Your engagement survey just came back. Trust in leadership: 78%. Psychological safety: 72%. You're relieved. The culture work is paying off.
Then, three weeks later: your VP of Engineering quits with two days' notice - "I've been unhappy for months." Your best product manager leaves - "I raised concerns about the roadmap six months ago. Nothing changed." A project fails because nobody told the team lead about a critical dependency. Everyone knew. Nobody spoke up.
Survey scores measure what people say. Trust determines what people do. Right now, you have high survey scores and low actual trust.
Key takeaways
- 86% of executives believe employee trust is high - only 67% of employees agree (PwC, 2024)
- Executives report 87% positive psychological safety perceptions; individual contributors report 69% - an 18-point gap (Perceptyx, 2024)
- When psychological safety is genuinely high, attrition risk drops from 12% to 3% - a 4x difference (BCG, 2024)
What trust surveys actually measure
PwC's 2024 Trust Survey found that 86% of executives believe employee trust is high - but only 67% of employees agree. That's not a perception gap. That's two entirely different lived experiences of the same organisation.
Only 69% of individual contributors agree. That 18-point gap sits between the people with the most power and the people with the most to lose by speaking up. Source: Perceptyx, 2024
Most organisations confuse trust signals with trust reality. Trust signals are: survey scores above 70%, nobody openly complaining, "positive" culture. Trust reality is: do people surface problems early, or hide them? Do they admit mistakes, or cover them up? Do they tell leaders hard truths, or what they want to hear?
You can have excellent trust signals and terrible trust reality. In fact, the worse trust reality gets, the better signals often look - because people learn that honesty is dangerous.
The four trust failures
When diagnosing organisations through voice-led methods, four patterns of trust breakdown appear repeatedly. They're worth naming - because once you see them, you can't unsee them.
1. The commitment fade
CEO commits to bi-weekly acquisition updates. Week 2: on time. Week 4: three days late. Week 6: nothing. Week 8: still nothing. Three months later: surprise announcement. The team's reaction: "You said you'd keep us informed. Why should we believe you next time?" It's not the missed updates individually. It's the demonstrated pattern: my commitments are conditional.
2. The safety theatre
Town hall: "I want hard questions. Nothing is off limits." Then someone asks about layoff rumours - and you deflect. Everyone just learned: hard questions aren't actually welcome. Two weeks later you're frustrated nobody asks tough questions. You don't connect this to what they witnessed when someone tried.
3. The integrity gap
Values: "Integrity above all." Quarter-end: sales backdates a contract - approved. Product skips a security review - approved. Marketing exaggerates a capability - approved. The team learns: "Integrity above all" means "integrity until results are at risk." Your best people quietly start looking for exits.
4. The punishment learning
Head of Product raises a concern. You don't fire her. But she gets checked in on twice ("Are you feeling aligned?"), excluded from early planning, gradually sidelined. Six months later she sees another problem. She stays quiet. The issue surfaces late, expensive to fix. Everyone saw what happened to her. They all learned: speaking up gets you marginalised.
High-safety environments: 3%. A 4x difference - and that doesn't count the people who stay and go quiet. Source: BCG, 2024
What to do next
Stop trusting the survey scores in isolation. Map actual behaviour instead. When was the last time someone surfaced a problem early? When did someone admit a mistake before being caught? When did someone challenge a decision and get genuinely thanked?
Test your own response to bad news. Next time someone brings you a problem, notice your reaction. Do you get defensive, minimise, or question their judgement? Or do you thank them and engage seriously? Your response in that moment teaches everyone watching whether honesty is actually welcome.
Audit your commitment follow-through. List every commitment you made in the last six months - including the casual "I'll get back to you." What percentage did you actually honour? Below 80% and you're teaching people your commitments are conditional.
Separate diagnosis from response. The four patterns above require different fixes. Commitment fade requires behavioural consistency over time. Safety theatre requires leaders to change how they respond in the moment. The integrity gap requires decisions under pressure to match stated values. Punishment learning requires visible evidence that honesty doesn't cost careers. Fix the wrong one and you may make things worse.
Frequently Asked Questions
Why do engagement surveys show high trust even when the reality is low?
Engagement surveys have a structural flaw: they ask people whether they feel safe speaking up, but people who don't feel safe are also unlikely to say so in a survey. The result is systematic overestimation of psychological safety. It's not that people are lying - it's that the survey is measuring something other than what you think it's measuring.
What is the actual cost of low organisational trust?
BCG found that employees in low-psychological-safety environments face a 12% annual attrition risk - four times the 3% seen in high-safety environments. Beyond turnover, costs compound through problems caught late, decisions that weren't challenged, and projects that fail because critical information wasn't shared upward.
How long does it take to rebuild trust after it's broken?
Trust breaks fast and rebuilds slowly. A single commitment fade or visibly punished honesty can undo months of culture work. Rebuilding typically requires consistent behavioural change over 6-18 months - and leaders who name what broke publicly, rather than quietly starting over.
What's the difference between psychological safety and being nice?
Psychological safety is the belief that you won't be punished for speaking up with ideas, questions, concerns, or mistakes. It has nothing to do with niceness. Highly psychologically safe teams often have more conflict - because disagreement is surfaced rather than suppressed. Being nice in a low-trust environment is how people protect themselves while staying quiet.
Want to know where your organisation's trust actually breaks down - not just what scores your survey returns?
Anonymous voice diagnostics. Results in 2-3 weeks. Get in touch to start.