Data Analytics – Carving Out the Next Level of Engagement for the OTT Sector
Data analytics is and will continue to be a big part of content development, consumption driving activities as well as engagement initiatives for the VOD business.
Media & Entertainment industry has been using data analytics for a long time. Programmers have used the ratings (audience measurement) data to get a pulse of the viewers and then create content to maximize viewership. And media planners use the same data, to maximize the reach and impact of advertising. They also use several other tools for cross-media analysis and planning. It is interesting to note that most, if not all, tools used by our industry, have two things in common – 1) they are sample based and 2) they don’t give real time results.
And then it all changed when Digital Video services arrived. Over-the-top (OTT) digital video on demand services allowed the two connection and census-like measurement, which not only meant that you can know what ‘everyone’ was watching , but do this measurement with as much detail and accuracy as you wanted to – and all of this in real time. Viewers, besides video consumption, are constantly engaging in real time feedback which is actively decoded, analyzed and acted upon to be measured and retested again. While data analytics is widely used across most internet based services, the digital VOD space has been among the frontrunners in this. Not only are the content offerings being customized, segmented and packaged basis user’s preference and consumption patterns, now content creation itself is based on huge amounts of data and insights drawn from it.
In order to understand how some of the successful VOD players are using analytics, we need to understand the various stages involved in the process. In fact these stages form the maturity curve of the analytics process and business intelligence tools in the company. Of course the pre-step in all of this is to get a good database system, where data points are allocated based on which data is collected. Here, one needs to ensure that the data collected is agile, scalable and ever evolving.
The first stage is Descriptive, where you capture what happened. This is always in hindsight. The second stage, one needs to look at the ‘why’ of what the data is saying, and this is called Diagnostic. Very often people stop at this stage and say they are using analytics well. But the real magic starts from stage 3. Once the data is analyzed, the third stage is Predictive - where one attempts to predict and understand what is going to happen through the trends analyzed. The 4th and final stage is often called Prescriptive – where Machine Learning and Artificial Intelligence drive actions based on predictions and to influence ‘how to make something happen’. Recommendation systems are a good examples of this stage.
While these steps are important in understanding how Data Analytics works, they need not be followed in a step by step manner. They can be executed simultaneously, thereby greatly reducing the time involved. To give an example from our own business, VOOT has content which can be broadly divided into the following: TV Network content, VOOT Originals and VOOT Kids. Using analytics we have been able to understand the audiences and their preferences across each of these segments on our platform. VOOT’s big data has given us deep insights from which we have created this new type of programming which we call content around content (CAC) around our Non-Fiction shows and this has been a big driver for consumption on our service. Basis data, we have been able to sharpen this further across different kinds of CAC - character stories, spin-off shows and un-cut content.
There are numerous other examples - including those for VOOT Kids’ consumption on a parent’s phone, cross language and cross genre content consumption, timing for publishing new episodes and impact, timing & tonality of notifications - where we have derived great learnings from analytics.
To get the full benefit of Analytics, data must be analyzed at the granular level. This includes understanding data that stands out and comprehending the reason behind it. For example, if a user comes to VOOT and accesses content older than few months, we need to understand this usage pattern. Why is this content being accessed? Are these users always watching old shows? At what point did they want to go back to some older shows? Are they searching those episodes through in app search or scrolling via published verticals? Why did they play a particular old episode, is there a storyline link between the current episode and the old one or they liked the episode thumbnail and clicked? Only after knowing the detailed behavior of the users will we be able to use this for any meaningful next steps. We in fact, have decided to re-commission / extend an old show (that stopped airing on TV) basis understanding of such user behavior on VOOT.
Data analytics is and will continue to be a big part of content development, consumption driving activities as well as engagement initiatives for the VOD business. Today, knowledge garnered from this data is reaching a new high every day, to the point where Data is rightfully considered the new oil and Analytics, the oil drill.
Disclaimer: The views expressed in the article above are those of the authors' and do not necessarily represent or reflect the views of this publishing house
Around The World