Employee engagement benchmarks: Australia & New Zealand

Key takeaways

• A single external benchmark tells you almost nothing about your organisation on its own. It's a comparison point, not a verdict.
• Benchmarks get built from very different companies, industries and question sets, so "the industry average" is softer than it sounds.
• Your own trend, over time and broken down by team, tells you far more than any outside number ever will.
• The right use of a benchmark is as a conversation starter with your leadership team, not a scoreboard to chase.
• If a number makes you feel either smug or panicked, that's usually a sign you're reading it as a verdict instead of a starting point.

A CEO once showed me his engagement score and asked, with real concern, whether 71 was good. I asked him good compared to what. He didn't have an answer, and to be fair, most people don't.

That's the trouble with benchmarks. Everyone wants the number. Almost nobody asks what it's actually being compared to, or what to do with it once they have it.

Why chasing a single number misleads

Benchmarks look precise. A published figure, a percentile, a clean comparison against "the industry." That precision is mostly an illusion, and it's worth understanding why before you let one number change how you feel about your organisation.

The comparison group is rarely who you think. "Industry average" often blends companies of wildly different sizes, maturity and geography. A benchmark built mostly from large multinationals tells you very little about a 40-person Australian business.

The question wording varies. Different survey providers ask different questions under the same label of "engagement." A score of 75% on one instrument and 75% on another can reflect genuinely different things.

Self-selection skews the dataset. Organisations that contribute to benchmark databases tend to be the ones already investing in this area, which quietly shifts the "average" upward.

A single number hides everything interesting. An organisation-wide score of 71 could mean everyone is at 71, or it could mean one team is at 90 and another at 45. The average erases exactly the information you'd need to act.

None of this means benchmarks are worthless. It means a benchmark answers a much narrower question than most people think it does, and treating it as a verdict on your organisation is where the trouble starts.

What to compare against instead

If an external number isn't the right anchor, what is? In my experience, three comparisons matter far more than any published benchmark.

Your own trend over time. Are you moving up, down, or flat, compared to your own last three or four readings? That trajectory tells you whether what you're doing is working, which no external number can, because it doesn't know what you're doing.

Team by team, not just organisation-wide. A single company-wide figure hides the variation that matters. Two teams sitting ten points apart is a far more useful signal than either team's distance from an external average.

Before and after a specific action. If you make a visible change, redistribute a workload, fix a broken process, the most useful benchmark is your own score before against your own score after. That comparison tells you whether the thing you did actually worked.

This is really the same idea as the model I set out in workplace culture: what it is, and how to actually change it: listen continuously, act visibly, close the loop. Your own trend is what lets you see whether the loop is actually closing.

Using a benchmark as a conversation starter, not a scoreboard

There's still a legitimate use for an external benchmark. It's just not the one most leadership teams reach for.

Bring a benchmark into the room to prompt a conversation, not to settle one. "We're a few points below what's typically reported for our sector, what do we think is actually going on here?" is a useful question. "We beat the benchmark, so we're fine" is usually the wrong conclusion, because it stops the conversation right when it should be starting.

The healthiest way I've seen leadership teams use a benchmark is almost as a permission slip. A number that's below where they'd like gives them the internal cover to ask harder questions of their own trend and their own team-by-team data, which is where the real answers live. Used that way, the external figure has done its job even though it was never precise enough to be the answer on its own.

Be equally sceptical of a benchmark that flatters you. A comfortable score against an external average can quietly stop a leadership team asking the harder question of whether one team is well below the others, or whether the trend over the last year has actually been flat.

A quick way to check your own numbers properly

Next time you're looking at an engagement or culture result, before you ask "is this good," ask these instead.

• Compared to what? Your own last reading, a specific team, or before and after a change, not an unnamed industry average.
• What's hiding inside the average? Break it down by team, tenure or role before you decide what the number means.
• What did we do since the last reading? If nothing changed, don't expect the number to have moved, whatever a benchmark says is typical.
• What's the smallest visible action we could take on this, this month? A benchmark conversation that doesn't end in an action is a benchmark conversation that wasted the room's time.

Consistent, well-worded questions asked regularly are what make this kind of comparison possible in the first place. I've set out what makes a question worth asking, and a full bank grouped by theme, in pulse survey questions that actually tell you something.

Frequently asked questions

Where do published engagement benchmarks come from?

They're typically built from the aggregated results of organisations using a particular survey provider or instrument, which means the dataset reflects whoever happens to be a customer of that provider, not a random or complete sample of the workforce. That's worth knowing before you treat a published figure as an objective market rate.

Should we stop looking at external benchmarks altogether?

No, they're a reasonable sanity check and a useful conversation prompt. The mistake isn't looking at one, it's treating it as more precise or more meaningful than it actually is, or letting it substitute for tracking your own trend.

How often should we check our own trend?

Regularly enough to catch a shift before it hardens into a bigger problem, which usually means more often than once a year. I've written about why the annual model struggles with exactly this in annual engagement survey vs continuous listening.

What's a realistic score to aim for?

I'd push back gently on the question itself. A fixed target score, chosen without reference to your own history, tends to encourage chasing the number rather than fixing what it's telling you. Aim for a trend moving in the right direction, off your own baseline, rather than a specific figure borrowed from somewhere else.

Next time a benchmark lands on your desk, what will you actually do differently because of it, and if the honest answer is nothing, what was it for?

Want to see how continuous, team-level data replaces the single annual number entirely? Have a look at Whyser Culture.