It's been a long summer. So much time on the beach, so little time to blog. In any case, I've recently moved more deeply into the realm of Web Analytics. Think of it like Big Brother for the Internet. Someone is watching what you do, and in the case of a big communications company here in Calgary, that someone is me.

Don't worry, it's not as 1984 as it sounds. The website I manage now gets about a billion page views a year, so the I'm not too concerned about individual users. (All kidding aside, the only thing Web Analytics can see about you is your IP address, which can't narrow your location down much below the city level).

If you're looking for some basic information on how to get started with Web Analytics, check out my introductory article here. There's also a good overview look with my Executives Guide to Web Analytics. As I delve deeper into this mysterious world though, I'll share some insights here on my blog.

The one thing about analytics is that if done well, it can help executives make decisions that can have significant impact on the bottom line. My elevator pitch these days is "I help companies improve how their website works, making it easier for consumers to buy from them, thereby making the company way more money".

Tall order? Not as tall as you might think. Most bigger websites aren't tracked all that efficiently. If they are tracked at all, it's all about page views. Yes Mr. President, our website views are up 10%. An astute president will say.. so what, what does that mean to my business?

The reality is that Key Performance Indicators (KPI's) need to be chosen in such a way that REAL decisions can be made from them. What is more useful? How many page views your website got? Or what percentage of your email campaign recipients bought something versus the last campaign? How about if there is a glitch in your order form that is stopping people from buying? Analytics can tell you all these things, and much more but deciding what is useful, actionable information and what isn't will be your key to success.

So how do you choose a good KPI? Work backwards. What are your companies overall revenue and growth projections? How is the website contributing to them? Once you have those questions answered, you can delve into the specific parts of your website and use analytics to determine your best bet for KPI's. There are no perfect KPI's, it's simply a matter of what relates to your company.

Some of the more common advanced KPI's include share of wallet, homepage bounce rate, average depth of visit (how many pages did a visitor look at), revenue per visitor and the list goes on.


I'll post some more experiences soon as I go through the process of identifying qualitative KPI's for the billion page view per year site that I'm working with. Stay tuned for more!

Labels: google analytics, web analytics

A couple of ongoing topics I've ran into lately...

Recently Google Analytics launched their new and upgraded interface. After a few minutes of playing around with it, I've found everything I've used in the past. A great deal of attention in group forums though has been focused on the disappearance of visitor statics by the hour. It now appears that Google either forgot or ignored it but thanks to user outcry, they are now working on putting it back in.

The hourly traffic analysis is interesting on a few levels, but also raises many questions. Should you target
AdWords advertising during your low hours to boost traffic? Or is there bigger bang for your buck to advertise during your busiest hours to attract the largest possible interested audience?

If you're an e-commerce website, particularly a regionally targeted e-commerce website, this could be really good information. If you're a flower shop selling flowers only to people in a specific city, you will probably get hits at specific times of day, and then it may be good advice to use Google AdWords to boost this trend (attract visitors when they are looking).

Here's hoping the hourly report comes back soon!

Another interesting topic I've seen around lately is the length of URL title for search engine optimization purposes. Now there are two arguments here. One is that a description URL, like...www.calgary.com/downtown_calgary_flower_delivery.html is great from an SEO point of view. The counter argument is that no one will
remember that or type it in themselves. Compared to www.calgary.com/flowers which someone might actually recall from other advertising (radio, tv, or whatever) and type in themselves.

My theory is do both. I've used this strategy successfully for search engine optimization purposes AND advertising simplicity. All you need to do is create both pages, and then redirect the simpler, easier to remember address to the more complex one. Don't put any text on it or duplicate the page as you might be penalized for having duplicate content. A simple meta-refresh tag will do the trick. Also make sure
there is a text link on your main page or site map so Google can find and index the page easily.

Voila, a great search engine optimization URL plus a great advertising campaign. Google sees and indexes only one page with appropriate content. For more on SEO, check out my SEO Articles.

Good news for me and my fellow Calgary walkers and bikers, the weather is finally warming up!

Labels: google adwords, google analytics, search engine optimization, SEO