Values that do work — past the wall poster
Why most stated values don't shape behavior, the structural tests of a working value, and the cadence that keeps values alive past the rollout.
On this page▾
- Patrick Lencioni's HBR essay 'Make Your Values Mean Something' identified four value types: core, aspirational, permission-to-play, and accidental. Most published value sets are an undifferentiated mash of all four.
- The test of a working value: does it force a real trade-off? 'We value integrity' fails — nobody publishes the opposite. 'We value transparency over comfort' passes — it tells you what gets sacrificed.
- Values that survive 5 years tend to be 3–7 in count, paired with anti-values (what we don't value), and operationalized in hiring rubrics, promotion criteria, and performance reviews.
- The single most predictive question: 'In our last 10 promotions and 10 terminations, can you trace each decision to a specific value?' If not, the values are decorative.
Most company values are written during a leadership offsite, etched onto a wall, mentioned at the all-hands twice a year, and forgotten by Year 3. The reason isn't that the words were wrong — it's that the words were never connected to a decision.
The four types of values
| Type | Definition | Honest treatment |
|---|---|---|
| Core | Behaviors the company actively rewards and protects, in good times and bad | These are the 3–5 you publish and operationalize |
| Aspirational | Behaviors the company wants to develop but doesn't yet model | Be honest that they're aspirational; track progress; revisit |
| Permission-to-play | Baseline behaviors required to be in the industry (honesty, respect) | Don't publish — they're table stakes |
| Accidental | Behaviors that emerged unintentionally, often from founder personality | Audit them; some are worth keeping, some are limiting |
The trade-off test
Patrick Lencioni's diagnostic: a real value forces a trade-off that some companies would not make. 'We value customers' is not a value because every company says it. 'We value craft over deadlines, even when deadlines hurt' is a value because most companies would not say it — and most importantly, most companies would not act on it. The trade-off is what gives the value teeth.
Stripe: 'Move with urgency and focus.' Trades off: comprehensiveness, comfort. Netflix: 'No vacation policy.' Trades off: structure, predictability. Amazon: 'Disagree and commit.' Trades off: prolonged debate, comfort with consensus. Each value implies what gets sacrificed.
Operationalizing values
- 1Hiring rubricEach value is assessed in the interview loop with a specific scorecard. A candidate strong on one value but weak on another is a no-hire, not a 'we'll develop them.'
- 2Promotion criteriaPromotion to manager and above requires explicit values demonstration evidence. Calibration sessions surface when promotions are happening despite values gaps.
- 3Performance review structurePerformance review includes a section on values demonstration. High-performing-but-values-failing employees get flagged for the hardest conversation.
- 4Public momentsWhen the company terminates a senior person for values reasons, the all-hands message names the value violated. The CEO's willingness to do this is the single strongest signal.
The annual values audit
- Trace the last 10 promotions: which value did each promotion exemplify?
- Trace the last 10 terminations: which value did each termination protect?
- Survey: 'Can you give me a recent example of someone living this value?' If <60% of employees can answer per value, the value is decorative.
- Calibration test: in your last hiring debrief, was any of the discussion organized around values? If not, the values are post-hoc rationalization, not selection criteria.
- External signal: do candidates mention specific values during interviews? If so, your career site is doing real work; if not, the values aren't visible.
Frequently asked questions
How often should we revisit our values?
Core values: every 5–7 years, with serious resistance to change. Aspirational values: annually, since they're meant to graduate to core. Operating principles (more granular, more dynamic): annually.
What if a long-tenured high performer violates a value?
This is the test. If they keep their job, you've taught the company that values are negotiable for high performers — which means values don't apply at the top, which means they don't apply at all. The hard call is the only call.
Should employees vote on values?
Co-create, don't vote. Leadership owns the values (because leadership has to enforce them); employees inform them through structured input. Pure-democracy values usually end up as the least common denominator.
- Make Your Values Mean Something (Patrick Lencioni, HBR 2002) — Harvard Business Review
- Built to Last (Collins & Porras) — HarperBusiness
- Netflix Culture Memo (Reed Hastings) — Netflix
Read next
All playbooksEvery company has a culture. Most are accidents. This is how to design culture as an operating system — values that behave like rules, rituals that ship…
The cultural equivalent of technical debt, the four debt types, and the only intervention that consistently moves the needle: removing people who model the…
Amy Edmondson's research at HBS and Google's Project Aristotle converged on the same finding: psychological safety is the single biggest predictor of team…