Marketing Analytics: What annotation data can tell you about video subscribers

In the past year, plenty of charts, projections and infographics all show video content heading in one general direction – up and to the right.

This makes sense given the use of video content is also growing as more marketers adopt video into their marketing mix, lest their message is left behind by the projected 77% of all Internet users who will be viewing video content online by 2016.

With the adoption of any new strategy comes the part where the devil is in the details – how do you measure video content and what can the data tell you?

This was a challenge Luke Thorpe, Audio and Visual Manager, MECLABS, was facing in early 2012 when MarketingExperiments Web clinics transitioned to a video format. Shortly after the transition, Luke started implementing best practices into his video editing and uploading to YouTube.

 

One of those practices is adding annotations to the end of videos. Annotations appear when your video ends and it allows users to view other videos you have on YouTube with a single click. Annotations also allow you a certain amount of creativity that Luke explained.

“I have a friend over at GoPro cameras and they were using annotations that included video thumbnails and music. When I saw their video, I wanted to experiment with something similar on our site. I thought it was cool and so my thinking was in line with that whole expression ‘imitation is the highest form of flattery.’”

 

Creativity is one thing, finding a way to measure it is another

Related:  Data Analysis 101: How a nonprofit used data to secure a critical business decision and help find 125 missing children

After a few months of adding annotations, Luke started to notice a spike in the number of subscribers to the MarketingExperiments YouTube channel. But, he wasn’t sure why the gain rate was increasing because YouTube’s dashboard analytics currently don’t allow you to directly measure the impact on subscriptions through adding annotations.

Also, given at the time he didn’t have any tracking in place, our options were limited in what he could learn, so he asked me to take a look at the data.

Here’s what we found:

 

First, we compiled the daily views and subscriptions to see if any correlation existed.

We found that the likelihood of a correlation existing between daily views and subscribers at around 76%, but the caveat here is the ever cautionary tale of statistics. Correlation is not always causation.

 

Explore every option  

Next, we took subscriber gain stats for the month prior without the annotations and compared that data with subscribes after Luke started annotating videos and found that there was a correlation between views and subscriptions.

 

What can we learn from this?

There are a few things we can glean from the results:

  • There’s no surprise here that a correlation exists between an increase in views and subscriptions. It makes sense after all that they grow together as more views will likely produce more subscribes.
  • We need to discover which one of these two variables is the responsible driver for the upward trend.

The only way to truly know this would be to increase our tracking and analysis efforts to really capture the true cause for the change in behavior.

Related:  Data Analysis 101: How a nonprofit used data to secure a critical business decision and help find 125 missing children

It’s interesting to consider that small changes can have a big impact, but the key rests in identifying the elements of change and understanding how they work together so you can leverage that knowledge to better serve your customers.

When I asked Luke what he thought about the findings in the data, he explained that annotations help to “close the loop” of sorts in your video content.

“Adding annotations keeps people in your channel longer and engages more of your content,” Luke said. “The views via the annotations might not be very high, but without them, the additional views would be zero.”

 

Related Resources:

Analytics and Testing: Understanding statistics symbols and terminology for better data analysis

Web Analytics: What browser use can tell you about your customers

Marketing Analytics: 4 tips for productive conversations with your data analyst

Optimizing for Multiple Personas — MarketingExperiments Web clinic replay

You might also like

Leave A Reply

Your email address will not be published.