Should different teams have different OKR cycles?

Short answer: Give the company one shared execution cycle — usually a quarter. Put year-and-longer goals on strategic objectives. Let people check in weekly or monthly with update cadence. Do not stand up a separate live cycle per team.

That is not a limitation we papered over. It is the method.

The request that sounds reasonable

A VP of Engineering wants monthly OKRs. Sales wants the fiscal year. A platform team wants a four-week sprint goal. Leadership wants two or three objectives that last eighteen months. Someone asks: can’t we just create multiple cycles?

You can create as many planned cycles as you want. You cannot run them as parallel town squares.

In Gahmbit, a cycle is the company’s shared clock: who has committed, who is Missing, what parents what, and what the review is about. Only one cycle is active. Coverage, the cascade, the digest, and the retrospective all key off that window. An objective belongs to exactly one cycle. A parent and a child must share it.

If Engineering lives on a monthly cycle and Sales on a yearly one, you do not get flexibility. You get two scoreboards that cannot see each other.

Why a shared clock is the point

OKRs only work as a company practice if the cascade can meet.

Company objectives are supposed to be the thing team objectives align to. Team objectives are supposed to be the thing individual objectives align to. That graph is a lie if the nodes do not share a clock. A monthly team OKR cannot honestly parent to a quarterly company OKR that lives in a different cycle. A yearly Sales objective will not show up when the rest of the company asks “who hasn’t committed this quarter?” Sales will show as Missing — not because they are idle, but because their pledges are on another calendar.

That is the failure mode. It looks like accommodation. It is fragmentation.

Different departments already have different work rhythms: sprints, quota years, launch trains, board packets. Those rhythms are real. They are not a reason to give each department its own strategy calendar. The town square exists so those local rhythms still answer to one set of public commitments.

When every team picks its own cycle length, three things happen:

1. Coverage goes quiet. People with OKRs on the “other” cycle look empty on the active one. Missing stops meaning “hasn’t committed” and starts meaning “on a different clock.”
2. Alignment becomes a screenshot. You can no longer click from a company objective to the team work that is supposed to land it this period. Cross-cycle parents are invalid on purpose.
3. The review dies. There is no single agenda. Engineering is mid-month. Sales is mid-year. Nobody is looking at the same pledges, so the meeting becomes status theater.

A company that cannot name the window it is currently scoring is not running OKRs. It is running a collection of local goal lists.

Cycle, cadence, and horizon are three different jobs

The confusion is usually vocabulary. People say “cadence” when they mean three different things.

Job	Question it answers	Where it lives in Gahmbit
Cycle	What window are we committing and scoring together?	One org cycle. Default quarter. Week / month / year presets change this cycle’s length for the whole company — not for one team.
Update cadence	How often should this owner check in?	On the objective and its key results: weekly, biweekly, or monthly. Freshness and overdue, not goal length.
Horizon	What are we still trying to be true in 18 months?	Strategic objectives. Cycle-independent. Quarterly OKRs align up. They are not scored by rolling up children.

Once those are separate, the “different teams, different lengths” request usually dissolves.

• Sales can still think in years. Write the year as a strategic objective. This quarter’s pipeline, win rate, and ramp sit on the shared cycle and point up.
• Engineering can still ship in sprints. The cycle objective is the quarter’s outcome. Weekly check-ins keep the KRs honest. A four-week push is an objective you close when it is done, still on the quarter — not a side calendar for one team.
• Leadership can still hold two to five long goals of uneven length. That is what strategic objectives are for. They may last a year, or three. Cycle OKRs are how this quarter pays them down.

Shorter than the cycle is not a new cycle. Longer than the cycle is not a new cycle. Faster updates are not a new cycle.

How to set this up

1. Pick the company execution window

Create one cycle. Prefer quarter unless the whole company is truly on another rhythm (a small org that runs monthly OKRs together; a rare annual-only shop).

