The OKR & Strategy Execution Report 2026

We first ran this survey in 2022, with 40 companies. This year we asked the same core questions of 187, and we tracked down 37 of the original 40: 31 answered the survey again, and 6 more answered a short follow-up on the status of their OKR program. Three of the five measures we can compare went down. The 2022 baseline is 40 companies, so a difference of about 10 points between the two years is within its margin of error. This section looks at why the biggest drop happened, and then at the companies that moved the other way.

How much of the drop comes from younger programs

The 84 companies that have run OKRs for two years or longer sit at 86%. The 103 younger ones sit at 67%. A third of this year’s sample, 59 of 187 companies, is in its first year of OKRs, a group the 2022 survey left out: in 2022 a company needed at least a year of OKRs to take part, and this year one finished cycle was enough. Counting only the 128 companies a year or older, the group the 2022 survey covered, 80% say OKRs met or exceeded expectations, against 88% in 2022: 5 of the 13 points go away, and the rest is within the margin of a 40-company baseline. Two cautions. The 2022 survey did not record program age, so this matches the rule for taking part, not the age mix. And the older groups here contain only the programs that lasted long enough to reach them, which flatters them. We cannot check the recommendation or the impact rows the same way.

Where the numbers held up

The high-impact group, companies rating OKR impact 3.5 out of 5 or higher in all four impact areas, is 22% of companies this year (41 of 187), against 18% in 2022 (7 of 40). The two years apply the same bar to similar rather than identical scores, and the difference is within the margin of a 40-company baseline, so the fair reading is that the group has held its ground while the average fell. Three of the six habits it shares repeat findings of the 2022 study: a clear strategy, frequent check-ins, and treating OKRs as something you keep adjusting rather than something you install once.

Of the 31 companies that took part in both surveys, 26 still run OKRs in some form (4 of them only in part of the company) and 5 have stopped. We compared each of the 26 with its own 2022 answers, not with the rest of the sample, using one respondent per company in both years so the two scores are built the same way: 11 rated their program higher in 2026, 6 the same, and 9 lower.

This year we added five questions on how clearly a company’s strategy is written down and understood: the Strategy Clarity Index. Scores on it separate the companies more sharply than anything else we measured, including how often they check in, their industry, how leadership reviews progress and how long their cycles run. One thing could flatter the index, so we tested it. The index is the program owner’s own judgment, given on the same form as their own rating of OKR impact, so part of the link could be one person answering two questions the same way. Measured against the employees’ ratings alone, which the program owner’s answers do not feed, the link is weaker (r = 0.55 rather than 0.62) and still the strongest in the study.

OKRs need a strategy to point at

OKRs cannot create a clear strategy. They can only turn one into goals. When the strategy is unclear, OKRs become an elaborate way of writing down confusion.

OKRs ask teams to set ambitious goals that support the company’s priorities. If nobody has spelled out those priorities, teams either invent their own, which pulls them in different directions, or default to whatever is loudest in their inbox.

The most common way to set OKRs is a cascade: the CEO’s goals become the next level’s goals, and so on down to every team. It is tidy, and companies that use it score lower than companies that work top-down and bottom-up: leaders set the priorities and teams propose how they will contribute. 95 of the 187 companies cascade, 77 let teams propose, and 15 do neither in any organized way.

Why handed-down OKRs score lower

01 · Ownership

When a team’s goal is handed down as someone else’s Key Result, the team is working toward something it did not shape, and ownership suffers. This is an argument about how cascading works, not a measurement: the survey records which approach a company uses and how it scores, not how often individual OKRs get updated.

02 · Rigidity

A cascade assumes the strategy will not change mid-cycle. It does, and a rigid hierarchy cannot absorb that. Teams either drop their OKRs or keep chasing an outdated strategy. When teams propose their own OKRs, they can go back to leadership with a new proposal when things change.

03 · Blind spots between teams

Most important work needs several teams to coordinate. A cascade hands each team its goals from the level above, so no team has to look at what the teams beside it are doing, and each optimizes its own slice. When teams bring their drafts to leadership, the drafts can be looked at side by side before anything is agreed, so overlaps and dependencies between teams show up before the cycle starts.

Every company in this study tracks its OKRs in the same strategy execution software, so the question here is not whether a tool helps. It is whether leadership looks at the OKRs there, or has the progress rebuilt somewhere else. We asked how leadership reviews OKR progress each cycle. 73 of the 187 companies (39%) review straight from the live OKRs or from reports the software produces, 82 (44%) do that for some reviews and rebuild progress in slides by hand for others, and 32 (17%) rebuild it in slides for every review. Their OKR impact scores fall at each step, from 3.5 out of 5 to 2.9 to 2.3.

Older programs are more likely to review straight from the live data, so we split each group by program age, and the gap holds within both. Among programs two years and older, reviewing from the live OKRs rates 3.7 against 2.6 for slides at every review; among younger programs, 3.2 against 2.1. What the survey cannot separate is commitment: the companies that drop the slide deck may also be the ones that take OKRs most seriously. The company that publishes this report sells that software, which is a reason to read this section with care.

Six in ten companies still rebuild OKR status in slides every cycle

114 of the 187 companies (61%), the two slide groups above, rebuild OKR status in slides for leadership at least once every cycle, even though the same numbers already sit in their strategy execution software. The person who runs the program spends a median of about 6 hours per cycle on it. The other 73 companies (39%) review only the live data.

The split lines up with the gap between leaders and employees. Where rebuilt slides come into at least one leadership review each cycle, leaders rate OKR impact 1.8 points higher than their own employees do. Where every review runs on the live data, the gap is 0.8 points. Weighted together, the two groups give the report’s overall gap of 1.4 points. One reading is that a slide shows leaders a tidied-up version of progress the employees doing the work do not recognize. The survey records the gap, not the reason for it.

Only 1 in 3 companies lets every employee see every OKR

64 of the 187 companies (34%) make every OKR visible to every employee. 90 (48%) restrict some of them, and 33 (18%) let each team see only its own. Companies where everyone can see every OKR rate impact 3.4 out of 5; companies with team-only visibility rate it 2.4. Visibility is usually a setting rather than a project, which makes it the cheapest of the three changes in this section. Whether the setting drives the gap, or a more open culture drives both, the survey cannot separate.

Most OKR research only looks at programs that are still running. Following the companies from our 2022 survey lets us look at the ones that stopped. 8 of the 37 companies we could account for, 22%, had dropped OKRs entirely by 2026, and another 5 had paused or scaled back. Adding 24 earlier programs that this year’s companies told us they had shut down gives 32 shut-down programs. Their reasons fall into five repeating patterns. None of the five is about how OKRs are written or scored; each is a decision made, or skipped, around them: who sponsors the program, whether there is a strategy for it to point at, where the work actually lives, and whether anyone looks back at a finished cycle.

We measured stopping two ways. Four years after 2022, 24 of the 37 programs we could account for were still the company’s main goal system, 16 of them unchanged; the four-year section counts 26 of its 31 as still running because it includes programs that scaled back. And of all 32 shut-down programs, 21 had run between six months and three years when they stopped, the Walk stage, when the system is only half in place.

Why the programs stopped

32 cases: the 8 programs from the 2022 group that stopped, plus 24 companies in this year’s sample that told us about an earlier OKR program they had shut down before the current one. We gave each case one main reason. The counts behind the shares are 13, 7, 6, 3 and 3, and a leadership change leads in both groups: 6 of the 8, and 7 of the 24. Two researchers coded the reasons independently and agreed on 28 of the 32; they settled the other 4 together.