Here at The Filter, we specialise in providing bespoke data science solutions for TV and OTT clients. A critical aspect of our product offering is our in-depth understanding of many different genres of content, and the different approaches that should be taken when personalising each of these genres. Although our bread and butter is the personalisation of movies and TV shows, for many of our larger clients we recommend sports titles as part of the overall customer experience. There are notable differences between users’ interactions with sports content and their interactions with other genres, meaning changes are required when designing sport-specific personalisation products. Whilst these changes might seem small, there is nothing worse than an average recommendation and so we have found that they have a disproportionately large effect on sports engagement and, ultimately, overall retention.
1. Only show sports to users who like sports
There is a very clear divide between people who will regularly engage with sports content, and people who never will. In a service that covers multiple genres, sports are often located on their own page to cater for this fact. However, we have recently started recommending sports content on the home page and it is working exceptionally well. Importantly, however, this approach should only be used in certain situations; only if the user watches sports and only at the right times, e.g. Saturday afternoons.
2. Live sports should also be personalised
There is a perception that personalisation is only valuable for on-demand platforms. However, within the world of sport you should not neglect the importance of live events. Indeed, most sports fans know the exact time their team are playing and when exactly they need to tune in. However, this customer journey can still be greatly improved; if a fan has to hunt for their live event that is starting in a few minutes, it adds friction to the UX and will result in frustration. If you are a sports destination with multiple events happening at once, covering multiple sports, and across multiple channels, then ensure the live match the viewer will be most interested in is at the forefront of the UI to improve the customer’s experience of your service.
For example, for one of the OTTs we work with, we found that within three views we can accurately predict which team an individual supports, and use this knowledge to make that team front and centre to all their touch points when live.
3. Sports is all about recency
Sports fans also want to know the latest stories that are circulating about their team or sport. Since there are many services which offer such content, you might not be able to own all of this engagement, but you need to make sure that you are high on the user’s go-to list of providers. When live events are not on, personalised swim lanes for that individual’s team or sport should contain all available related, and importantly recent, content. This includes highlight packages (especially when the user’s favourite team has won), post-match analysis, replays and interviews. You should also include wider commentary on rival teams, such as ‘goals of the week’ and other big news stories from that sport.
But the real key here is weighting the data science models up for recency. Recommending a highlight package from an event that occurred over a week ago will not drive significant engagement.
4. Sports fans often like multiple sports
Most sports fans have a favourite sport that they will devote time and effort to following. However, there are often secondary sports that they will engage with. We have found that here in the UK, rugby and cycling have a strong cross over. As a content provider, it is worth carrying out this correlation analysis to ensure you have content that caters for these overlaps. From a personalisation perspective, we make sure that some of these secondary sports are interwoven into any “For You” rail but with less focus on recency than for the user’s primary sport.
5. Sports metadata should be sport-specific
None of the above analysis will be possible to achieve unless you have the right metadata to support the data science workflows. The generic data structure that movies and TV shows are stored in will create unnecessary ambiguity if applied to sports content. For example, does the ‘genre’ metadata field contain the name of the sport, or is that found within ‘description’? Which metadata field is the outcome/final score of the event stored in? Where can I find the teams that were involved?
To cater for these differences and remove this ambiguity, we believe that storing sports metadata in a sport-specific hierarchy can greatly improve our data science workflows, and ultimately improve the quality of our personalisation products. Such a hierarchy might include, amongst others, the following attributes:
Sport / Date / Competition / Round/ Teams or Players / Winner / Score
6. Final thoughts
There are many other nuances when it comes to the personalisation of platforms that offer sports content. The labelling of the user interface needs careful consideration, and you must also consider the UX of logged-out or new customers, which can make up a large proportion of the overall user base.
With any of the above suggestions, the real key is to continually perform A/B tests to ensure that the changes being made are having the desired effect. By measuring the metric you are trying to improve (whether that be user engagement, retention, bounce rate or even acquisition), you can be certain that you are improving the performance of your platform. This data-driven approach is considerably more powerful and beneficial than a purely subjective evaluation conducted by a handful of individuals.
Here at The Filter we would love to help you on your journey of sports personalisation. If you would like to know more, please email firstname.lastname@example.org
About The Filter
The Filter offers personalisation of key customer touch points based on proven data science. We put the right content in front of a customer at the right time. More views means more loyalty, lower churn and more revenue.
The Filter prides itself on working closely with clients both during set-up and afterwards to make sure that the client’s customers are always getting the optimal recommendations. We are an agency not a SaaS provider. We have a genuinely deep TV understanding which means our many data science solutions work exceptionally well for your viewers. We work with you to tailor our AI to your customer experience, your audiences and your content. We also integrate effortlessly into your data ecosystem allowing you to benefit from the power of personalisation easily and quickly.