A vast majority of discourse in the web analytics world is about orders and conversions and revenue. There is not enough of it about non-ecommerce websites, metrics and KPI's.
I hope to shift that imbalance a little bit with this post. :) With a tweak here and a tweak there along with a understanding of your outcomes and you can measure any type of website effectively.
I am not going to talk about lead generation websites here since metrics similar to ecommerce (conversion rates etc) could still apply there. The types of sites I am thinking of are pure content sites where people come, consume content and they leave. And maybe they come back again, and again, and maybe again. :)
Think of a technical support website (though in that specific case read this: Measuring Success for a Support Website). Or a news website. Or a blog. Or a site full of white papers. Or think of a "brand" site or other such things. Bunch of content and no real online outcome.
My recommendation: Measure the four metrics that are under the "Visitor Loyalty" button in Google Analytics (or in your favorite web analytics application). Loyalty, Recency, Length of Visit, Depth of Visit.
The goal is to use web analytics data to interpret success of a visit to your website.
I am embarrassed to admit that until a couple months ago I did not even realize how well these were laid out in GA. I blame it on the V1 interface!! :)
There is one singular reason I loved 'em: they showed distribution and not simply averages for each of the metric!
The metrics mentioned above have been recommended in the past for use for non-ecommerce sites. But it is illustrating of the distribution of the metrics in this case is the key for gleaning insights. A escape from "averages hell" if you will, a world where insights were hidden.
Visitor Loyalty: During the reporting time period how often do "people" ("visitors") visit my website?
The number you are used to seeing is "average visits per visitor". That is usually one point something. It hides the truth.
In this case like a the curtain is removed from the "average" picture and you can see the front and back loading of the trend. It helps you understand what the median is, where are your outliers and if you were aiming for a particular number then what are the specific number (or percent) of people in that bucket.
For example you update your website ten times each month. If you have 100% loyal visitor base then they should be visiting your website ten times each month. Are they? What's your number? Is it going up over time?
Action: 1) Identify a goal for your non-ecommerce website for the # of visits you expect from the traffic to your website in a given time period (say week, month etc). 2) Measure reality using above report. 3) Compare your performance over time to ensure you are making progress, or potentially not as in my case…
Recency: How long has it been since a visitor last visited your website?
Sounds confusing? Don't worry it is cool (it even has a psychedelic border! :)……
Simple example: www.cnn.com probably wants the visitor recency metric to be 1 day ago, which means every visitor comes to the website every single day to consume all the wonderful content that is on the website.
As would be the case for a jobs site. Or craigslist. Or any website that wants lots lots of repeat visits. Using this simple report you can now see how you are doing when it comes to the distribution of visitors in terms of their propensity to visit your site.
Action: What kind of site are you? If 67% of your audience (that's 0 days above) is consistently new over months then should your site content / design / merchandizing be different? Should you "sell" harder the value of repeat visits to your audience? Is that number going up or down?
All questions that now you can answer and as you create incentives on your site for people to visit more frequently you can actually measure success!!!
Length of Visit: During the reporting period what is the quality of visit as represented by length of a visitor session in seconds.
Average time of site is perhaps the most common web analytics metric on earth. Some people have recently taken this on as the new word of god.
But it has always been frustrating to me how hard it is to get away from the average and measure the distribution of the visits to check if the average time on site is 50 seconds because one person visited for one second and the other person for 100 seconds. The average hides so much. Here's a better alternative……
Ain't that better? I think so. So many things jump out at me, but notice that either I lose 'em right away or if some how I can suck them in for one minute then they tend to stay for a long time. Hurray! I have a better idea of how to interact with my visitors.
Action: 1) Identify what the distribution is for your website for length of visits. 2) Think of creative ways to engage traffic – what can I do to keep you for sixty seconds because after that you are mine! 3) Should I start charging more for ads on my site – if I have 'em – after 60 seconds? 4) If you are a support website then should you be embarrassed if 20% of your audience was on the site for more than ten minutes!
Create your own goals, measure success for percent of visits that are long and percent that are short.
Depth of Visit: During a given time period what is the distribution of number of pages in each visit to the website.
This is the brother metric to the Length of Visit metric above. Though given the recent press from random bloggers (self included) a brother that everyone is in a rush to disown. It is important to remember most of the web is still pages.
You are used to seeing average page views per visitors, above is something that is a lot more helpful. I was also able to get this exact metric from my indextools implementation…..
Action: There has been so much said about this already so I'll spare your the pain. You can easily imagine how wonderful and fantastic this data is as you go about analyzing experience of your customers (and so much more powerful, a million times more, than average page views per visitor!).
Recommendations for all of the above metrics:
Getting the data above is just the first step. Once you have the above distributions for your key metrics:
- Socialize them to your key stake holders and decision makers to make the realize what is really happening on your website. Get beyond living in the average world (where Bill Gates and I make a average of 45 million US dollars per minute!!).
- Absolutely positively work with your leadership to create goals and then measure against goals over time, especially as you make changes to your website.
- Segment the data! For Visitor Loyalty or Length of Visit what are the most important acquisition sources? What are the keywords that drive valuable segments of traffic to the website? As you look at longer time periods what pieces of content do people with longer visits consume? And so on and so forth. Segmentation is key to insights that will drive action.
Convinced these are wonderful metrics to measure success of your non-ecommerce website? I'll be the first to admit that all four metrics are "simple" yet presented differently they can be
You have a content site, you have visitors, now you can set clear goals and measure success of your website. Every single web analytics tool should provide one click access to a clear visualization of this data (check you web analytics tool, it probably has this already).
Please share your feedback via comments? Critique 'em, add your favorites to them, remind me what I have missed in this story.