Meaningful analytic benchmarks and insights for 100+ websites?

[JCCA]

Hey all!

I work for an association that has over 100+ member businesses. These businesses are dispersed through out North America and each have their own website, totally independent of each other in look, feel, functionality, etc.. While most of these businesses offer the same core businesses such as a gym, summer camp and preschool, none of the website structures, calls to actions, etc. are the same. I have a feeling that there could be some interesting and meaningful analytics that I could pull together if I had access to all their Google Analytics accounts, but I'm not quite sure what they would be. Some ideas I have -

1 - Percentage of traffic from social media - will give insight as to which businesses in the association are using social successfully.

2 - New vs. Returning traffic - a high percentage of returning traffic might be indicative of more useful websites.

3 - Mobile usage - How mobile savvy are the customers of different businesses in the association.

Any other analytics (and quite frankly, more meaningful/impactful than the ones I posted above) that you guys can think of?

Thanks for your help!

Chris

[DJ123]

What would the goal be if you had all the data you wanted?

[JCCA]

DJ123,

I guess that is the question I am asking. If given the opportunity to have all that data, what kind of meaningful insights could I get from it? Ultimately I want to be able to report back to the members of my association and tell them how to make their websites better. Sorry for the vague response, but I ultimately want to know is it worth the time to collect all the data and if I do, what are some meaningful things I could do with it.

Thanks!

Chris

[DJ123]

In addition to what you said, I would also get:

1. unique visitors by time period x

2. possibly geographic area of visitors

3. bounce rate

4. time of day that visitors visit

I would then take all those metrics and then compare them among similar businesses, then similar businesses in similar sized markets and see what raw differences you can see.  One of the problems which you would need to overcome is the difference in markets (both in size, preferences, etc).  There are local differences which would need to be accounted for if you were to be able to have actionable items.  Another challenge would be that every site is probably targeting different keywords to different cities and targeted demographics.  That is another thing that you would need to normalize for.  Also, each site is probably running on different platforms, etc so you would need to try to think about that aspect of the data because it would introduce noise into the data.

If you had the data, you may want to look into something called Multiple discriminate analysis: http://en.wikipedia.org/wiki/Multiple_discriminant_analysis due to how many variables you are going to have.  If you cannot do it yourself, there are statisticians out there who could help you set up the data in the right way.  They are not cheap though...

[JCCA]

Great insights and suggestions - thanks for replying!

I'm hung up on the different size markets as well. Our association has sorted the 100+ businesses into different size brackets, but those are based on revenue as opposed to size of city (although there is definitely a correlation.)

Analytics like bounce rate though would be a benchmark regardless of the demo/size of the business' location.