Why industry benchmarks are usually the wrong target

Somebody will always tell you the average conversion rate for your industry. It is worth asking where that figure came from, who paid for the research, and whether the businesses in it resemble yours at all.

Talk to us about your site

Published benchmarks are usually drawn from the customers of whoever published them, which is a self-selected group, and averaged across businesses of wildly different sizes and models. Compare against your own history instead, and against your own pages and channels, which are the only genuinely like-for-like comparisons available.

Where the numbers come from

Most benchmark figures are produced by companies that sell software or services, drawn from their own customer base. That base is self-selected: businesses already investing in the thing being measured.

This does not make the figures dishonest. It makes them unrepresentative of businesses in general, and particularly of a small local business that looks nothing like the average participant.

Why averages hide more than they show

An industry average blends businesses of different sizes, prices, markets and models. A conversion rate averaged across a business selling something inexpensive and one selling a substantial project describes neither.

The same applies to averages across channels. Traffic from a branded search and traffic from a broad informational query behave completely differently, and an average across them describes a business that does not exist.

The circular quotation problem

There is a further hazard worth knowing. Many widely repeated marketing statistics have been quoted so often that the original source is difficult to find, and occasionally does not exist in the form claimed.

Before using any figure, ask who measured it, when, and on whom. If those three cannot be answered, do not build anything on it, and certainly do not publish it as though it were established fact.

  • Who collected the data, and what do they sell?
  • When was it collected, and is it still current?
  • Which businesses were included, and do they resemble yours?
  • Is the original study findable, or only the quotation?

What to compare against instead

Your own history is the only genuinely like-for-like comparison you have. The same business, the same market, the same customers, measured the same way.

Within your own site, compare pages against each other and channels against each other. Those comparisons are meaningful because everything else is held constant, and they point at decisions you can actually act on.

Where benchmarks are legitimately useful

Two situations. When you have no history at all and need a rough sense of whether a figure is plausible. And when the number is a hard fact about a platform rather than an average about businesses.

Even then, treat it as a rough orientation. The moment you have three months of your own records, they are more informative than any published average about your industry.

The same discipline applies to what you publish

If you would not accept an unsourced figure in a supplier's proposal, do not put one on your own website. Statistics with no named source are the fastest way to lose the trust of a careful reader.

Where a figure is genuinely needed and cannot be verified, either attribute it clearly to whoever claimed it or leave it out. Both are better than presenting somebody else's marketing as a fact.

How Licheo handles this

We do not publish estimated results or industry averages as though they described your business, and we do not put unsourced statistics in proposals.

The comparisons we report are your own: this period against the same period last year, and one of your pages or channels against another.

Part of a larger guide

This page is one part of Revenue per visitor. The other parts:

Questions people ask

Are industry benchmarks useful at all?
Only roughly, and mainly when you have no history of your own. Most are drawn from the customers of whoever published them, which is a self-selected group, and averaged across businesses that may look nothing like yours.
What is a good conversion rate for my industry?
The honest answer is that the question cannot be answered usefully. It depends on what you sell, at what price, to whom, on which page. Your own site over time is the only meaningful comparison.
How do I check whether a statistic is trustworthy?
Ask who collected it, when, on which businesses, and whether the original study is findable rather than only the quotation. If those cannot be answered, do not build a decision on it.
What should I compare my results against?
Your own history, period against equivalent period, and your own pages and channels against each other. Those hold everything else constant, which is what makes a comparison mean something.
Should I use statistics on my own website?
Only ones you can attribute to a named source with a date. Unsourced figures are the fastest way to lose a careful reader's trust, and they age publicly once the year attached to them slips into the past.

Compare against something real

Your own history and your own pages, reported with sources named and no borrowed averages standing in for measurement.

Talk to us about your site Read the full guide