Employee Engagement Statistics 2026: All 12 Elements Measured — and Twelve Numbers We Removed

Last updated: October 2026. This page has been rebuilt against primary sources. Twelve statistics it previously carried — a global engagement rate that was rising when Gallup measured it falling, a "quiet quitting rate" that is not a published metric, an average tenure of 2.4 years against a measured 4.1, a generational engagement table, a work-model table and a set of "% impact" driver weights — were either contradicted by the publication they named or could not be traced to any publication at all. They are listed with their replacements in section 2. Everything that remains carries the publication it came from and the period it covers, and anything that is our own arithmetic on published data is labelled as ours.

1. The engagement benchmark cheat sheet

Every headline figure on this page in one table, with the publication and the period it covers. This is the block to copy if you are citing engagement data in a deck, a board paper or an article.

2. Twelve numbers we removed from this page, and what replaced them

An earlier version of this page carried the statistics below. We went looking for each one's primary source. Three were contradicted by the publication they were attributed to. Six could not be traced to any publication at all. Two were real measurements of something else, restated as something they do not measure. One was a claim of first-party research we have never conducted. They are listed here rather than quietly deleted, because anyone who copied them from this page deserves to know.

3. Global employee engagement, 2009–2025

Gallup has measured engagement through the World Poll since 2009. The series is the single most useful thing in engagement data because it is long, consistent and global — and because it shows that the story of the last three years is a reversal, not a continuation.

Source: Gallup World Poll, engaged and actively disengaged as published; the not-engaged column is the residual (100 − engaged − actively disengaged). Gallup has no global data for 2010, 2011 or 2017.

The decline since 2023 is a drift into indifference, not into hostility

Three points came off the engaged share between 2023 and 2025. Where did they go? Not, mostly, into active disengagement, which rose only one point from 15% to 16%. Two of the three points went into the not-engaged middle, which grew from 62% to 64% (our decomposition of Gallup's published shares).

That distinction matters operationally, and it is the opposite of how the decline is usually described. Actively disengaged employees are a containment problem. The not-engaged are a management problem — people who are present, compliant and uncommitted, and who respond to clarity, feedback and recognition rather than to intervention. Nearly two-thirds of the world's workforce now sits in that middle category, and it is the largest it has been since 2015. For what that looks like on the ground, see our quiet quitting statistics for 2026.

In relative terms the fall from the peak is larger than it looks in points: three points off a 23% base is a 13% relative fall in the engaged share of the world's workforce in three years(ours). Gallup puts the cost of low engagement at $10 trillion a year, around 9% of global GDP.

4. US employee engagement, 2000–2026: flat at 31%

The US series is longer and is now quarterly. Gallup measured 31% engaged and 18% actively disengaged across the first half of 2026 — unchanged from 2025, and unchanged from 2024. Engagement peaked at 36% in 2020 after a decade of steady growth, fell to 31% by 2024, and has sat there ever since.

Gallup publishes a conversion that makes the series concrete: each percentage point represents roughly 1.6 million full- or part-time US employees. The five points lost since 2020 therefore amount to about 8 million fewer engaged American workers — which is the figure Gallup itself publishes, and a useful check that the multiplier is being applied the way Gallup applies it.

One number inside the flat topline did move. In the first half of 2026, 49% strongly agreed they know what is expected of them at work, up two points — the first improvement in the element that had suffered the largest five-year decline. Gallup attributes part of that to managers having to explain how AI fits into people's jobs (see section 9). Clarity is still a long way below its 61% peak in 2015, and small declines in employees feeling their opinions count and feeling connected to the organisation's purpose offset the gain.

5. The ceiling has not moved: best-practice organisations have held 70% for eleven years

This is the finding we think matters most in the 2026 data, and it comes from a series most engagement articles ignore. Alongside the national and global rates, Gallup publishes the average engagement rate inside its best-practice organisations — the winners of its Exceptional Workplace Award. That line has been at or above 70% every year since 2015.

Read the columns separately and the implication is hard to avoid. The ceiling did not move. The average did. Over the sixteen years of the series the best organisations went from 62% to 70% while the US national rate went from 28% to 31% — and since 2015 the best-practice line has been effectively flat at 70–73% through a pandemic, a hiring boom, a hiring freeze and the arrival of generative AI. Whatever has been happening to national engagement, it has not been happening inside the organisations that manage engagement deliberately.

The gap in people is the number worth quoting: 39 points × 1.6 million employees per point = about 62 million US workers who would be engaged if the national average performed like the top of the market (ours). That is more than the entire engaged population today.

