Quantifying Content Velocity in Adobe Analytics

If publishing content is important to your brand, there may be times when you want to quantify how fast users are viewing your content and how long it takes for excitement to wane. This is especially important for news and other media sites that have content as their main product. In my world, I write a lot of blog posts, so I also am curious about which posts people view and how soon they are viewed. In this post, I will share some techniques for measuring this in Adobe Analytics.

Implementation Setup

The first step to tracking content velocity is to assign a launch date to each piece of content, which is normally the publish date. Using my blog as an example, I have created a SAINT Classification of the Blog Post Title eVar and classified each post with the publish date:

Here is what the SAINT File looks like when completed:

The next setup step is to set a date eVar on every website visit. This is as simple as capturing today’s date in an eVar on every hit, which I blogged about back in 2011. Having the current date will allow you to compare the date the post was viewed with the date it was published. Here is an example on my site:

Reporting in Analysis Workspace

Once the setup is complete, you can move onto reporting. First, I’ll show how to report on the data in Analysis Workspace. In Workspace, you can create a panel and add the content item you care about (blog post in my example) and then break it down by the launch date and the view date. I recommend setting the date range to begin with the publish date:

In this example, you can see that the blog post launched on 8/7/18 and that 36% of total blog post views since then occurred on the launch date. You can also see how many views took place on each date thereafter. As you would expect, most of the views took place around the launch date and then slowed down in subsequent days. If you want to see how this compares to another piece of content, you can create a new panel and view the same report for another post (making sure to adjust the date range in the new panel to start with the new post’s launch date):

By viewing two posts side by side, I can start to see how usage varies. The unfortunate part, is that it is difficult to see which date is “Launch Date,” Launch Date +1,” Launch Date +2, ” etc… Therefore, Analysis Workspace, in this situation, is good for seeing some ad-hoc data (no pun intended!), but using Adobe ReportBuilder might actually prove to be a more scalable solution.

Reporting in Adobe ReportBuilder

When you want to do some more advanced formulas, sometimes Adobe ReportBuilder is the best way to go. In this case, I want to create a data block that pulls in all of my blog posts and the date each post was published like this:

Once I have a list of the content I care about (blog posts in this example), I want to pull in how many views of the content occurred each date after the publish date. To do this, I have created a set of reporting parameters like this:

The items in green are manually entered by setting them equal to the blog post name and publish date I am interested in from the preceding data block. In this case, I am setting the Start Date equal to the sixth cell in the second column and the Blog Post equal to the cell to the left of that. Once I have done that I create a data block that looks like this:

This will produce the following table of data:

Now I have a daily report of content views beginning with the publish date. Next, I created a table that references this table that captures the launch date and the subsequent seven days (you can use more days if you want). This is done by referencing the first eight rows in the preceding table and then creating a sum of all other data to create a table that looks like this:

In this table, I have created a dynamic seven-day distribution and then lumped everything else into the last row. Then I have calculated the percentage and added an incremental percentage formula as well. These extra columns allow me to see the following graphs on content velocity:

The cool part about this process, is that it only takes 30 seconds to produce the same reports/graphs for any other piece of content (blog post in my example). All you have to do is alter the items in green and then refresh the data block. Here is the same reporting for a different blog post:

You can see that this post had much more activity early on, whereas the other post started slow and increased later. You could even duplicate each tab in your Excel worksheet so you have one tab for each key content item and then refresh the entire workbook to update the stats for all content at once.

Check out Part 2 of this post here: https://analyticsdemystified.com/featured/quantifying-content-velocity-in-adobe-analytics-part-2/