Math is the curse of the Internet.
True, it’s the most trackable medium ever devised and what you can unearth by sifting through the sand dune of data created by every Web site is startling. Just about everything you’d ever want to know about your organization’s Internet success or failure is viewable numerically, if you know where to look.
The problem is that the Internet has no self-control. It’s like a freshman at a fraternity party. If some is good, more must be better. If one ad on a Web page is good, why not 10? If a handful of e-mail ads are OK, why don’t we just spam everyone into oblivion? If five pieces of data can be crammed into a statistics report, why not 105? All this overkill is like Anna Nicole Smith; self-destructive, annoying, and difficult to
understand. (author’s note: This was written long before the tragic passing of Mrs. Smith)
To try to make sense of all that math and figure out what’s happening on their Web site, many companies examine statistics reports generated by their in-house team or Internet services provider. But the sheer magnitude of the available data makes it tough to fathom. It’s like reading X-rays without training. My wrist could be broken in four places, but until Dr. Radiology breaks out his laser pointer and circles the fractures, it would look like all systems go to me.
People responsible for Web site success are often perplexed by these statistics reports that have the intuitiveness of cave drawings. What’s important? What’s irrelevant? What does it all mean? So, one of two things happen: A quick, Grenada-like surrender (many companies tell us that they have access to site statistics but never look at them); or a fixation on the number that’s the easiest to understand – the visit.
But the problem is that visits are the most overrated statistic in the land of the Internet. The number of visits has nearly no bearing on the success of your Web site. The presence of a person on your site doesn’t mean he or she will buy something, fill out a lead form, use your online customer service tools, or do anything else that has a tangible financial benefit to your organization.
If you own the pet store at the mall and have cute little puppies in the window at the front, how many people stop and look at the puppies every day? The lab-grade sea of bacteria created by everyone pressing their noses up to the glass indicates that indeed a lot of people take a peek. But how many of those folks actually come in the store and buy something? A very small fraction.
Focusing on Web site visits is like focusing on puppy lookers. Interesting, but not a barometer of success. Instead, of quantity, focus on behavior. What did people do once they came to your Web site? That’s what counts.
To appropriately focus on behavior, you need to have access to data that assesses and evaluates activity of your site visitors. As recently as one year ago, using math to understand the behavior of your Web site visitors was like heckling Randy Johnson ¬ only for the courageous. But now, there is a new tool we recommend for this effort. For approximately $600 to $2,000 and a few hour learning curve (we can show you how), an inexpensive software program called ClickTracks can give you all the behavior and results-based metrics you could ever want.
Instead of just telling people in your company how many visits your site had, you can throw out fascinating tidbits like, “7.5% of the people coming to the site from the banner ad campaign filled out the lead form, compared to 12.2% of the people coming from the e-mail program.” Statistics like those describe results and effectiveness, and since the whole point of a Web site is to make money, save money or both, those are the numbers that matter.
A good rule of thumb is that when evaluating online behavior, look for metrics that are percentage or ratio based, rather than whole numbers (like visits).
Pay attention to numbers like “Length of Visit” which measures how long people spend on your site. Are they engaged and looking at multiple pages, or are they horrified or confused by the first page they see and thus quickly traveling elsewhere. For the record, an average visit of less than two and a half minutes merits examination.
If there are certain pages or tools on your site that are important to your company, measure how many visitors go to those pages. This is called the “Acceptance Rate.” If your whole Web site is based on people using your online demo, and fewer than 15% of your visitors ever see the demo intro page, measure your Web designer for a noose necktie. Acceptance Rate shows the percentage of people that see something.
A related metric is “Conversion Rate” which measures the percentage of people that do something. The percentage of visitors that complete your online demo and fill out the lead form, for example.
You should also investigate what percentage of your site visitors came to the site through the front door. This is an important measurement if you put all of your news and special offers on the homepage only. You’d be surprised how many people never see the homepage—usually 25% to 45% on most sites.
Of course there are nearly infinite varieties of other behavior metrics you could examine. But to get a good snapshot of what’s really happening on your Web site, just figure out the three to five tasks that your organization would want site visitors to do, then find the ratios that describe how often those tasks are indeed being accomplished.
The Internet isn’t doing you any favors with its fire hose of data. But, if you drink from that hose with a straw built from meaningful, results-oriented, behavior-focused numbers instead of less important metrics like visits, you’ll have a much clearer understanding of how your site is really doing, and how to improve it over time.