6. All twelve engagement elements, ranked

Engagement is not a single question. Gallup's Q12 measures twelve elements, and it publishes the percentage of US employees who strongly agree with each one. This is the most useful table in engagement data and the one almost nobody reproduces: it tells you which part of the employee experience is actually failing, rather than that "engagement is low".

Source: Gallup Global Indicator: Employee Engagement, US quarterly readings. Sorted by the August 2026 level. Gallup's commentary dates the Q05 pre-pandemic reading to March 2020; the published chart labels it January 2020, which is the label used here. Overall satisfaction is not one of the twelve elements and is shown for context.

Eleven of the twelve are below where they were before the pandemic

Put the January 2020 column next to the August 2026 column and the pattern is unambiguous. Eleven of the twelve elements have fallen, 73 percentage points in total, a mean decline of 6.1 points per element (our calculation on Gallup's published readings). Not one element is higher than it was in January 2020.

The twelfth is the interesting one. "I have a best friend at work" is at 21%, exactly where it stood in January 2020 — the only element the last six years did not move. It is also the only element on the list that an employer does not supply. Expectations, equipment, development, feedback, recognition and a sense of purpose all arrive through the organisation, and all eleven of them are down. The one thing employees provide for each other held its level.

The two steepest falls are both manager-delivered

The two largest single declines are tied at ten points: knowing what is expected of you (56% to 46%) and someone at work caring about you as a person (47% to 37%). Gallup identified these same two as the largest five-year declines in its 2025 analysis, and both are delivered by a manager rather than by a policy.

Our arithmetic on the size of that concentration: those two elements account for 20 of the 73 points lost, or 27.4% of the entire decline, from 16.7% of the items — an over-representation of 1.64×. Set that against the fact that Gallup attributes at least 70% of the variance in team engagement to the manager, and against manager engagement itself falling from 31% in 2022 to 22% in 2025, and the national engagement decline starts to look less like a workforce problem and more like a management-capacity problem. We develop that argument with the recognition data in our employee recognition statistics, and with the manager-specific series in our manager effectiveness statistics.

7. What disengagement costs, with the arithmetic shown

Two published totals anchor the cost question. Gallup puts the global cost of low engagement at $10 trillion a year, about 9% of global GDP, and the US cost of employees who are not engaged or actively disengaged at $2 trillion a year in lost productivity.

Those numbers are too large to act on. Here is the per-employee version, derived from Gallup's own two figures and nothing else:

• Gallup's conversion of 1.6 million employees per percentage point implies a reference workforce of 160 million US employees.
• At 31% engaged, 69% — about 110.4 million people — are not engaged or actively disengaged.
• $2 trillion ÷ 110.4 million = about $18,100 per non-engaged employee per year.
• At that rate, one percentage point of national engagement is worth about $29 billion a year (1.6 million × $18,100).

The eleven outcomes engagement actually moves

The defensible version of the business case is Gallup's Q12 meta-analysis, now in its eleventh edition: 736 studies across 347 organisations, 183,806 business units and 3.35 million employees in 53 industries and 90 countries. It reports the median difference between business units in the top and bottom quartile of engagement.

Source: Gallup, Q12 Meta-Analysis, 11th edition (2024). These are differences between business units, not between individuals, and they are correlational within a predictive design — which is why the table is the right thing to quote and "disengaged people produce 18% less" is not.

Two of these deserve a note. The turnover effect is larger in organisations that already have low turnover (51%) than in high-turnover ones (21%), which inverts the usual assumption that engagement work pays off most where churn is worst. And the two productivity measures — 18% in sales, 14% in production records — are the smallest numbers in the table, well below absenteeism and safety. The cheapest wins from engagement are attendance, safety and retention, not output. Our absenteeism statistics and turnover statistics cover those two outcomes in detail.

8. Employee engagement by work location, 2026

This is the question most engagement articles get wrong, usually by asserting that hybrid workers are the most engaged. Gallup measures it directly for the US.

Source: Gallup, Indicator: Hybrid Work, work-arrangement table dated September 2026 in the page's own markup. See the note below on a second, conflicting table on the same page.

