Goals that outlive the cycle
Some goals belong to the year. Some belong to the job. Tying every goal to a review cycle makes the second kind disappear each time a cycle closes.
A review cycle has a start and an end. Most goals do not.
“Bring time-to-hire down to 30 days” is not a first-quarter goal that stops mattering in April. It is part of what the job is for. But if goals only exist inside cycles, that measure gets retyped every cycle, its history resets, and the number you compare against is whatever somebody remembered to carry over.
Two kinds of goal
Taqeem keeps them apart, in the same list.
- Cycle goals belong to one review period. “Ship the new onboarding flow by June.” They open with the cycle, get rated in the review, and close with it.
- Standing goals belong to the person and the position. “Keep first response under four hours.” They live outside any cycle, stay open between them, and carry their history forward.
Both are built the same way — a measurement, a frequency, a target and the threshold that raises a RAG alarm. The only difference is whether a cycle owns them.
What carry-forward actually does
When a cycle opens, the standing goals are already there. Nobody retypes them. The person reviews the targets and spends the conversation on the goals that are genuinely new this period.
When a cycle closes, the review takes a snapshot of every goal it rated: the target, the result, the weight it carried. That snapshot belongs to the closed cycle and does not change afterwards. The standing goal keeps running past it.
So a standing goal holds two histories at once — its own continuous record, and a series of frozen readings, one per cycle. That is what turns “better than last year” from a claim into a fact.

Alarms between reviews
A KPI on a job profile carries a threshold. When the recorded value crosses it, the goal turns amber or red, and the person and their manager see it.
On a cycle goal that mostly matters at review time. On a standing goal it is the entire point: the measure is being watched between reviews, which is the window in which anything can still be done about it. A goal looked at twice a year is a report, not a goal.
Set the threshold where you would actually want a conversation, not at the target. A goal that turns amber the moment it is not perfect is a goal people learn to ignore.
Half our KPIs were red all year and nobody moved on any of them. We had set the alarm at the target.
When to detach a goal
Detaching takes a goal off a cycle so it keeps running on its own. Three cases are worth it.
- The measure outlived the period. It was written as a cycle goal, the cycle is closing, and the number still matters next quarter. Detach it and it carries.
- The role changed mid-cycle. The person moved position. Goals that came with the old job close with it; the ones that belong to the new position detach and follow the person.
- It was never about the period. Someone put a standing measure into a cycle because that was the only form on screen. Detach it once and it stops being retyped.
The case where you should not detach: a goal nobody wants to be rated on. Detaching does not remove accountability, and a standing goal with no owner and no alarm is just a line of text.
A shape that works
For most positions, three to five standing goals and two or three cycle goals is enough. The standing ones come from the job profile’s KPIs, so they are already written. The cycle ones are the things that are true about this year and not the next.
If every goal on a person’s list is a cycle goal, the list is describing a period rather than a job. If every one is standing, nothing about this year is being decided.
Bring one role and the KPIs behind it. We will set the goals up with you on the call.
Book a demoGet your HR right, once.
Bring the org chart you have today. In one call we’ll show you where it breaks, and what it looks like when it’s alive.