Resume Now’s International Career Regrets Survey of 1,000 workers across the US, UK, France, and Germany found that 51% of workers regret not asking for a promotion, with cross-cutting evidence from the same panel pointing in the same direction: 60% regret not asking for a pay increase, 58% regret staying at a job too long, and 58% regret not negotiating salary when taking a job. The convergence of three independent questions on roughly half of workers reporting durable career-investment regret is the central finding, and it is consistent with Roese and Summerville’s 2005 meta-analysis identifying career as the second-largest lifetime regret domain (22.3% of biggest regrets, behind only education). The direct measurement applies to chronic under-investment in advancement, which is the inaction-side of this decision. Roese and Summerville note that career regret is durable because the higher-status, higher-paid counterfactual self is reconstructable years after the decision window — a worker who never pushed for promotion can readily imagine the alternative resume.
The action-side estimate is harder to pin down. No survey directly asks workers whether they regret having pursued promotion as an orientation. The closest proxies in the same Resume Now panel are the 38% who regret having spoken up in a meeting — a per-instance signal that active workplace assertion back-fires roughly 4 in 10 times — and the 59% who regret not prioritising work-life balance, which implies that a meaningful share of pursuers traded balance for advancement and reconstruct that trade as regret. Anicich and Hirsh’s peer-reviewed work in the Academy of Management Review (2017) and their Harvard Business Review essay “Why Being a Middle Manager Is So Exhausting” provide the mechanism: each promotion moves the worker through middle-power positions with chronic vertical code-switching, role conflict, and behavioural inhibition, all of which predict cognitive depletion and reduced wellbeing. Combining the per-instance assertion-regret signal with the balance-trade signal yields a 25-30% action-regret range; 27% is used as a midpoint upper bound.
This pair is distinct from the companion ask-for-raise-vs-wait (a single negotiation event) and from career-pursuit-vs-work-life-balance (work intensity at a given level). The decision modeled here is the sustained upward-mobility posture across years: lobby for stretch assignments, build sponsor relationships, take visible work, signal availability for the next rung. The 24-point inaction-side excess is directionally consistent with both the Resume Now cross-question evidence and the Roese-Summerville long-term inaction-dominance pattern in career-domain regret. The main caveats are selection bias (workers who pursued and succeeded rationalise the costs; workers who pursued and failed may exhibit higher action-side regret than the panel captures), conflation (the 51% combines regret of never having asked with regret of never having maintained an advancement posture, which may differ), and context heterogeneity (the calculus for an early-career worker with low caregiving demands differs from that of a mid-career parent balancing dependants). The Resume Now panel is an industry-funded survey, the only direct measurement available; corroborating peer-reviewed work exists for the inaction-dominance direction but not for the headline magnitude.