How to Avoid the Common Pitfalls in Web Analytics

MoreVisibility - November 4, 2010

Did you ever play the game “Pitfall” on your Atari 2600 back in the 80’s? I sure did! I remember borrowing it from a neighbor when I was about 6 years old, and I remember not wanting to give it back. I would split my video-game time between that and “Asteroids”, and for the record, I did score more than 20,000 points on “Pitfall” once (true 80’s geeks will know exactly what I’m talking about).

“Pitfall” for Atari 2600 was all about avoiding traps, holes, alligators, rolling logs, and basically anything that you could possibly think of while you desperately tried to collect gold bars, jewelry, and bags of money in under twenty minutes. Fast-forward 25 years into the future, I find myself trying to avoid traps and pitfalls in web analytics to help clients make more money and become more profitable – all of which in seemingly much less than the twenty minutes allotted in the popular 1980’s video game.

Pitfalls in web analytics come in all shapes, sizes, and colors, and it’s very difficult to dodge them all without bumping into one from time to time. The three most common pitfalls that I encounter in web analytics are listed below in the form of statements and questions that could easily lead to your demise.

Pitfall #1: “How many hits did we get?”
Why it’s a pitfall: If you’ve ever been asked this question and have answered it without even critically analyzing the question itself, then you fell into the mother of all pitfalls that the web analytics industry faces today. In 2010 – and as we move into the thick of the next decade – it’s just not going to cut it to report on “Hits”, “Visits”, “Uniques”, or whatever statistic needs to be reported on.
How to avoid it: You have to give your data some context, you have to segment and compare your data over time, and you absolutely have to be able to retrieve some sort of meaningful, insightful information from your data. If you – or the person requesting it – has their heart set on getting the total number of uniques, try wrapping some data around that figure to give it some background and shine it in a different light.

Pitfall #2: “Everyone is leaving my website from the home page!!”
Why it’s a pitfall: People have to leave your website at some point. Visitors can’t possibly stay on your site forever, so there will be a time that your site’s visitors will need to get back to work, go to lunch, or go to sleep. Since your home page is most likely responsible for the largest share of page views, it’s not farfetched to assume that your home page will also account for the highest percentage of website exists. It doesn’t mean that your home page is bad, ugly, or broken – it simply is what it is.
How to avoid it: Instead of focusing on the number of exits from your home page or any page on your site, focus on the number of bounces and the bounce rate. A bounce is a visitor who views only one page on your web site, and leaves your site without visiting another page. Bounce rate is not a perfect measure of a page’s appeal or performance, but it’s a thousand miles ahead of measuring exits. With bounces and bounce rate alone, you can at least point yourself in the right direction and know which page(s) need your attention. After all, you do want visitors to view more than one page on your site, right? Start by picking out a couple of important pages with higher-than-average bounce rates and determine ways to optimize that page to lower that percentage.

Pitfall #3: “My conversion rate should be 2.50%!”
Why it’s a pitfall: This pitfall is two-fold. On one side of the coin, a bar has been set for a statistic which may never be reached (there are many reasons why). On the other side of the coin, this train of thought usually leads down the primrose path of boxing oneself in to only caring about those 2.5% of visitors who do convert. What about the other 97.5% of your web site’s traffic – will you be ignoring this massive majority because they don’t fill out the inquiry form or buy something from your online store?
How to avoid it: The truth of the matter is this: people visit your website for many different reasons (emphasis on the word “many”). One of those reasons is to shop online or fill out your inquiry form. Other reasons could include researching your company, reading your articles, subscribing to your blog, following you on Twitter, downloading your annual share-holder report, upgrading to browser version 3.5.12, writing a review for your product, finding out how to get their money back…you get the idea (hopefully not the last reason, but you get the point). You can’t just measure one desired outcome on your website to know if you’re succeeding or failing online – you must set up multiple Goals / Outcomes / Conversions to understand how the visitors to your website are engaging and interacting with everything you have to offer. There is a great hidden value in learning what else people are doing on your website, and it goes far beyond the ultimate purpose of your website.