Skip to main content

Hello there, 

I would like to know how many of my readers read my articles completely in other way how many percent of from all my articles are read from A to Z ?

To do that, i have an event A that tell me that my user are viewing the article and then after some scroll down and some time spend on the page an other event (B) told me that they have finished the article. 

So my guessing was to do a funnel chart from event A to event B, however how to to be sure that the event B is from the same article than event A ? It’s also possible that he starts reading one article, did’nt like it, went on anohter one and this time finished it and so trigger event B.

Can we solve this by using “holding a constant” by the url or the name of the article, to be sure he stays on the same article ? And should I count by Unique or by total ? 

Can you help me with this by answering me if i’m doing it correctly or not ? 

 

Thanks in advance ! :)

 

Here is the screen of what I think is right : 

 

Holding the URL or page title constant like that would work yeah. Unless you think you have readers coming back and re-reading the same article multiple times (which is a very small percentage of users in my experience at major media publications with the rare exception of some evergreen content), then I'd stick with uniques. Would be interesting to see the different between those though, for example if you’re writing instructional guides you might have a lot of repeat visits and that’s an interesting insight. 

 

Depending on article length though, the 15 minute window is a relatively short time, and I’d also say, depending on your products feature set, using a scroll depth event is a good proxy but not a 100% guarantee it was an article read. While a good amount of people did, there’s plenty of scrolling activity that can happen on webpages (jumping to comments for contentious activity, wanting to see other links at the bottom of the page, etc) that can account for it. 


Reply