Looking at how analytics work, vendor solutions and rules for choosing an appropriate supplier.

The VOD industry has undergone a number of transformational phases since 4oD, the UK’s first broadcaster on-demand service, launched all the way back in 2006. After their initial rush to get an MVP onto as many platforms as quickly as possible, service providers have variously turned their attention to UI and UX upgrades, improving (or replacing) outdated legacy systems, scaling up with CDNs and integrating payment platforms and programmatic advertising.

In the last twelve months however it seems that we’ve entered a new stage of the product development cycle – the analytics phase. Now more than ever broadcasters are looking to understand, at a granular level, exactly how their customers are consuming their content and what improvements can be made in real-time to increase revenues and reduce churn.

Working with Rags Gupta, General Manager EMEA at Ooyala and Robert Farazin, Chief Product Officer and Founder of TVbeat, two companies that have built specialised video analytics products, I’ve produced the following introductory guide about this part of our industry.

Feel free to tweet me your thoughts @consultVodkr.

#1. What Is Analytics in relation to Online TV?

There are actually several flavours of analytics, which I’ve outlined below:

  • Behavioural Analytics – at its simplest, this allows you to track how much content has been watched by how many people. This is typically no longer enough for service providers who, in some cases, may have millions of users with varying tastes in TV shows and movies. More powerful, 360-degree tools are now available which allow you to track everything from the “device the content was viewed on, the location, operating system, syndication partner performance and drop-off rates” according to Rags Gupta of Ooyala;
  • Ad-Insight – a sub-section of the above, this type of data analysis gives publishers a comprehensive, real-time look at how ad-placement, duration and frequency correlate with video engagement. By monitoring KPIs in ad-supported content, companies are better able to optimise their ad-strategies to reduce drop-off and increase revenues;
  • Service Analytics – It’s one thing acquiring content and making a UI look great, but providing a consistent, uninterrupted stream to users is vital in order to promote engagement and reduce churn. Service analytics measure back-end processes, ensuring the quality of experience for each user is high, with minimal buffering or drop-off. Some solutions are able to use this data in real-time to make infrastructure improvements at scale – check out Conviva, for example.

#2. Why Is Analytics Important?

As more and more viewers watch their content online, it’s becoming increasingly important for service providers to understand exactly how, what, where and when their audience is consuming video. Having a better grasp of how your customers interact with your product enables better business decision-making.

I’ve outlined some potential use cases below. Note that this is not a comprehensive list – many departments can utilise analytics, from product development to marketing to finance. Drop me a line @consultVodkr to tell me about your experiences.

  • Content Acquisition – Having a robust analytics solution built into your player can help those responsible for buying programming make more informed decisions based on the tastes of their audience. Robert Farazin of TVBeat told me that:“Content spend is usually the highest expenditure within a broadcaster, representing around 40% of overall budgets. There is so much content nowadays that you cannot buy everything, so you really need to optimise for your users. If you don’t have analytics, then you don’t know what your subscribers are watching.”
  • Real-Time Platform Diagnostics – A study released by Conviva last year found that 75% of viewers will abandon a poor video experience within four minutes, often blaming the player itself. Integrating a service analytics solution that can report on streaming quality in real-time allows you to stay on top of technical issues even before they happen.
  • Upsell Opportunities – Increasing engagement is obviously a major KPI for all broadcasters and by leveraging analytics tools you are better able to group the customers you need to address at any given time. For example, if a new season of The Walking Dead has just landed on your service, you may like to segment viewers who have watched it in the past and send out an appropriate campaign to them.
  • Product Development – Understanding the features that your customers use (and, more important, the ones that they don’t use) allows you to refine your roadmap and potentially save resources on unneeded design and development.
  • Pay-TV Packaging – Here’s Farazin again:“For operators, using analytics can help you cluster the right channels into the right packages. Offering everything in one bundle can be cost-prohibitive for some users so segmentation is key but how can you do this if you don’t know what the viewers like to watch? You need to understand which channels have an affinity to each other.”

