Benchmarking Against Yourself

Where a published average comes from

An industry benchmark is calculated from whichever businesses a company had data about, which is usually its own customers.

Those are the businesses large enough to buy that company's product, in whichever markets it sells, measured however its tool counts.

So an average conversion rate for trades describes a set of businesses you are not part of.

Comparing yourself to it tells you how you differ from that sample, which is not a question anybody asked.

The four reasons it does not transfer

• Scale: their traffic is not yours.
• Definitions: a conversion means something else.
• Market: a different country or city.
• Selection: only businesses that measure.

The second does most of the damage. A published conversion rate frequently counts newsletter signups, downloads, and video plays alongside purchases, so a business counting only enquiries appears to be performing far worse than it is.

The comparison that works

Which is the same period a year earlier, in your own figures.

Same month, same definitions, same site, same market, same business.

Everything that would confound a comparison is held constant except the thing you want to know about.

That is the only comparison available to a small business that answers a real question.

It requires a year of records before it works, which is the reason to start keeping them rather than a reason not to.

In the first year, record and resist concluding, which is difficult and correct.

Anybody offering a verdict on a business with eight months of data is guessing with a chart attached.

A worked example

A business was told their conversion rate of about two per cent was below the industry average of four.

Their two per cent counted enquiries only, on a site with no downloads, no signups, and no shop.

The published four per cent came from a survey of businesses counting several kinds of interaction, most of them considerably larger.

Against their own previous year, the same figure had risen from about one and a half.

So the honest reading was that the site had improved by a third, and the benchmark had suggested it was failing.

They stopped using the external figure entirely.

Compare against your own segments too

Which is the useful internal version.

Your service pages against each other, since one converting at twice the rate of another is a fact about your pages rather than about anybody else's.

Mobile against desktop, which frequently reveals a real problem.

Enquiries from search against enquiries from your listing, since those behave differently.

Each of those comparisons holds everything constant except the one thing being compared, which is what makes it readable.

They also point at a specific action, which an industry average never does.

The only useful external comparison

Since not all of it is worthless.

A direct competitor's site, read as a page rather than as a statistic, tells you what they publish that you do not.

That is observation rather than benchmarking, and it produces a list you can act on.

Their review count against yours is also directly comparable, since it is the same number measured the same way in the same place.

Both are specific to your market and neither requires trusting somebody else's sample.

Beyond those two, external comparison has very little to offer a small business and a great deal of confusion to introduce.

Watch for benchmarks in a sales conversation

Since that is where they mostly appear.

A figure showing you underperforming an average is a persuasive opening and it is rarely offered with its methodology.

Ask what the sample was, how the metric was defined, and which market it covers, which is a fair question with usually no answer.

Then ask what your own figure was a year ago, which is the comparison that would justify the work if anything does.

A supplier who has looked at your own history is worth listening to and one quoting an industry average has not.

Targets beat benchmarks

Which is the constructive alternative.

Rather than comparing to an average, set a figure you would like to reach and a date.

Base it on your own trend rather than on anybody else's performance: a fifth more enquiries than last year is a target, and matching an industry average is not.

That gives you something to measure against that is achievable, relevant, and yours.

And it survives every change in how anybody else measures anything.

Your own history is a benchmark somebody else cannot see

Worth stating as the positive case rather than only as a criticism of the alternative.

Three years of your own figures is a comparison set nobody else has, covering exactly your market, your customers, and your definitions.

That is a genuine advantage over a competitor working from published averages, since you can tell whether a change helped and they cannot.

It also compounds, in the sense that the fourth year is more informative than the second and the tenth considerably more than either.

A business that started recording in 2020 is now able to answer questions that no amount of money can buy an answer to.

The counter-case

Some external figures are informative.

Knowing that mobile is a large majority of traffic in your sector, or that response times under an hour convert better, is generally true enough to act on.

Very new businesses also have no history to compare against and need some external reference point.

And a figure wildly out of line with every published range is worth investigating rather than dismissing, since it occasionally indicates something genuinely broken.

Compare against the same month last year, compare your own pages against each other, and treat industry averages as sales material.

What to do

1. Use the same month last year.
2. Keep the definitions constant.
3. Compare your own pages against each other.
4. Compare mobile against desktop.
5. Ask for the methodology behind any benchmark.
6. Count competitor reviews, which is comparable.
7. Set a target rather than matching an average.

Step five ends most of these conversations, since a published average offered without its sample and its definitions is a marketing figure rather than a measurement.

Recording something to compare against is covered in setting a baseline you can compare to.

Frequently asked questions

West Coast Media Solutions Inc. provides web design, web development, hosting, digital marketing, and business consulting to organisations across Canada, drawing on more than twenty-five years in the field.