wrong.
Emily
Small Business Owner
No, and it never will. Turning “94% compliance” into a rating is an automated employment decision, the number sits beside the rating as context, and a human chooses the rating.
No. Every endpoint and every chart goes through one scope resolver against a real department registry, not a free-text field an employee can type on themselves.
The subject and their reviewer, once the card is decided. Both sides see the same evidence pack at the same time, asymmetric evidence is how a review tool becomes a grievance.
Nothing. The reviewer, department, job title and rating scale are frozen onto the card when the cycle is issued, so a mid-cycle move cannot rewrite a review in progress.
Two narrow things: it drafts the employee's own summary as an editable draft, and it SMART-checks a KPI you are writing. It does not write reviews and does not rate anyone.
Never. Performance is the most sensitive data in the product and is visible to members of your organisation within their scope, and to nobody else.
