A nice bottle of champagne and a survey

Engagement happens in the work

Imagine getting to your desk to find a rather lovely Marks & Spencer hamper, a bottle of champagne and a £20 gift card. What a nice treat. Well, one company I was brought in to help had just done this very thing, right before the engagement survey was run. It was a kind gesture, and I don't doubt it was meant kindly and with positive intent. But, in the days that followed, managers also grew conspicuously pleasant. The survey ran. The scores, as you'd expect, were excellent. And a few weeks later, everything went back to how it had been – fractured, stressful and in need of help.

I'm not telling that story to be unkind – about the company, or about surveys – far from it. People are simply very good at reporting that things feel good when you've just handed them champagne. And that, in miniature, is the thing I keep going over about how we measure engagement.

Because most organisations run these surveys for entirely decent reasons. Someone asked for them. They care about how people feel. They want to do the right thing. Listening genuinely matters, and asking people how they're doing is a reasonable, humane instinct. So the question I find myself returning to isn't whether the surveys are accurate. It's subtler, more equine in nature, and a little more awkward: why have we come to treat measuring engagement as though it were the same thing as managing it? Those are two very different acts – and it's easy to do the first while sincerely believing you've done the second.

There's something reassuring pleasing about a number. A messy, human, hard-to-solve problem becomes 47%. And 47% can go in a board pack; it can sit beside last year's 57% with an arrow pointing to it; it can be explained, defended, benchmarked within the business, broken down per department and compared.

Something deeply human takes on the look of a financial instrument. Which is useful if the aim is to report that something has been attended to, and it moved in some direction. It's less useful if the aim is to make the work and climate better for the people doing it – because a snapshot of the weather is not the same thing as changing the weather. Unless of course you're Storm from X-Men.

You see, engagement doesn't happen when the survey opens, or the couple of weeks leading up to it. It happens on an ordinary Tuesday in January. It happens at 10:17, when someone needs a decision and can't get one. At 11:42, when two teams can't work out who owns a piece of work and both are genuinely blocked. At 2:05, when somebody has a good idea but knows it'll take six meetings to get permission to try it. At 3:30, when a manager offers some honest, useful feedback instead of dodging the conversation. At 4:45, when someone's having a hard time at home and their manager listens, and tells them not to come in until it's resolved. Those tiny moments, stacked up, are the experience of work. A survey arrives months later and asks people to compress all of it into a handful of questions.

None of which makes the number meaningless – only that it can mislead. I once watched an organisation where the engagement score had become the most important measure in the company, treated as a proxy for the health of the whole business, sitting above even the commercial results that were there to keep the business alive. The scores were high. The business was running out of money. People were engaged, they liked their colleagues, they felt listened to – and the company was in real trouble, because engagement and value creation, it turns out, are not the same thing. We've simply grown used to treating them as if they were. To be fair, the marketing around engagement surveys is good.

It doesn't help that once a number matters, everyone begins, innocently, to behave towards it. Managers become a little warmer in the weeks before the survey. Small gestures are made. People's behaviours morph towards what is measured, which may not be what is needed, nor desired. Nobody has to be dishonest for this to happen – it's just what any measure does once something important is riding on it. Which means the survey may be capturing rather less about the year than about the fortnight immediately before it opened.

Try a small thought experiment. Imagine that tomorrow every engagement survey vanished – no annual, no pulse, no score, no red-amber-green dashboard. What would you do differently? If the answer is quite a lot, or nothing much, that's worth sitting with. There's no right answer. But it's a powerful to think about.

Perhaps the more useful question isn't how engaged are our people? Perhaps it's simply: what is it like to work here on an ordinary Tuesday? Can people make progress? Can they get a decision when they need one? Do they understand what actually matters? Can they do work they're proud of, and turn a good idea into something useful? Can they see that their contribution lands somewhere real? Can their manager see what's getting in the way? Are they trusted to make decisions? Do their insights, ideas and opinions get heard and acted upon? These things are much harder to fit on a dashboard – and closer to the thing we all actually care about – doing good work, improving it and being understood in the workplace.

None of this is a knock on HR, and it certainly isn't a case against surveys. I've spent time in a HR leadership, rolled out surveys myself and worked with plenty of great HR teams. A good HR function does things no single manager could: it builds the systems and conditions in which good work becomes possible, it supports managers who need supporting, and it spots patterns across a whole organisation that no individual could ever see. But engagement itself is made somewhere HR can't quite reach – in the work, in what happens between people every day: when obstacles get cleared, when priorities are clear, when feedback is useful, when a manager genuinely notices what's going on around them. Those are daily, managerial acts. A survey can point toward them; it was never going to produce them.

And that, when you sit with it, is rather good news. It means engagement isn't a substance to be extracted once a year and reported on. It's made – or missed – in ordinary moments, which puts it within reach of every person, on every ordinary Tuesday, without waiting for a dashboard's permission or a report to be produced. Listening still matters, surveys and all; they simply work best as one modest input into engagement, rather than a stand-in for it.

Because the purpose of work is not to produce a high engagement score. It is to turn ideas into value, and to feel that the doing of it was worth it. If we want people to care about that, we might spend a little less time measuring how engaged they are – and a little more making the work worth engaging with.

The survey isn't the point. The work is.