“Engaged employees are productive employees.” It is repeated so often at HR conferences that it passes as a fact. Then comes a quarter where the survey says 82% favourable and delivery numbers are down 15%, and the belief comes under strain. Let us take it apart properly, because the way you resolve this tension decides whether you waste a year on the wrong fix.

Myth one: the survey score measures how hard people work

Most engagement surveys ask how people feel: whether they would recommend the company, whether they feel valued, whether they trust leadership, whether their manager is supportive. Those are sentiments. They say something about the climate, and that is not the same as measuring output. A team can be pleasant, trusting and relaxed, and also be working on the wrong priorities, with too few skills, using slow tools, or with unclear targets. Warm feelings and results can sit side by side.

There is a genuine body of research linking engagement with performance at unit level, and consultancies such as Gallup publish such claims. But a statistical link across many units is a long way from saying that, in your team, this quarter, one explains the other. Averages do not tell you about your case.

Myth two: high scores mean people are happy with everything

Look at who answered and how. If the response rate is 55%, the missing 45% may be the people least happy. If the survey is not truly anonymous in a small team, people give safe answers. If questions are generic (“I am proud to work here”), they score high even when the working conditions have changed. And employees who like their colleagues and their manager will answer positively about the company, even when they are quietly overloaded.

Also check the date. A survey taken in March describes March. If a big client changed requirements in April or a reorganisation happened in May, the score is describing a company that no longer exists.

Myth three: if productivity fell, engagement must be the cause

Here the direction of cause can run either way, and often something else drives both. Productivity can fall because of things engagement does not touch.

  • A skills gap after a technology change, so the work genuinely takes longer.
  • Rising workload with the same headcount. People remain loyal and keep trying, but output per person falls once the team is stretched thin.
  • Process problems: approvals stuck, tools slow, handoffs unclear.
  • A change in the work mix, such as more complex tickets or smaller deals, so that the metric itself is no longer comparable.
  • High turnover of experienced people, replaced by new joiners still ramping up.
  • A measurement artefact: someone changed how output is counted.

Reverse causation is also real. Teams that perform well often feel good, and then report high engagement. So the survey may be a result and not a cause.

How to investigate

Start by checking the productivity data itself. Is the metric defined the same way as last year? Has the mix of work changed? Is the fall everywhere or in specific teams, clients or locations? A drop concentrated in one account points to something in that account.

Next, split the engagement scores by the same cuts. Look at the individual questions, not just the overall index. If “I have the tools and resources to do my job” or “my workload is manageable” scores lower than the glowing headline, there is your lead. Engagement indices often bury the awkward items under the comfortable ones.

Then talk to people. Short, small-group conversations with managers and employees from the teams where output fell, asking what has made work harder in the last six months, will give you more than another survey. Pair this with some hard data: overtime hours, leave patterns, internal transfer requests, time-to-resolve and rework rates.

What to take from all this

Engagement is worth measuring, but as one input among several. If you want to talk about productivity, measure productivity, defined carefully, and track it alongside the survey and not in place of it. When the two disagree, treat the disagreement as the interesting finding. It usually means that the survey is answering a different question from the one the business is asking.

Figures and names in the examples are illustrative.

Leave a Reply

Trending