By Agostino Canepa, Director of Product Management, Telestream iQ
There’s no arguing with the fact that the OTT streaming platforms have presented service providers with unprecedented challenges in service quality management. With technology changing at a much faster pace than ever before, creating, operating, and maintaining a complex and sophisticated video network is very difficult to do well and very easy to do poorly.
Service providers’ main challenge is to maintain Quality of Experiences (QoE) not only related to the key metrics on how the content looks and sounds but also to the delivery of video content through the network as well as to the availability and performance of content in the network. There are many customer-impacting use cases of poor quality and interruptions like buffering or failed starts.
OTT service providers need more than ever a video quality monitoring and analytics platform that will allow them to find the root cause and understand what’s going on and where, so they can achieve the end-to-end performance essential to competing in a transformed premium video marketplace.
A Growing Pantheon of Suppliers in the Video Streaming Distribution Chain
No category of third-party vendors in OTT service delivery is more common to operations than CDNs. Adding to the challenge, service providers typically rely on multiple CDNs to provide backup support within a given service footprint and to extend service coverage across multiple geographic regions.
CDNs were designed to enable more efficient and, therefore, lower-cost use of backbone transport and internet peering capacity by providing regional caching locations for storing the most popular on-demand content. Now that live content has become a big component of the service mix, CDNs can’t always be counted on to maintain acceptable OTT streaming quality and performance.
When high volumes of live-streamed traffic are onboarded, CDN traffic flows can saturate allocated capacity, resulting in session start-up and in-session buffering delays at massive scales, not to mention crashes. According to the video intelligence firm Nice People at Work (NPAW), stream crashes during or after startup are five times more likely to occur with live versus on-demand content.[i] The difference, of course, is due to the traffic spikes caused by concurrent viewing.
This is especially damaging when consumers are tuned into live-streamed sports, which are occurring with ever greater frequency worldwide. ResearchAndMarkets predicts the live-streamed sports segment will grow from $18.1 billion in 2020 to $87.3 billion in 2028, which equates to a compound annual growth rate (CAGR) of 21.26%.[ii] In the U.S., 61% of people who use OTT services transmitting linear programming watch live-streamed sports, according to Parks Associates.[iii]
People want to see a live event at the same time it’s being played out on legacy TV outlets, which prevents the notorious spoiler effect that occurs when broadcast viewers react to something happening on the field before online viewers have seen it. This is especially important when people are engaging in group viewing through social media or when the viewer is on a second screen in proximity to a TV set showing the same event. Out-of-sync delivery with broadcast is also damaging to new micro-betting applications that allow people to gamble on unfolding events.
About two-thirds of consumers worldwide are more likely to stream a live event if they know it will line up with broadcast latency, according to a survey conducted by Limelight Networks, the CDN operator that was recently renamed Edgio.[iv] CDNs are doing better at this but are still a long way from consistent performance, especially when the streaming volume gets into the millions.
Service providers have no control over what goes on inside a CDN. But they certainly can employ intelligent video quality monitoring and analysis on CDN output to detect and hold operators accountable for subpar performance.
The Tight Coupling Between Live & Time-Shifted Content Use Cases
Another new element to the new OTT environment impacting OTT streaming quality management has to do with service providers’ opportunities to facilitate time shifting of live-streamed content. With the aid of advanced asset management systems and tiered storage, they can add appealing trick-play, catch-up, and network DVR (nDVR) features to live-streamed services on a shared-resource basis with file-based content.
The technology allows operators to automatically coordinate how the flash memory capacity in CDNs and their own edge clusters are used in conjunction with short-term solid-state Network Attached Storage (NAS) and longer-term core file storage to maximize speed and efficiency with no perturbations in the live-stream flow. As ABR segments exit packagers they can be temporarily cached locally and fed to NAS storage for compilation into complete files for catchup viewing over a set period, after which they can be shifted to longer-term storage in response to users’ nDVR commands.
This avoids creating separate workflows to turn every live stream into a stored file for on-demand access. And it reduces use of long-term storage requirements for content that’s not meant to be part of a VOD service offering. This added complexity in the time-shifted workflow implementation makes finding issues more difficult and requires advanced monitoring tools to detect and isolate those issues.
The Movement of Transcoding & Packaging to the Network Edge
Transcoding and packaging related to per-stream personalization of some content elements is moving closer to end users. This adds new complexities to OTT streaming quality management, given the complexities of synchronizing incoming graphic and video overlays as well as ads with core content across the gamut of streamed bitrate profiles.
Most significantly, precise performance with Dynamic Ad Insertion (DAI) has become crucial to service providers’ bottom lines now that addressable advertising is driving the higher cost-per-1,000 viewer impressions (CPM) rates that underlie many OTT monetization strategies. Where traditional linear TV 30-second spots typically garner $10-$25 CPM rates, addressable ad CPMs routinely run in the $40-$50 range.[v]
Implementation of these capabilities is the natural outgrowth of the massive shift of TV ad spending into the OTT sector. Globally, Digital TV Research projects the total annual ad revenue flowing into Advertising Video-On-Demand (AVOD) services and into a growing class of Free Ad-Supported Streaming TV (FAST), services will go from $27 billion in 2020 to $66 billion in 2026.
With unicast OTT streams, advertising can be targeted to users based on demographics, the viewing context, personal preferences, or other parameters. This is proving to be a major draw to advertisers, who, facing diminishing returns from traditional linear TV advertising, are feeling greater urgency than ever to get maximum bang for their advertising dollar as live-streamed viewing takes off. According to a survey conducted by the Video Advertising Bureau, 71% of responding buyers said they were willing to pay higher CPMs with targeted placements.[vi]
In this environment, the ability to accurately track ad performance with respect to maintaining OTT streaming quality parity with core content and accurate placement is critical to success. Beyond ensuring ads are properly transcoded for these formats, service providers must be able to assess whether mechanisms essential to accurate placement are producing the intended results. This should allow them to coordinate with their own and third-party ad management workflows to automatically execute adjustments in ad breaks and insertions of missing SCTE-35 ad markers.
The Right Approach to OTT Streaming Quality Management
Any approach to OTT streaming quality management requires strategic positioning of advanced video analysis probes in the video delivery chain. When used in conjunction with an overarching intelligence layer, the required visibility in the OTT streaming quality can be illuminated. The goal is to turn metrics into actionable events to address issues that can occur at any point from contribution feed, encoding, packaging, and CDNs down to the subscriber’s device.
Telestream iQ’s portfolio of video analysis probes has long dominated the OTT streaming service. With analytics and placements tailored to the complexities of ABR streaming over the open internet, the combination of Telestream iQ analysis probes along with ARGUS provides enhanced visibility and insight into live streaming and on-demand OTT services. ARGUS is Telestream iQ’s powerful centralized video quality monitoring solution that enables automated surveillance of each video transition with data aggregation from monitoring probes in the video delivery chain. Find out more key benefits and features of ARGUS.
We hope you will allow us to help you with your video monitoring needs. Interested in learning more? Contact us to discuss our video quality monitoring and analytics services.
[i] NPAW, Global Streaming and Broadcasting Report, February 2020
[ii] ResearchAndMarkets, Outlook on the Sports Online Live Video Streaming Global Market, October 2021
[iii] Parks Associates, 61% of Livestreaming Viewers Watch Sports, May 2022
[iv] Limelight Networks, The State of Online Video 2020, October 2020
[v] Harmonic, Guide to Launching a FAST Channel, February 2021
[vi] Ad Week, Study Projects Addressable TV Advertising Will Double by 2018, November 2016