Content is king. We all know that; to a digital marketer it’s like “slip slop slap”.
Having relevant, well written content on our sites is all part of our greater content marketing strategy these days, but the key is knowing which content is performing well (or not so well). There’s no point having it if you don’t know what it’s doing. Knowing how it’s performing helps us to better design our future content and optimise lower performing content.
There’s a number of different metrics you can use, like ‘time on page’, ‘bounce rates’, ‘views’ etc. but they won’t necessarily help you to understand the overall conversion performance. And you shouldn’t just add a conversion metric, such as ‘sign ups’, to a page report because it wont give you what you’re looking for.
One of the more advanced features related to content performance is success event participation – how well does the content participate in this particular success event. But, to understand it means getting a little more familiar with success event allocation.
Linear allocation.
By default, when you add a metric like ‘sign ups’ to a traffic report, SiteCatalyst will give each page a portion of the conversion credit based on path length. For example, if I view 10 pages in my user journey and sign up on the 10th page, each page will receive 0.1 conversion credit. Someone who views 5 pages before signing up will cause it to assign equal credit of 0.2 to each of the 5 pages.
Linear certainly has its place, but it can be very deceiving when looking at a report and trying to understand the values.
Enter Participation metrics.
In SiteCatalyst, there’s this thing called Participation.
You can enable Participation across any success event. Not many people use it. Participation assigns the value of the success metric to each of the pages viewed. For example, if I view 5 pages or 10 pages prior to signup, participation would allocate 1 signup to each page I visited. This gets even more important when you look at revenue participation (you don’t want to divide revenue by path length).
The percentage in the metric column shows you the percent of times this page participated in a conversion. For example, in the above image, Page 1 was seen in 33% of conversions, whereas Page 4 was only seen 6.7% of the time, prior to a conversion. So Page 1 is performing better than Page 4.
Content performance.
Getting back to understanding content performance, hopefully you can now see that we have a metric that can help us see things a bit more clearly. This helps show which pages are participating in converting prospects. High values will naturally be across all of you sign up pages, or shopping carts etc. Low values will be on non-participating pages.
I like to export the pages report with two metrics, Page Views and a Participation metric, such as Sign Ups, or Orders, or Applications, or whatever. Then I use Excel to produce a scattergraph to visualise the data. I eliminate the blatantly obvious ones, such as the sign up funnel pages, or the cart pages, and focus on the other areas. Typically I’d use a log-scale on the Page Views which helps to spread out the low performing content. The end result might look something like:
Three clusters begin to emerge. We’ve got three pages in the group that are performing really well, participating in lots of conversions. Then we’ve got a lot of pages that I’d call the opportunity cluster, they have a reasonable amount of page views in the given time-frame, but they’re not really participating in conversion. These would be good candidates for further analysis and optimisation.
The third cluster is where we have the low value content – content that doesn’t have many page views and doesn’t contribute to conversions. If your landing pages, or some that are considered top pages, are in this set then you’ve got a problem.
Some of these pages may surprise you. If that happens, it gives you an area to look at more closely. What is it about these pages that makes them perform less? Try doing some testing if they’re supposed to be high value pages.
You can also focus the analysis on different sections of your site instead of doing it across all pages. In Excel you can use the Filter capability to narrow down the list of pages to specific areas, sections etc. Your chart will re-draw, automatically highlighting opportunities and weaknesses.
So, participation metrics should definitely be used in your content performance analysis. And just by using metrics you take away a lot of the confusion.