The best way to measure employee engagement

A chap over on Quora posted a question asking what the best method to gauge employee engagement was. In answering I took a slightly different perspective because I believe engagement itself is pretty hard to quantify with a method. Engagement is a state of mind, by nature something that cannot be absolutely observed or determined externally. In lab conditions I’m sure certain sets of answers correspond with a model described level of engagement, but in the real world you’re contending with all sorts of human filters that will introduce bias as people do (or don’t) give the answers they think they should.

If a survey comes out from the boss or HR, are employees really going to be honest about an answer that might cost them their job? At the very least, a sense of duty or loyalty to the team and company will probably inhibit ‘pure’ responses.

It’s much easier and instructive to look at the signs and symptoms  – the behaviours expressed that will show the lack or presence of engagement. By looking at actual behaviours, rather than trying to determine state of mind, instinct and experience tells me that they are likely to be a more accurate predictor of engagement.

Here’s five examples:

  1. Anyone, at any time saying the immortal words “that’s not in my job description”.
  2. Look at the level of absenteeism. High levels of absence are a symptom that people need smaller excuses not to show up for work. If they were more engaged they’d want to be working and finding ways to show up despite their issues.
  3. Analyse your error rates/corrections/bugs/returns/escalations or whatever your enterprise qualifies as something going wrong (the lean movement would call this ‘failure demand’ because it’s a wasteful demand on your resources derived from a failure to do things properly). Errors and problems in delivery are a symptom that people don’t care enough to get it right the first time and likely don’t care enough to try and correct the underlying issue that gave rise to the problem in the first place.
  4. Track how many suggestions for improvements and how distributed they are across your teams. If no-one is suggesting any improvements no-one is engaged to care enough. If only a small fraction of the employees are generating ideas to make things better then you only have pockets of engagement.
  5. Stand in the middle of your workplace just before 5.30 (or whatever time ‘official’ hours of work end). If on the stroke of 5.30 everyone gets up and starts leaving you have a problem. Unless they’re all super-organised machines of efficiency, it probably means they’ve been clock  watching since 5pm, winding down, not starting anything, started packing  their bags at 5.15, shutting down the computer at 5.20, waiting for the  instant they can be out the door ‘legally’. Engaged people don’t clock-watch because the time they finish isn’t important – doing the job right is.

There’s also the added benefit that signs and symptoms – observed behaviours – are easier to talk about and address because they’re out in the open rather than a state of mind. e.g. “I saw you earlier throwing a report across the room, how come you’re that worked up?” vs “Why are you pissed off and not engaged with this project?”. The former gives you an ‘in’ that can’t be refuted (everyone saw the report fly across the room) and allows the employee to unpack their thinking on it. The latter assumes what’s going on in their head and pre-judges, which leaves much less room to address the underlying issue(s) in a way that won’t feel confrontational.

By getting people to talk about expressed behaviour first, you can get to the motivations in a more constructive way. If you pile in trying to talk about their inner world, making assumptions and judgements, all you’ll do is get their back up and make things worse.

All this alternative ‘method’ of determining employee engagement requires you to do is undertake is one of the most valuable, insightful functions as a manager: simply shut up and observe. Watch and listen closely and you’ll learn tons.

If that doesn’t work you can always take a leaf out of Amazon’s book and simply ask “How much should you offer employees to Pay to Quit?” I’m guessing the size and range of numbers you get back would be a pretty good indicator!