OKR Check-In Templates · 7 min read
The Monthly OKR Check-In: A Template for Trajectory, Not Deltas
I've run OKRs for about a decade, and the monthly check-in is the one I see people get wrong most often. They run it like a longer weekly. It isn't. The weekly is about what moved and what's blocked. The monthly is about whether you still believe the target, and that's a completely different conversation. This template is built for that one.
By Max Bondarenko · Last updated June 2026
The template, up front
No buildup. Here's the whole thing. A monthly OKR check-in answers five questions per key result, and the last one is a decision, not a status. If you only copy one part of this post, copy these five and the confidence rule underneath. Confidence is a number from 1 to 10 that says how sure you are you'll hit the target by quarter-end. Track it every month so you can see the line move.
Five questions for every key result, monthly
- 01Where's the number now, where does it need to land, and how confident am I we get there (a single score, 1 to 10)?
- 02Which direction has my confidence moved over the last 4 weeks: up, flat, or down?
- 03What did the last month actually teach me about whether this target was the right one?
- 04Is the gap between today and the target closable with the time and people left in the quarter?
- 05Decision: do I persevere, change the work, or revise the target? Pick one and say why.
Why each question is there
The confidence score is the spine of the monthly. A weekly check-in tells you a metric went from a 22% baseline to 24% against a 40% target. Fine. But a number can climb while your confidence drops, because you learned the channel driving it is about to dry up. Confidence captures judgment the raw metric can't. I score it 1 to 10, force a reason in one sentence, and I don't let anyone park at 5. Five means "I haven't thought about it." Make them commit to a 4 or a 6.
The trend question matters more than the score itself. One reading is a snapshot; four weeks of readings is a story. A KR sitting at confidence 6 that climbed from 3 is healthy. A KR at 6 that fell from 9 is the one I want on the table, even though the number looks the same. The monthly is the first place you can see that slope, because a single week is too noisy to trust.
The decision question is the whole point, and it's where I've watched the most value get left on the floor. Years back I ran a team that set a target it became obvious was wrong by about week 4. The market had shifted under us. Everyone could feel it, nobody said it, and we spent eight more weeks grinding toward a number that no longer meant anything. If we'd had a real monthly with a forced persevere-or-revise call, we'd have caught it at the first review. Revise a wrong target early, in week 4, not week 12. A target you change in month one is a course correction. The same change in the last week is just a confession. Same edit, totally different cost.
How to score confidence
Keep the scale dead simple. Confidence is your honest read on hitting the target by quarter-end, 1 to 10, and the only thing that matters at the monthly is which band you land in and which way the line is moving. Here's how I read the three states and what I do about each.
| On track (7 to 10) | You believe the target lands with the time and people left. Action: leave it alone. Don't reallocate toward it, don't over-discuss it. Protect the focus for the KRs that need it. |
|---|---|
| Behind (4 to 6) | Reachable, but not on the current path. The work, not the target, is the problem. Action: change the work this month. Reprioritize, pull in help, or cut a competing commitment. Recheck in 4 weeks. |
| At risk (1 to 3) | You no longer believe the target, or the trend has fallen two months running. Action: make the call now. Either commit real resources to a turnaround or revise the target down to something honest. Don't let it limp. |
Reading the confidence score at a monthly check-in
How to actually run it
Run the data part async by default. Every KR owner writes their five answers in a shared doc two days before you meet, score and reason included. No status read-aloud in the room. If the meeting is people taking turns narrating numbers everyone could've read, you've rebuilt the weekly and wasted the month. The async write-up is the check-in; the meeting is only for the decisions that need more than one head.
Keep the live session to 30 minutes and invite only the KR owners plus whoever can actually reallocate people or money. You don't need the whole org watching. Walk only the KRs at "Behind" or "At risk," because the "On track" ones don't need air. And skip the monthly entirely in the first two weeks of a quarter and the last week. Early on there's no trend to read yet; at the very end the decision window has closed and you're just scoring for the record. The monthly earns its keep in the middle of the quarter, which is exactly when most teams stop paying attention.
Questions people actually ask
How is a monthly OKR check-in different from a weekly one?
The weekly is about motion: what moved, what's blocked, what's next. The monthly zooms out to trajectory and confidence, and it forces a decision the weekly never asks for. You can't read a real trend off one week of data, so the monthly is the first point where you can see whether your confidence is climbing or sliding.
What confidence scale should I use?
I use 1 to 10, where the score is how sure you are you'll hit the target by quarter-end. Pair every score with a one-sentence reason and don't let people sit at 5, since 5 usually means they haven't thought it through. The exact scale matters less than tracking the same one every month so the trend is honest.
When is it right to revise a target instead of pushing harder?
When you no longer believe it, or the confidence trend has dropped two months running. Revise early, in week 4 rather than week 12, because a target you change in month one is a course correction and the same change at the end is just a confession. Pushing harder only makes sense when the target is still reachable and it's the work that's off, not the goal.
Should the monthly check-in be a meeting or async?
Async by default. KR owners write their five answers and confidence scores two days ahead, and the live 30-minute session covers only the KRs at Behind or At risk. If the meeting is people reading numbers aloud that everyone could have read in the doc, you've turned the monthly into a slow weekly and lost the point.
Stop running check-ins in a spreadsheet
Okiar is free during beta. Voice check-ins, AI projections, team health — live in minutes.
Start free →