Three readings, the third of which we have not seen published anywhere else:

• Remote wins engagement; hybrid wins wellbeing. Exclusively remote employees are the most engaged group at 37%, but hybrid employees are the most likely to be thriving in their lives at 54%. These are different outcomes and they do not peak in the same arrangement, so "which model is best" depends on which one you are managing for.
• The spread is 9 points, not the 20-point chasm usually implied — from 37% remote to 28% on-site in non-remote-capable jobs.
• Thwarted preference is not what is driving the gap. On-site employees whose jobs could be done remotely score 29%, and on-site employees whose jobs could not score 28% — a one-point difference. If being made to come in against the logic of the job were the mechanism, those two groups would diverge sharply. They do not. The gap sits between remote or hybrid work and on-site work generally, which points at autonomy and commute rather than at return-to-office resentment.

The arrangement shares, as of May 2026, were 52% hybrid, 26% exclusively remote and 22% on-site among remote-capable employees — against a stated preference of 61% hybrid, 33% remote and just 6% on-site. Our hybrid and remote work statistics for 2026 cover the arrangement series, the productivity evidence and the policy-ownership data in full.

9. AI and engagement: 73% of the gain is management, not technology

Gallup's July 2026 release is the first large measurement of AI against engagement — 43,262 US employees surveyed in February and May 2026, of whom 21,724 work in organisations that have integrated AI. The headline is that the national rate did not move while AI adoption accelerated. The detail is more useful than that.

Source: Gallup, Employee Engagement Remains Flat as AI Adoption Accelerates, 21 July 2026.

Our decomposition of that last row is the number to take away. The full stack of conditions produces 53% engagement against a 31% national rate — a 22-point gain. Adoption on its own accounts for 6 of those points, or 27%. The remaining 16 points, 73% of the total, come from the management wrapper around the technology: an integration plan, frequent use and a manager who actively supports it.

Two further comparisons make the point sharper. The 18-point gap between employees whose manager supports their AI use and those whose manager does not is 3.6 times the entire five-point national decline since 2020 (ours). And 53% is 56% of the way from the national average to the 70% best-practice ceiling in section 5 — achieved by three management practices, none of which is an AI licence.

The productivity data points the same way. Among weekly AI users in adopting organisations, only 17% give AI the highest possible rating for its effect on their productivity. That rises to 33% where the manager actively supports AI use (against 9% where they do not), and to 50% where a clear plan, frequent use, manager support and an engaged employee all coincide. Gallup's conclusion, which the numbers support, is that making AI available is not the same as integrating it.

10. Managers are now less engaged than the people they manage were in 2022

Globally, 22% of managers were engaged in 2025, down nine points from 31% in 2022, including a five-point drop in the final year. Non-managers were at 19%. The three-point manager premium is all that is left of a gap that used to be substantial.

Scaled against the workforce decline, the asymmetry is the story: the workforce fell from 23% to 20%, a 13% relative fall, while managers fell 31% to 22%, a 29% relative fall — more than twice as fast (our calculation). Gallup also reports that leaders report more daily stress (+7 points), anger (+12), sadness (+11) and loneliness (+10) than individual contributors, and that in best-practice organisations 79% of managers are engaged.

That matters because of the attribution Gallup makes directly: at least 70% of the variance in team engagement is down to the manager. The channel through which engagement is produced is deteriorating faster than the population it serves. We trace the consequences for recognition specifically in our recognition statistics, and the manager series itself in our manager effectiveness statistics for 2026. For the burnout data behind the stress figures, see our employee burnout statistics.

11. Engagement by country: the eleven-point gap inside "North America"

Gallup reports engagement for ten world regions and for individual countries on a three-year rolling aggregate. Benchmarking against a region hides something important.

The United States and Canada region is reported at 31%, the highest in the world, and it is the number that gets used as "North America" in multinational benchmarks. Split it and the United States is at 32% while Canada is at 21% — an eleven-point gap, a factor of 1.52×, inside a single region (ours, on Gallup's country aggregates). Canada's rate is identical to Australia's and below Sweden's. Any organisation using a "North America" engagement benchmark is using a US benchmark and applying it to a Canadian workforce eleven points below it.

Source: Gallup, State of the Global Workplace: 2026, country data aggregated from 2023, 2024 and 2025. Selected countries; Gallup publishes all of them.

The European figures are the ones worth sitting with if you run a multinational programme. The United Kingdom at 10%, Germany at 11% and Poland at 7% are a third or less of the US rate, and Europe is the lowest-engagement region on earth at 12% while being near the top for life satisfaction. That combination — disconnected from work, satisfied with life — is examined through the thriving-to-engagement ratio in our recognition statistics, where Europe scores 4.08 against a global 1.70. Programmes built on US engagement norms mis-target badly there, and a locally relevant reward catalog matters more than the communications wrapper.

12. Disengaged, and staying: the exit door looks closed

An engagement decline usually arrives with a turnover problem. This one has not, and the reason is in the labour-market data rather than in the engagement data.

