OKRs, KPIs, Hoshin, EOS. One organisation.
Most large organisations aren't choosing a goal framework. They already have four, each with defenders, and each defender is partly right. Convergence isn't picking a winner. It's keeping what each one does best and agreeing what has to be common.
If this sounds familiar, you are in good company.
Head office runs OKRs. A manufacturing region runs Hoshin, with an X-matrix older than some of the staff. Finance swears by its balanced scorecard. The business you acquired two years ago runs EOS, with Rocks and a Level 10 every Monday at nine sharp. And everyone, everywhere, has KPIs.
None of this is a failure. It's what happens when capable people solve goal setting locally. The cost shows up at group level: numbers that don't read across, reviews that don't line up, and a leadership team translating between four dialects every month.
Teachers, not rivals.
Hoshin Kanri
Catchball: objectives thrown back and forth between levels until both sides genuinely commit. Aligned, not cascaded, proven at industrial scale decades before OKRs existed. We keep catchball, and Hoshin's wall between breakthrough goals and daily management.
Balanced Scorecard
Measures as causal hypotheses, not a dashboard of whatever happens to be countable. The scorecard watches the whole business. OKRs move the part that matters now. They divide the labour cleanly.
EOS and Traction
Issues as a scheduled habit, not an embarrassment, and a weekly meeting that is a well-run check-in by another name. Keep the rhythm. Let Rocks grow into OKRs where ambition matters. Rocks are commitments, OKRs are bets, and a healthy portfolio holds both.
KPIs
Run-the-business health, which nothing else should replace. A KPI becomes an OKR only when it is broken and needs a step change, or you are deliberately pushing it to a new level. When the work lands, it goes back on the scorecard with a new threshold.
Agree four things. Leave the rest alone.
01The shape of a goal
What a good objective and key result look like, written the same way everywhere, so a goal in one region reads correctly in another.
02The wall between run and change
Health on a KPI scorecard, change in OKRs. Hoshin, EOS and OKRs all drew this line independently. Three eras, one wall.
03The review rhythm
One set of clocks for the whole group, so evidence arrives at leadership at the same time from every division.
04Measures that read across
A handful of measures that mean the same thing in every market. Everything else, including what people call things, can stay local. Nobody needs to rename the X-matrix to feel aligned.
When everyone finds the same thing, it is a finding.
Different decades, different industries, the same conclusions. That's why convergence is possible: the frameworks agree on far more than their defenders do.
Legacy is the barrier, not capability.
The teams that find convergence hardest are rarely the least able. They're the ones where the old system has been doing the most jobs for the longest time, woven into how people manage, meet and report. Every habit has a history, and most of the histories are reasonable.
So you lead with the why, change by change, and you show where the work that used to live in the old system will live instead. The teams newest to it, with the least to unlearn, usually go fastest. Plan for both.
See how one group did it →Translate first. Replace almost nothing.
01Survey before anyone changes anything
Map what each part of the organisation runs, and what it is genuinely good at.
02Name the strengths out loud
Before asking a team to change, tell them what they do best. It changes the whole conversation.
03Agree the four commons
Shape, wall, rhythm and shared measures. Then stop.
04Translate, then converge
Map local artefacts onto the common shape. Retire duplicates only once the common version is proven better.
05Volunteers first
The quiet majority watches the first team through. Choose that team carefully.
Q&A
Can OKRs and Hoshin Kanri work together? +
Yes. Hoshin's catchball and its separation of breakthrough objectives from daily management fit OKRs naturally. Many groups keep the X-matrix as a planning artefact and run the breakthrough objectives as OKRs on a shared review rhythm.
Should we replace KPIs with OKRs? +
No. KPIs monitor the health of the business; OKRs change it. Keep a KPI scorecard for run-the-business work, promote a KPI to an OKR only when it needs a step change, and move it back when the work is done.
How do you merge EOS and OKRs after an acquisition? +
Keep the EOS weekly rhythm, scorecard and issues habit, which are genuinely good, and let Rocks mature into OKRs where the ambition is larger. Then align both onto the group's goal shape and review cycle.
Do all divisions have to use the same framework? +
They need the same goal shape, the same separation between run and change, the same review rhythm and a few shared measures. Beyond that, local artefacts and vocabulary can stay. Uniformity is not the goal; reading across is.
We'll show you how it works and share case studies.
Then you decide.