Com Hem has seen first-hand how much subscribers value Google Assistant on its recently launched Android TV Operator Tier based platform. Regarding Google Assistant as a key product differentiator, Com Hem has invested in broadening the capabilities of Google Assistant, and other operators will follow its lead. This article looks at what is possible for all Pay TV providers using Assistant, including when they move beyond an on-device implementation to a cloud-to-cloud integration. With this latter scenario, someone could say ‘Record this’ and the platform will know what you are watching automatically and set the PVR into action. There are front-end strategies that can also surface operator content ahead of third-party search results.
This week’s ‘Inside Video Tech’ feature focuses on Verizon Media, which provides media and entertainment companies with end-to-end solutions for top quality streaming, encompassing a streaming platform and CDN. The company chooses the cloud as the tech development that has had the biggest recent impact on the video market. Verizon Media champions a multi-CDN approach to protect content against slowdowns and downtime with any one delivery provider, and believes a key industry challenge now is to ensure hyper-personalised experiences, particularly for advertising.
Advertisers are rushing to get on board digital video, and demand appears to be outstripping supply. With a shortage of inventory, how can we match demand sources with their target audiences as efficiently as possible? Header bidding is a major innovation that opens up more demand source, can result in higher CPMs for sellers and help advertisers reach the right audiences through video. Open source technologies like Prebid.org mean no one should be left behind. But if header bidding is going to work smoothly with video, reducing latency must be front of mind for the industry.
Looking beyond scale and quality of inventory, some of the biggest challenges in advanced TV advertising are the absence of unified cross-channel measurement, differing views on what class of ad inventory connected TV belongs to, connected TV inventory that is not clearly identifiable when sold and, in Europe, relatively immature connected TV advertising sales strategies. Change priorities include media sellers powering cross-device measurement solutions and an industry-wide move towards a single, consistent metric that establishes media/marketing efficiency.
Using a CDN-enabled SaaS solution, as opposed to satellite, programmers can meet the demand for linear content on connected screens. A key benefit of this cloud solution is the capability to insert the accurate metadata needed not only for rights management and blackouts, but also for monetisation and dynamic ad insertion, without requiring dedicated on-premises equipment and with little upfront cost for distribution of local channels to virtual Pay TV operators. Supporting a wide range of use cases, with increased speed and efficiency, SaaS solutions are setting a new benchmark for OTT delivery of linear channels.
TV users who spend an above average amount of time watching on weekdays tend to be more likely to churn compared to those who spend more than the average time watching during the weekend. That is one insight from a recent study that confirms that the best predictors of churn are viewership and engagement. TV personalisation also helps keep subscribers. And brand new subscribers who engage with personalisation tools are up to 140% more likely to stay as customers.
Television has transformed sports such as football, with millions of fans tuning in week in, week out to see their team. It is clear...
Growing OTT consumption emphasises the importance of Wi-Fi Quality of Experience. Operators need to proactively prevent problems and avoid customer calls and truck rolls. For this, they need visibility into the end-to-end service delivery, data usage and device environment, and the use of prescriptive and automated remediation techniques, working in real-time. Combining TR-069 data with usage behaviour and machine learning is a large part of the solution, and early implementations of this approach show tangible operational and business benefits.
OTT means content providers can reach world markets, but they need to ensure compliance with local regulations. In order to achieve this with large volumes of video, you have to look beyond manual classification. Next-generation QC checking technology, harnessing AI, ML and computer vision algorithms, means you can automatically label and categorise content. By identifying objects, actions, scenes and spoken or visual words in streaming content, you can flag the presence of nudity, violence, alcohol, smoking, vaping and different levels of sexual activity, among other things.
The issue of credential sharing has become too big and too expensive for media companies to ignore. Tracking technology is one solution, but digitally-savvy consumers could cover their tracks with a VPN that alters their IP address and so masks their true location. And because legitimate users may use a VPN due to online privacy concerns, broadcasters cannot simply block all VPN users. Instead, you have to take a more granular look at viewing patterns and user behaviour to flag suspicious use.