In July 2003, Best Buy Store 484 in Manchester, Connecticut, got its employee survey results back. Two items scored especially low, “At work, my opinions seem to count” and “My associates or fellow employees are committed to doing quality work.” The overall score had gone up since the previous survey. Those two had gone down.
“It was honest,” the store’s general manager said later, “and it hurt.” Best Buy measured engagement with Gallup’s Q12, a 12-item survey, and put the score on each manager’s scorecard. The story comes from Gallup’s own Management Journal, in 2004.
At the next quarterly all-employee meeting, he asked everyone to put a sticky note on the item they thought the whole store should work on. The notes piled up on “committed to doing quality work.” Many carried a suggestion. “We want ‘team close,’” some said. “We want everyone to be treated the same, and we want to walk out together as a team.”
The store closed at 9:30 p.m., but some associates were still there near midnight while others were long gone. It depended on which assistant manager was on duty. What the staff wanted fixed wasn’t a policy from head office. It was the last hour of the day.
Twelve questions, none about pay
Employees rate each Q12 statement from strongly agree to strongly disagree. Gallup’s published list starts with “I know what is expected of me at work” and “I have the materials and equipment I need to do my work right.” It runs through “In the last seven days, I have received recognition or praise for doing good work” and “I have a best friend at work,” and ends with “This last year, I have had opportunities at work to learn and grow.” Gallup sorts them into four levels of needs, from basic clarity to personal growth.
None of the twelve mentions pay. Gallup describes them as issues it has found to be “actionable (changeable) at the supervisor or manager level,” though only one contains the word “supervisor.” By Gallup’s account, the list came out of research in the 1990s that included more than 1,000 focus groups. Marcus Buckingham and Curt Coffman’s 1999 book First, Break All the Rules, built on a Gallup study of more than 80,000 managers, was the first book to present the twelve, its publisher says.
Teams that score high on these twelve questions tend to do better. Why they do better is the harder part.
The room, not the building
Gallup estimates that managers account for at least 70% of the variance in engagement scores across business units. Writing about Best Buy, it pointed to the wide variation store by store as a sign that engagement “is created and maintained at the local level.”
A chain can ship every store the same shelves, the same prices and the same training videos. It can’t ship a closing routine. That gets decided by whoever is on duty tonight. Item 11 asks whether someone has talked to you about your progress in the last six months, which is where one-on-ones come in. The People Leave Managers page takes the same idea to quitting.
Who did the counting
In a 2002 meta-analysis, James Harter of Gallup, Frank Schmidt and T. L. Hayes pooled 7,939 business units in 36 companies in the Journal of Applied Psychology. They found links with customer satisfaction, productivity, profit, turnover and accidents “large enough to have substantial practical value.”
Gallup’s 11th edition, in 2024, covered 183,806 units in 347 organizations. Top-quartile units beat bottom-quartile ones by a median 23% in profitability, Gallup reported. Yet in its own table, the average observed correlation between a unit’s engagement and its profitability was 0.10, where 0 means no linear relationship and 1 a perfect positive linear one. Gallup’s case rests on consistency: “The direction of the effect is predictable,” it says, “but the size of effect across companies varies somewhat.”
Three cautions come with those numbers. First, Gallup sells the survey, its items are proprietary, and Gallup’s own researchers led the meta-analyses. Second, the Q12 may overlap with plain job satisfaction. The 2002 paper itself measured “satisfaction-engagement,” and the 2024 report says the Q12 composite has “high convergent validity with affective satisfaction.” A 2019 review for the UK’s Enterprise Research Centre, by Veronica Lin of Queen’s University Belfast, left the 2002 study out of its review of business-unit engagement, saying it measured “workplace facets that relate to job satisfaction and can be actioned upon by managers.”
Third, the arrow could run backward. A University of Maryland team led by B. Schneider studied attitude surveys (not the Q12) at 35 companies over eight years and found overall job satisfaction was predicted by return on assets and earnings per share more strongly than the reverse. A 2010 study led by Gallup researchers, covering 2,178 business units in 10 organizations, found the stronger arrow ran from employees’ perceptions to results, with the reverse weaker. Lin’s review concluded that at the business-unit level, “the causal order is not clear.”
Reading your own score
Imagine a warehouse supervisor whose team’s lowest item is “I have the materials and equipment I need to do my work right.” Her first idea is a team lunch. Instead she asks at the next shift meeting what’s missing, and hears that the handheld scanners die before the shift ends, so people queue for the two chargers. The survey couldn’t tell her that. It told her where to ask.
The number also invites a shortcut. Once a score sits on a manager’s scorecard, as it did at Best Buy, it can become a target in its own right, raised without fixing the work, the trap described in Goodhart’s Law.
In Gallup’s telling, Store 484’s next survey, in January 2004, put it in the top 10% of workgroups in Gallup’s database. What the associates had asked for was smaller than any program: to walk out together as a team.