#3. Do You Really Need an Analytics Solution?

One major broadcaster I spoke to at TV Connect last month told me that they were still using Google Analytics to track user behaviour – for free! This then begs the question: do you really need a specialised analytics solution? Subject to the size of your user base and content catalogue, I’d argue that you do – whether you build it yourself or buy from a vendor. Sure, Google can give you the top-line stuff but expert online video analytics products come with some distinct advantages.

Gupta says that:

“TV analytics solutions that have been tried and trusted in the market for multiple years provide a strong foundation and have gone through the bugs and hurdles to make it a better and more efficient tool than it was the year before. And an analytics provider that has been on the market for multiple years will have had the luxury of seeing an enormous number of use cases, scenarios and solutions. Know that when you buy off-the-shelf, you’re also buying years of experience and expertise. Chances are the supplier will have a good knowledge of how the data can meet your business goals and objectives.”

#4. How do you Choose the Appropriate Vendor?

If you do decide to enlist an external vendor, Gupta and Farazin, both suggested some questions and considerations that you may like to include in your RFP:

  1. Can the analytics tool support and track performance across all of the devices that my audience is currently consuming content on today?
  2. How sophisticated is its ability to query multiple data sets?
  3. How do I draw actionable insights from the data points provided? Put another way, is the vendor willing and able to support my team, providing routine discussions about the numbers generated and how it can influence my business decision-making?
  4. Can I view data in real-time?
  5. Is the technology scalable across a long period of time? Can the vendor provide examples of comparable instances over a two-year period?
  6. Is the solution available as a standalone product or part of a larger video player package? If the former, can it fit into my existing technology stack? How long will the integration take and how involved will I have to be?

For Farazin, the best analytics system is able to ingest multiple data sets, in real-time and then export this information to as many third-party services as required.

“It’s becoming increasingly important for clients to integrate other parts of their architecture to the API of their analytics solution. This can include the CRM, campaign management, recommendations or ad-serving tools, amongst others. If you have a great analytics base that can collect and store data in real-time, and you can push the information to other components, you have the best system”.

#5. The Future

So what does the future look like for analytics? “Extremely bright,” says Gupta, as more and more broadcasters become interested in leveraging data to make their businesses more profitable. As analytics solutions continue to evolve and become more deeply integrated across the video supply chain, they will enable broadcasters to pinpoint exactly how, where and when their customers are consuming content, in turn leading to a more personalised viewing experience. With more information on an individual user’s watching habits will come the ability to serve up just two or three titles that are relevant to them at that precise moment in time.

From an infrastructure stand-point, big data will allow providers to optimise delivery in real-time and at scale. Streaming issues such as buffering, drop-off or server crashes (remember whenSling TV went down during its broadcast of the March Madness college basketball semi-final?) will be predicted before they happen and the appropriate preventative steps taken. The entire process will be automated, freeing up resources that can be diverted to other areas of the business – content creation/acquisition, marketing and user-facing product development.

Tweet me any thoughts you have @consultVodkr.

If you’d like help choosing the right vendor for your online video service, talk to us! We’re VOD geeks here at VOD Professional and we’ve written multiple reports and strategy documents for companies like Netflix, the BBC, Channel 4, BBC Worldwide, Liberty Global and UKTV.


Kauser Kanji has been working in online video for 19 years, formerly at Virgin Media, ITN and NBC Universal, and founded VOD Professional in 2011. He has since completed major OTT projects for, amongst others, A+E Networks, the BBC, BBC Studios, Channel 4, DR (Denmark), Liberty Global, Netflix, Sony Pictures, the Swiss Broadcasting Corporation and UKTV. He now writes industry analyses, hosts an online debate show, OTT Question Time, as well as its in-person sister event, OTT Question Time Live

Get OTT Briefings Every Week!

Sign up for my newsletter to stay up to date with stories, analysis, events and reports from VOD Pro.