How to
How to run OKRs without migrating your tools
Short answer: Keep Jira, Linear, Notion, the docs, the personal to-do list. Put objectives and key results in one org-visible square with a shared cycle and loud Missing. Do not migrate the work OS. Do not pretend a task tracker is a strategy hub.
OKRs are pledges. Tasks are how a pledge gets done. Those are different jobs.
The request that sounds reasonable
The rollout committee wants “one tool.” Someone demos a work OS that can hold issues, docs, goals, and a chat overlay. The pitch is familiar: if everything lives in one place, alignment will follow.
It does not. Alignment follows shared commitments, not a shared issue schema. Companies that try to make the tracker into the town square usually get a more expensive tracker and a quieter strategy. The OKRs become epics. The people who did not write any disappear into “no issues assigned.” The quarter becomes a backlog.
The other failure is the opposite: OKRs in a slide deck, work in six apps, and a Friday status meeting that tries to join them by memory.
You do not need a third failure — a new work OS — to fix the second.
What the research says about tool sprawl
Knowledge workers already spend most of the day on work about work: status chasing, app switching, searching for the latest version, meetings about the meetings. Asana’s Anatomy of Work Index (10,000+ workers) put that share at about 60% in the original study and 58% in the 2023 global index. The 2023 release also found knowledge workers using 8.8 apps on a typical day (about 10 at director level and above), and that missed messages rise as the app count rises. The 2024 State of Work follow-up reported 75% of workers describing digital exhaustion — up from 64% the year before — as AI tools landed on top of the existing pile rather than replacing it.
That is the backdrop for “just add our OKR module to the work OS.” You are asking a tired org to learn another system of record for tickets and to treat that system as the place strategy is judged. Most people will keep the tracker they know and ignore the goals tab. The goals tab then looks empty, so someone filters it to the writers, and you are back to a highlight reel.
SaaS inventories tell the same story at company scale. BetterCloud’s State of SaaS 2025 put the average firm at 106 cloud apps after two years of light consolidation. Zylo’s index puts the average buyer far higher — hundreds of paid apps in tech and enterprise cohorts — with spend still rising even when the app count is flat. The problem is not “we lack a place to type a task.” The problem is too many places that claim to be the source of truth.
Locke and Latham again, from the other side: goals work when they are specific and when feedback is attached to those goals. A Jira ticket titled “Phase 2” is a task. It does not tell the company whether the quarter’s pledge moved. A key result with a baseline, a target, a current value, and a check-in does. You can link them in a comment. You should not pretend they are the same object.
Industry OKR surveys keep finding tool fatigue among the reasons programs stall, next to missing owners and no weekly rhythm. The teams that last are not the ones that migrated every workflow. They are the ones that can name the commitments in under a week and check in on them without opening a new project taxonomy.
Town square vs work OS
| Job | Lives in | What “done” means |
|---|---|---|
| Ship the work | Jira, Linear, GitHub, the design file, the CRM | The ticket closes |
| Pledge the quarter | Gahmbit — objective + key results, one cycle | The KR moves; the score is honest |
| See who hasn’t pledged | People directory — Has / Missing / Draft / Pending | The blank is still on the page |
| Talk about the pledge | Comments, feed, review agenda | Org-visible, not a manager-only channel |
A work OS is good at the first row. It is a bad town square because it is organized around work items, not who has committed. It will always be easier to hide an empty person in a tracker (they simply have no issues) than in a directory that lists every teammate.
Gahmbit is the second, third, and fourth rows. Keep your tools. Share your commitments.
How to run the split
1. Write OKRs in the square, in one cycle
Company, team, and individual objectives go in the active cycle. Parents stay same-cycle. Year-and-longer bets go on strategic objectives. Check-in frequency is update cadence, not a second tracker. If you need the longer argument, it is here.
2. Point at the work; do not import it
In the key result or a comment, link the epic, the dashboard, the PR, the campaign. The evidence of a KR can live in another system. The pledge and the score do not. If the tracker is down, the company should still be able to see what it promised.
3. Check in on the KR, not on the ticket
Weekly or biweekly, the owner updates current, confidence, and a sentence the whole org can read. That is the feedback Locke and Latham say the goal needs. A burst of ticket closes is not a check-in. A check-in without a number is a status meeting.
