Are unique visitors a meaningful measure of your website's traffic?

Are unique visitors a meaningful measure of your website’s traffic?

Brian Clifton – author of the so-good you must go out and get a copy immediately Advanced Web Metrics with Google Analytics
(seriously, I made my whole team read it) – says counting unique visitors is meaningless. Despite his talents and reputation, I’m going to have to disagree with Brian here. Unique visitors can be tremendously useful to your business regardless of its inherent tracking issues Brian so rightly calls out. And I’ll explain why in just a moment.

More importantly, ignoring data simply because it’s “imperfect” is a huge mistake in business. I’m surprised to hear Brian suggest otherwise. You can use unique visitors to improve your business decisions. But it’s even more important you use this as an example of how to gain meaning from imperfect data and apply that knowledge to improve your decisions. And your business.

First, we have to start with some assumptions:

1. Any measure that reduces your uncertainty is better than no measure at all – This comes from Douglas Hubbard’s spectacular “How To Measure Anything: Finding the Value of ‘Intangibles’ in Business.” Hubbard argues – successfully – that measures exist to help you make decisions. Throwing out imperfect data that can help you make better decisions is pointless. Always. If the data doesn’t help you make decisions, than Brian’s right – you should ignore it. But your measure of unique visitors can usually help you make better decisions.
2. Unique visitor measures cannot be greater than visits. But visits can be higher than unique visitors – This is key. At most, your unique visitor metric will be the same as your visits metric. More often, it will be less. Tracking the relationship between these two provides a wealth of meaning to most businesses.

Mathematically stated, visits always must be equal to or greater than unique visitors. If every unique visitor cleared cookies every time they visited your site, or used a different browser, or a different computer, your ratio of visits to unique visitors would always be 1:1. But that’s not the case, is it? If you can’t grow visitors without also growing visits and you can grow visits without growing uniques, you have the basis of a measurement – and an opportunity to gain an edge on your competitors.

Let’s look at a real-world example. This site averages about 1.2 visits/visitor. Now let’s assume I have a 15% increase in monthly visits. If the visits/visitor metric holds steady at 1.2, then it’s likely I’ve seen a true increase in unique visitors. But if even as few as 10% of those new visits were redundant (i.e., the same visitor used multiple browsers, multiple computers, etc.), my visits per visitor metric would fall. Admittedly, not by much at first – just over a 1% decline. But it would fall. And if the trend continued, I’d have a sense something was going on. By contrast, if visits continued to climb at a faster rate than uniques, then clearly I’m getting more visits per visitor, which could indicate customers taking longer to decide, an improvement in “sticky” content, or any number of things, depending on my business.

Why does this matter? If your business relies on lead generation, such as asking visitors to fill out a form requesting a quote, you can’t expect every visit to result in a lead. For many businesses, you would only expect a customer to submit a form one time. Measuring visits gives an unrealistic picture of your business opportunity, while unique visitors – despite its flaws – can be a better measure because it reduces the uncertainty associated with your traffic.

The flaw in Brian’s reasoning reminds me of a great story in Hubbard’s book, in which he asked a group of executives how they would find out how many fish were in a recently re-stocked lake. After some thought, one well-meaning exec suggested draining the lake. Which is fine, if you want to find out how many dead fish you have. Instead, the best answer is to sample the population, tag some fish, then sample again. The percentage of tagged fish in your second sample should be about the same as the percentage in the lake overall and lead you to a solid – if imperfect – answer.

In this case, unique visitors is a bit like Hubbard’s tagged fish. They may not give you a perfect answer. But, if they help reduce the uncertainty around your business decisions, they just might be the right answer. Don’t throw them away like dead fish. Leave that for your competitors.

Are you getting enough value out of your small business website? Want to make sure your business makes the most of the local, mobile, social web? thinks helps you understand how to grow your business via the web, every day. Get more than just news. Get understanding. Add thinks to your feed reader today. Or subscribe via email.

And while you’re at it, don’t forget to follow me on Twitter.

Image credit: Mike Johnston via Flickr using Creative Commons Attribution 2.0 Generic. Mike Johnston’s images available here.

Technorati Tags:
analytics, douglas hubbard, douglas hubbard review, how to measure anything, measurement, metrics, online marketing, Brian Clifton