The Number Nobody Measures
You measure engagement, output, and turnover risk. You almost certainly don't measure loneliness, and it may be the signal that predicts the others before they move.
You measure engagement, output, and turnover risk. You almost certainly don't measure loneliness, and it may be the signal that predicts the others before they move.
People wait for a better role to be handed to them. The research says reshaping the job they already have moves engagement more than a new title does. It starts with one question in your next 1:1.
Structured interviews predict job performance about twice as accurately as the free-flowing kind. The difference isn't the questions. It's whether every candidate answers the same ones.
Companies write down a version of their culture, then act surprised when the real one diverges. The handbook is a wish. The culture is what happened the last time someone tested a stated value and nobody pushed back.
Generic recognition doesn't scale. Specific recognition doesn't happen often enough. One five-minute Friday practice solves both at once.
Mandatory seatbelts cut driver deaths. They raised pedestrian deaths. Drivers drove faster when they felt safer. The same pattern runs through workplace safety programs.
Most wellness platforms ask "how are you today?" That data averages to noise at the org level, because mood is yesterday's sleep. Climate (growth, agency, meaning) changes on month-scale.
Teams accumulate meetings, rituals, and processes that stopped making sense six months ago. Nobody removes them because nobody is asked to. One 45-minute meeting a month, with a single question, fixes that.
We think people noticed our slip. Mostly they didn't. Gilovich's research shows we overestimate how much others watch us by about twice the actual rate.
Most culture programs fail because they treat culture like a destination. It isn't. It's a gradient that shifts every day in small, mostly invisible ways. The question is which way it's drifting.
We remember the decisions that worked and revise the ones that didn't. Neither pattern makes us better judges. One four-line entry per decision, read back monthly, does.
When teachers were told random students were "academic bloomers," those students posted 10-25 point IQ gains in a year. The same effect runs through your team, whether you know it or not.
think-piece
The decision to leave is made months before the resignation. Your turnover dashboard is a history book. If you want to change retention, stop measuring it.
behavior-hack
Standard check-ins reward performance. "All good!" tells you nothing. A one-word format strips out the theater and gives you a trendable signal in 30 seconds.
research-drop
Your team is holding back. Not because they're afraid. Because they've decided it won't matter. Futility, not fear, is the top reason employees stay silent.
think-piece
Paying people for behaviors they were doing for free makes them do it less, not more, once the pay stops. Decades of research on this. We keep building programs that ignore it.
behavior-hack
Work that keeps intruding on your evening isn't about the work. It's about the missing plan for it. A three-minute closing ritual releases the loop.
research-drop
Knowing you underestimate how long things take doesn't stop you from doing it. The only known fix isn't better planning. It's using someone else's past, not your own.
think-piece
In our data, employees in the top quartile of wellbeing report more daily stressors than the bottom quartile. Not fewer. The standard playbook (remove stress, raise wellbeing) breaks on this finding.
behavior-hack
Most workplace programs don't fail on the idea. They fail on the friction. A 20-second extra step cuts participation roughly in half. Here's the audit that finds them.
research-drop
Being thanked doubles how likely you are to help the next person who asks. The effect passes through strangers. It's not the words. It's what they signal.
think-piece
Employee Assistance Programs have a 2-10% utilization rate. Not because employees don't need help, but because EAPs were designed for a version of stigma that no longer explains the problem.
behavior-hack
Improv comedians never say "no, but." They say "yes, and." When teams adopt this single rule for brainstorming, idea volume increases and the ideas that survive are more original.
research-drop
When Atul Gawande introduced a simple surgical checklist, complications dropped 36% and deaths dropped 47%. The mechanism isn't information. It's forcing a pause in autopilot.