4. Keep coverage on people, not on issue assignees
Missing stays loud. Assignees in Jira are not a roster of who pledged. Helpers in Gahmbit (Contributors, Watchers) are not a roster either. One owner per objective and per KR.
5. Score the pledge, not the backlog
Committed vs aspirational are two bars. Velocity in the tracker is a delivery metric. It must not become the OKR grade. A team can close 40 tickets and miss a committed 1.0. A team can close 8 tickets and deliver a moonshot KR.
6. Add Slack as a window, not a second square
When Slack is enabled, template digests can post an aggregate scoreboard to a channel. That is a view onto the square. It is not where OKRs are written, and it does not name individuals as overdue. Turning a digest off does not hide coverage in Gahmbit.
What not to do
Do not rebuild Jira inside the OKR product. We will not become your issue tracker. If a vendor offers “goals + work + docs + chat,” ask where Missing people go. If the answer is “they just don’t have issues,” walk away.
Do not keep OKRs only in slides. Slides are a broadcast. They are not a directory, not a cycle, not a check-in log, and not a retrospective with two means.
Do not require a migration to start. Create the org, invite or sync people, open a quarter, write company OKRs. The first honest picture is who is Missing. That can happen while the tickets stay where they are.
Do not buy a seat plan to unlock the square. The OKR loop is free. Optional AI is prepaid org credits when a model runs — never a gate for writing or reading commitments.
A worked example
Northstar Labs already has Linear, Figma, and a CRM.
- Product’s Q3 objective lives in Gahmbit. KR-1 links to the Linear epic. Check-ins happen on the KR every week. Ticket status stays in Linear.
- Sales’ Q3 objective lives in Gahmbit. The pipeline number is a KR, updated monthly. The CRM does not become the OKR.
- Design never “moved to Gahmbit.” They still open Figma. They have a KR on the same company objective because they own a measurable outcome, not because they migrated files.
Week two, four people are Missing. Nobody asks which app they failed to log into. They have not pledged. That conversation is the rollout.
If the committee still wants one tool
They want less switching. The data says switching is already the tax — and that another all-in-one usually adds a tab, not a subtraction. Subtract the status slide and the private tracker of goals. Keep the tools that ship. Give the pledges a square.
We would rather lose the “all your work in one place” checkbox than ship a work OS that hides who has not committed.
Sources
- Asana (2021). Anatomy of Work Index — ~60% of time on “work about work”; ~10 apps/day in the original global study.
- Asana (2023). Anatomy of Work Global Index — 58% of the day on work-about-work; 8.8 apps for knowledge workers, ~10 for directors+.
- Asana Work Innovation Lab (2024). State of Work — 75% report digital exhaustion (from 64%).
- BetterCloud (2025). State of SaaS — average 106 SaaS apps after light consolidation.
- Zylo (2025–26). SaaS Management Index — average buyer still in the hundreds of paid apps; spend up while counts flatten.
- Locke, E. A., & Latham, G. P. (2002). Goals need specificity and feedback attached to those goals — not to a proxy ticket count.
- Related: How to see who hasn’t written OKRs · How to score committed vs aspirational OKRs
FAQ
- Do we have to move Jira, Linear, or Asana into Gahmbit?
- No. Keep the tools that run the work. Write the commitments — objectives and key results — in the square where everyone can read them.
- Where should check-ins go?
- On the key result, in Gahmbit, on the shared cycle. The task board can stay where the tasks are. The pledge and the update are company-visible.
- Isn’t one more tool just more sprawl?
- A town square is not a tenth project tracker. It replaces the slide deck and the private spreadsheet that already tried to be the square — and failed because they were not org-visible or cycle-scoped.
- What about Slack?
- Connect it when you want. Template digests and shout-outs stay free. They post aggregate scoreboards, not a second place to hide work. Inference, if you enable it, burns prepaid org credits.
- Is the OKR loop free?
- Yes. Org setup, directory, cycles, OKR CRUD, check-ins, comments, and coverage are free forever. No credit card. No seat tax. Optional AI is prepaid only when a model runs.