The week / month / quarter / year control on Create cycle is the length of this org window. It is not “Engineering is weekly.” If you activate a yearly cycle while a quarter is running, the quarter closes. One active cycle is the rule.

Name it so people can say it in a hallway: Q3 2026, not Eng-Aug and Sales-FY27.

2. Write long-horizon goals above the cycle

Add strategic objectives for the company bets that outlive a quarter. Different lengths are fine. They are not attached to a cycle on purpose.

Do not score them by averaging the teams underneath. A strategic objective is graded later by its own long-horizon key results, or it stays unscored — a visible gap, not a borrowed number.

3. Put every team and individual OKR on that same cycle

Company, team, and individual objectives all go in the active cycle. Parent them to a same-cycle company or team objective, or to a strategic objective. If nothing fits, leave them Unaligned and findable. Do not move the orphan to a side calendar to make coverage look full.

Coverage is owner-keyed for this cycle. A person with no live objective and no live key result is Missing. That only stays honest if almost everyone’s live work sits in the same window.

4. Vary check-ins, not calendars

Set update cadence per objective or key result.

• A volatile KR can be weekly.
• The company default can stay biweekly.
• A slow, lagging measure can be monthly.

Overdue then means “this owner missed their check-in,” not “this team is on a different OKR year.”

5. Plan the next window without splitting this one

Create the next quarter as planned. Draft OKRs there. Carry work forward by copying into the new cycle when you are ready — a new row, without last quarter’s check-ins and scores. Move only to fix a mistaken cycle, not to let one objective live in two places.

A planned yearly cycle is a future window, not a parallel department.

6. Close work that finishes early

If a team’s push is done in week four, close that objective on the quarter. The cycle does not have to wait. The rest of the company still shares the window. Starting a weekly cycle for that team would take them off the cascade for the other eight weeks.

What not to do

Do not create one active-feeling cycle per department. Only one can be active. The others become a second, quieter town square. People on the quiet one look Missing on the loud one.

Do not hang one objective across two cycles. An objective belongs to one cycle. Continuing work next quarter means a new (or copied) objective. History stays on the closed cycle, where the retrospective can tell the truth.

Do not treat Teams as a permission wall or a second calendar. Teams are an attribution lens: which group wrote this team-level OKR. They are never scored, and they do not get their own cycle.

Do not paper over Missing because “that org is on the annual plan.” If their quarterly contribution is unwritten, the gap is the point.

A worked example

Northstar Labs runs Q3 2026 (1 Jul–30 Sep) as the active cycle.

• Strategic: “Become the default OKR square for companies that refuse hidden goals” — an 18-month bet. No cycle. No score until someone writes long-horizon KRs on it.
• Company, Q3: “Make coverage obvious in the first session.” Owner: a person. Aligns to the strategic objective.
• Team, Q3 (Product): “Ship the first-cycle checklist without hiding Missing.” Same cycle. Parent is the company objective. The team field names Product. Check-ins weekly.
• Team, Q3 (Sales): “Put 20 design-partner conversations on the board.” Same cycle. Aligns to the same company objective and to a multi-year “category of one” strategic objective. Check-ins monthly — quota is slower than shipping.
• Individual: ICs write Q3 objectives under those team rows. A four-week launch KR can hit 1.0 in August and stay on the quarter as done work.

Sales did not need FY2026 as a second live cycle. Product did not need August as a second live cycle. Both still have the horizon and the check-in frequency they actually needed.

If a team truly cannot share the window

Then you have an org-design problem, not a settings problem.

A holding company with two unrelated products might be two Gahmbit organizations. A single company that cannot name one execution window will not get a healthier cascade by adding calendars. Gahmbit will not pretend otherwise. Parallel cycles would make coverage and alignment look fine while the company is not actually looking at the same pledges.

We would rather lose the “flexible cadences” checkbox than ship a product that hides that.

Related: How to see who hasn’t written OKRs · How to score committed vs aspirational OKRs · How to run OKRs without migrating your tools