Posted September 18, 2012 by Bandish Patel in Tips

What Web Metrics to Track

When people start learning about Web analytics, it generally goes like this:

  1. They don’t know the difference between a hit and a pageview and don’t really care as long as they keep going up when they check their server log files.
  2. They then start learning the value of visitors and go out and find an analytics tool that will help them track that.
  3. The tool offers so many other features like tracking bounce rate and repeat visitors and so on that they start reviewing all the reports they can get.
  4. Finally, they realize that spending all their time reading statistics about their website is causing the pageviews (or hits) to go down and they give up on analytics in disgust. Or at least tone down what they look at.

I’ve been in all of those places with my websites, and what I’ve come to realize is this: analytics are important, but they’re not the most important part of my site.

One of my business teachers used to say “you get what you measure” – which is very true. You won’t see improvement in pageviews or visitors or anything else if you’re not measureing it. But if you let the measurement take over, you won’t get what you’re measuring either.

A New Path: Analytics Moderation

Instead of getting to step 4 in the path above and giving up, there is an alternate route you can take: moderation. Instead of trying to track and follow every stat on your website, you should determine which are the most important for your business and follow that one statistic only.

Yes, I said one statistic. If you’ve gone into stats junky mode, this can be very hard to do, but just following the one measurement that is most important to your business you can improve those results and grow your business, which is ultimately the goal of reading metrics.

What Metric Should You Track?

Track the most important metric for your business. This could be different for every company, but I’ll give you some examples:

  • This site on About.com is supported through advertising and those ads are supported by the number of people who view them and act on them. So the most important metric for this site is the number of visitors my site receives in a month. So that’s what I would want to track and that’s what I would want to improve.
  • An ecommerce storefront is supported by how many sales are completed. A great metric to track for an ecommerce site is how many people drop off without finishing their purchase. While you might think that visitors would be the most important metric, if you notice that there is a huge difference between visitors and customers with completed sales, you will realize that getting the people who come to your site to finish their purchase is more important.
  • A Web designer’s site might get a lot of pageviews and visitors, but very few people fill in the contact form. By tracking the paths through the website, the designer might be able to find out what is blocking visitors from becoming leads.
  • An information website gets most of it’s traffic through Google. Since inbound links are a huge part of high ranking in Google, tracking those links in order to improve them might be the most important metric for that site.
  • A social site might rely on other social media for promotion and customers. The more buzz this site gets on Twitter the more visitors come. So tracking Twitter leads might be the most important metric for this site.

The key is to stay simple. Track only one metric. If you’re not seeing improvements after 2-3 months, then perhaps you didn’t choose the most important metric. Don’t get hung up on what you used to track. Chances are you were on your way to tracking every single thing on your site, and that doesn’t help you improve it just helps you learn to read charts quickly.

Don’t Worry About Pageviews

Unless you are paid by the pageview, chances are they don’t matter nearly as much as other metrics you can track, like referals, visitors, entry and exit pages, and so on. It is very tempting to get hung up on straight pageviews because it’s so easy to understand, but ultimately, they don’t increase your bottom line (unless, as I said, you’re paid directly by the pageview).

For example, one month your site might get picked up by a very popular site in your genre. This could rocket your pageviews up to 10 or even 100 times your normal daily rate. In some cases this huge increase can even crash your servers or cause other unintended problems (this is typically called being slashdotted).

While this huge increase might seem cool, most of the time it doesn’t result in a large influx of new custoemrs to your site. The people riding the wave come in on a fad link, check out your site and leave, never to return. Yes, there were a lot of people doing that, but it’s more like being hit by a wave of locusts. And if you have any type of year-over-year comparisons, you’re going to be in trouble next year trying to beat the numbers from this year.

If you’re still inclined to track pageviews, just humor me for 2 months and track visitors anyway. If that’s too far away from pageviews for your comfort, then track pageviews per visit or pageviews per session. If you can increase the number of visitors to your site or get your current visitors to go to more pages in one visit, then your pageviews will grow anyway.

Bandish Patel