What a benchmark actually is
An industry average is a number derived from a sample you cannot see, of companies you cannot identify, measuring something they each defined slightly differently.
Utilisation is the clearest case. One firm counts billable over contracted hours. Another counts billable over available hours after leave. Those produce figures ten points apart for identical behaviour. An average across both means nothing, and it is published to two decimal places.
It is comforting in both directions
Above the benchmark, you relax. Below it, you have a project. Neither reaction is warranted, because the comparison did not carry the information to justify it.
The damage is worse in the reassuring direction. A firm at the industry average for margin, on work that is quietly getting harder, has been given a reason not to look at the trend that would have told it something.
Your own history is a better comparison
You control the definition, so it is consistent. The sample is you, so the mix is right. And the question it answers is the one that matters — not whether you are typical, but whether you are getting better or worse.
Four quarters of your own utilisation, measured the same way each time, is more useful than any published figure. It has a shape, and the shape is actionable. An average has neither.
What we do publish, and what we will not
We will not publish industry averages, and we will not build a feature that compares your figures to other customers' — even anonymised and aggregated. It is a popular feature and it is the same comfort in a nicer wrapper, with the added problem that we would be aggregating data people gave us for a different purpose.
What we do is show your own history with the definition visible, so a figure can be checked rather than trusted.
If you want a comparison, use the only honest one available: what you did last quarter, measured the same way. It is less satisfying than knowing where you sit in the industry, and it is the only one that tells you what to do on Monday.