When an employee looks at a survey invite and decides not to fill it in, they're not being difficult. They're being rational.
They've done this before. They answered honestly. Nothing changed. Or worse: something changed, and not for the better, and they've always half-wondered if what they said had something to do with it. Maybe that's paranoid. But paranoia doesn't need to be accurate to be effective. If it makes you hesitate, it works.
This is the situation most HR teams are trying to fight when they send a reminder email, adjust the survey design, or switch to a new platform. They assume the problem is the tool. The problem is the track record.
Trust in a feedback channel is earned the same way trust in any relationship is earned: by doing what you say you're going to do, repeatedly, over time. An organisation that says "this survey is anonymous" and then, three weeks after results come back, makes changes that somehow seem to affect only the team that scored lowest - that organisation has done something to trust. It doesn't matter that the changes were coincidental. It doesn't matter that they were pre-planned. The experience of the person on the other end is that honesty carries risk.
So they stop being honest. Or they stop answering at all.
The second source of scepticism is structural: in many tools, "anonymous" means managers see a report rather than individual names. It doesn't mean individual responses aren't stored, accessible to admins, or retrievable in an audit. Employees who understand this - or who simply don't know enough to assume otherwise - adjust their answers accordingly. They write what they'd be comfortable signing.
The third is the oldest: nothing happens. The survey goes out, results are shared in a deck at the all-hands, leadership acknowledges some themes, and then the year continues largely unchanged. People weren't wrong to be sceptical. The process confirmed it.
None of this is a design problem. A more beautiful survey won't fix it. Neither will a shorter one, a gamified one, or one that promises "your voice matters." Scepticism about feedback channels is a response to lived experience, and it can only be undone by different lived experience.
What that looks like in practice: a channel that is structurally anonymous, not just policy-anonymous. A consistent cadence so people know what to expect, not a survey that appears once a year and disappears. And a response - every time. Not a full action plan. Not a town hall. Just some acknowledgement that the signal was received and someone is thinking about it.
Trust is rebuilt by closing the loop, repeatedly, on small things. Not by declaring that the culture has changed.
Relay is built on the premise that the channel has to earn the honesty. If you want to know what your team actually thinks, start there.