• 46% of US employees say it is currently a good time to find a job — down ten points on the prior three-year rolling average, the largest move in any of Gallup's US workplace indicators.
• Quits are at 1.9% a month, 3.1 million people (BLS JOLTS, August 2026), against hires of 3.3% and openings of 4.3%. Annualised, the quits rate is about 22.8%.
• Globally, 52% say it is a good time to find a job, against a 55% peak in 2019.

So the not-engaged middle that grew to 64% of the global workforce is not, for the most part, leaving. It is staying, and it is the group that now determines productivity. That is the practical argument for treating engagement as a performance system rather than a retention tactic: the people the data is describing are going to be on the payroll next year either way. Our retention strategies and productivity statistics cover both sides of that.

13. Generations and age: what the data supports, and what it does not

We removed this page's generational engagement table because we could not find a primary measurement behind any version of it, and because the direction it implied is contradicted by the publications that do measure age. What the data supports is narrower and more useful:

• Younger workers have had among the sharpest declines since 2020, alongside hybrid workers and managers. Gallup names those three groups specifically.
• In the 2024 annual update, Gen Z employees were five points less engaged than the year before, with the declines concentrated in the most basic elements: clarity of expectations, recognition, materials and equipment, doing what they do best, feeling cared about and development opportunities.
• 44% of Gen Z workers describe their workplace as isolated or impersonal, against 32% of employees overall and 41% of remote employees (Gallup, 2025).
• Engagement among Generation Z and younger millennials has fallen about eight points since 2020, and among older millennials about nine — against five points for the workforce as a whole.

Note what that pattern is not. It is not a story about young people wanting different rewards; it is the same twelve elements, failing hardest for the people with the least workplace tenure to fall back on. The design response is individualisation at the point of redemption rather than a cohort rule at the point of design — let people choose. Our Gen Z workplace statistics carry the age-specific data, our tenure statistics the measured tenure-by-age figures, our employee onboarding statistics the first-year population those elements fail soonest for — about 30.0 million US workers, a fifth of everyone employed — and our rewards personalisation guide covers how far to take choice.

14. How to use this data

If you are turning these numbers into a programme rather than a slide, the sequencing matters more than the benchmark — our employee engagement strategy page covers the order we recommend running it in.

Engagement data without action is overhead. Four things follow directly from the numbers above rather than from general advice.

Two measurement habits also follow. Run the twelve elements quarterly rather than annually — Gallup publishes quarterly for a reason, and the 2026 clarity improvement would have been invisible on an annual cycle. And measure your own turnover and cost per hire rather than importing a national figure: the per-outcome quartile differences in section 7 are the defensible ones, and our engagement ROI guide works through the calculation. For the platform side, our engagement analytics tracks recognition activity against retention by team.

15. Methodology, derived metrics and sources

Every figure above is either a published measurement with its publication named, or arithmetic we performed on published measurements. The second category is listed here in full so it can be checked or disputed.

Metrics derived by Rewordin

What we deliberately did not do

• No figure on this page comes from a vendor blog, an infographic or a secondary round-up. Where we could only find a statistic in secondary content, we removed it — see section 2.
• We did not reconstruct the generational engagement table from plausible-looking numbers. Gallup publishes age-group changes but not a current engagement rate per generation, so we report the changes.
• We did not blend measurements with different bases. The global series, the US quarterly series, the three-year country aggregates and the remote-capable work-location sample are kept separate and labelled, which is why the US appears as 31%, 32% and 29% in different places on this page.

Key takeaways

• Engagement is 20% globally and 31% in the United States, and the US figure has not moved for three years. Any page reporting a rising global rate for 2026 is wrong.
• The decline since 2023 went into indifference rather than hostility: two of the three points lost landed in the not-engaged middle, now 64% of the world's workforce.
• Eleven of the twelve engagement elements are below their January 2020 level, 73 points in total. The two steepest falls are both manager-delivered.
• Best-practice organisations have held 70% for eleven straight years. The ceiling did not move; the average did. The gap is about 62 million US employees.
• AI is associated with higher engagement, but 73% of the gain is the management wrapper— a plan, frequency and an engaged manager — not the technology.
• Remote 37%, hybrid 34%, on-site 29% and 28%. The two on-site groups differ by one point, so thwarted remote preference is not what is driving the gap.
• The not-engaged majority is staying: job-market confidence is down ten points and quits are at 1.9% a month. Engagement is a productivity question now, not a retention one.

Frequently asked questions