You may, (or may not) recall that a couple of weeks ago in “The Basics of Analytics“, we started to look at Google Analytics, trying to decipher what these metrics really mean. Below is part two of that post.
As the name would imply, this is the rate at which people end up on one of your pages then ricochet off to another site entirely. The bounce rate is the percentage of people who view one of the pages on your site, then immediately exit. So… by definition, the number of pages per visit for “bouncers” is going to be a grand total of … one.
You can assume that if they land on one of your pages then jump to another site entirely, they didn’t get what they wanted, or landed on one of your pages by accident.
If you find that you have a high bounce rate, you likely need to work on your content, ensuring that on every page you have resonating copy that provides a reason to continue to explore.
Average time on site
Well… if this one is a struggle to understand, you might as well pack it up.
As you would assume, Average time on site means, very simply, the average total length of time that each visitor spends on your site. How do you break this down and what does it mean?
It could mean one of two things; a) you have a really interested prospect sitting on your site for ten minutes, genuinely engaged and enthralled by all of the information you have presented. Or b) your visitor is completely lost, desperately trying to find relevant information.
And how can you tell the difference? Look at your pages visited against total time on site. If they’re spending thirty seconds each on twenty different pages, they’re likely not all that engaged and may not be finding what it is they’re looking for. You would much rather have a page or two where your visitor is spending several minutes actually collecting and digesting the information you have provided. If you have twenty minutes on one page with little content, your visitor suffers from narcolepsy.
% New Visits
As the grouping name would suggest, % of New Visits tells you what percentage of your total visits are from new visitors.
Is a high percentage of new visits a good thing? Of course it is. It means that word about your site is travelling. Of course, if your site is one where you want continual revisits and your % of new is 99… that’s not a good thing. At all.
Make note that the new visit percentage is based on IP addresses. If Bob clears the cookies from his browser and visits your site, his visit will count as new. The reality is that people do, in fact (or should) clear cookies out of their browser with some frequency, so this number tends to be lacking in guarantee.
The best use of this number is as it relates to promotions – internet or otherwise. If you’ve engaged in a Google Adwords program, this number absolutely must increase, and should increase immediately. Likewise, any traditional advertising where your website address is promoted should result in a heavy increase in new visit percentage. If not, your promotion isn’t working. Plain and simple.
And there is your basic lesson on Google Analytics. My next post will go into more depth, covering some of the more granular tools that Analytics offers and (hopefully) helping you to decipher the information that will be of most help to you.
“If we wanted to figure out if a customer is pregnant, even if she didn’t want us to know, can you do that?”
This was the question posted to Target statistician Andrew Pole, as recounted in the New York Times article “How companies learn your secrets”. New parents have always been a boon for “one-stop” retailers like Target, as this life-change offers a small opportunity to alter engrained buying habits. “Coming in for diapers? Might as well pick up some supper too.” The competition for new parents’ attention is so fierce that Target wanted to get a jump on their competition and begin their marketing efforts before baby was even born.
And Pole delivered. He analyzed women’s purchase history and identified 25 products (pre-natal vitamins, maternity wear, etc.) that when considered together, generated a “pregnancy prediction score”. This allowed him to estimate who was a mommy-to-be, and approximately when she was due. Target then created custom “mommy-themed” flyers and coupons, disguised as regular flyers, and sent them specifically to these women, who were unaware of their data-gleaned origins.
And Target is not the only retailer doing this. As the article explains, “Almost every major retailer, from grocery chains to investment banks to the U.S. Postal Service, has a “predictive analytics” department devoted to understanding not just consumers’ shopping habits but also their personal habits, so as to more efficiently market to them.”
Does this story give you pause? Or does the prospect of a tailored-deal excite you? Either way, just wait. Those predictive analytics departments are about to get a whole lot more in-your-face thanks to mobile wallets.
The internet has been awash with talk of mobile wallets. (For the uninitiated, Jen’s post ‘The Next Step’ provides a great introduction). People are quick to talk about their convenience, the “coolness” of the technology, their lighter pockets. But less often do they mention the real reason for their existence; marketing.
A few weeks ago, I read an article that explained how RBC is poised to be the first Canadian bank to offer mobile wallets, possibly as a soon as October. Interestingly, this article also spoke at great length of the marketing advantages of the technology.
Retailers will now be able to collect purchase-data more easily than ever before, tailoring promotions to specific consumers. Combine this with your smartphone’s location-aware technology and retailers can now text you their latest flyer when you are near their store. Or alert you of their lunch special.
There are obvious advantages for both merchants and consumers alike, and it might be nice to walk into your favourite store and be provided with a list of on-sale items, or a coupon. We won’t know how this latest evolution of analytics will play out, but one thing’s for sure – we’ll all be watching. Oh, and so will they.
Do you think predictive analytics improve the retailer/customer dynamic? Or, are they an invasion of privacy?
“My website is getting a ton of traffic. I’m doing something right!”
You’re right. Maybe.
Here’s the catch. If you’re selling peanuts, don’t pat yourself on the back because you managed to enter the zoo. If you haven’t reached the elephants, it’s all for naught.
So… what is good traffic vs. bad traffic?
This is a little easier to distinguish. Good traffic comes in the form of people who are engaged. Bad traffic comes in the form of a visitor who stays for .01 seconds, and has absolutely no interest in going beyond that. Thusly, one can assume they have absolutely no interest in your product.
Google Analytics provides a wealth of data that allows you very easily determine how you’re really doing in relation to reaching your audience; the elephants.
This is the total number of new sessions that have begun on your site in the time period queried. This is not the number of people who have viewed it. If Bob enters the site, leaves to make a sandwich, then comes back, it’s two visits. One person.
This is the TOTAL number of pages that were viewed, regardless of who viewed them. Similar to Visits, if Bob decides to look at one of your pages, exits your site, makes a sandwich, then enters your site to see that same page again, it’s two views.
Here’s where it gets important. This is the first piece of data that tells you whether or not you’re getting to the elephants.
Pages/Visit tells you – as you might assume – the average number of pages that are viewed on any given visit. Like Visits and Page Views, it could mean that it’s Bob looking at two pages, leaving, then looking at the same two pages. So it’s really one person. But that’s not a bad thing. It means that Bob likely has at least moderate interest in something you’re selling. More than likely, it’s two people who have a level of interest beyond looking at your homepage. That suggests engagement. And quality.
If you find that your pages/visit is at 1.01, and it doesn’t move much, it means one of two things:
a) you’re being heavily targeted by a funky and evil machine sitting in a foreign land (often China or India). It has no interest in peanuts. It couldn’t care less how good they are. It doesn’t even know you sell peanuts, actually. It’s sole purpose is to collect information about your site. It’s searching for cracks to get through. It’s collecting email addresses that reside on your server. It’s considering you a prime candidate for future spamming. Though it’s big, grey, and ugly, it’s not an elephant.
b) Your content is lacking. Your homepage message isn’t resonating with visitors. You’ve managed to attract them somehow, but something was lacking in your homepage copy. They’ve obviously not found a reason to go deeper into your product. And that’s a problem. If you find yourself with great traffic numbers and low pageview per visit, it simply means that you’re not enticing anyone with the language you’re using.
Taking a few minutes to analyze not how many visits you’re getting, but starting to understand what’s happening once visitors are there will give you clear evidence of whether or not your site is “working”.
The cost of leaving your site as is can be immeasurable.
The cost to fix it?
The next post in this series will cover Bounce rates, Average time on site, and % of New Visits.