Quality of experience shapes brand perception. If a video is slow to start, buffers during playback, or suffers a noticeable drop in resolution, it can quickly erode trust, result in negative reviews, and drive subscribers away. Bitmovin’s 8th Annual Video Developer Report found that nearly 20% of video providers struggle to deliver a consistent quality of experience, while more than 24% find it difficult to identify the root cause of playback issues. Without the right visibility into performance, even a minor disruption can lead to a lasting impact on viewer satisfaction and retention.

The same report shows that it takes most video developers between 12 and 24 hours to identify the root cause of a playback issue. Traditional approaches rely on post-issue analysis and user-reported errors, which are too slow to protect the viewer experience at scale. The delay between a playback issue occurring and resolution leaves providers exposed, especially during high-profile events or peak viewing periods.

Understanding the Quality Challenge
Viewers expect seamless playback across any device, whether it’s a phone, tablet, smart TV, or set-top box. To meet this expectation, video services must support an expanding mix of platforms and hardware configurations, all while accounting for different operating systems, browsers, and firmware versions, which creates a complex testing environment. Manual testing is slow and expensive, while automated testing requires a lot of effort to stay current with evolving streaming protocols and codecs.

This is where viewer-side metrics become essential. Backend systems may report that everything is operating normally, but that doesn’t help when a user is stuck buffering or receives a lower resolution than expected for their device. Metrics such as startup time, buffering frequency, playback resolution relative to screen size, and error types, offer far more meaningful insight into real viewing conditions.

These data points allow teams to move beyond generalized performance assumptions to understand what users are actually experiencing, across all devices and environments.

The Case for Real-Time Observability
Observability is frequently used in the IT industry to identify, diagnose, and resolve issues in real time, yet in video streaming, it remains underutilized. Many providers still rely on static dashboards or user complaints to surface problems, creating a lag between failure and resolution.

As the video ecosystem becomes increasingly fragmented, this reactive approach falls short. Ensuring a seamless viewer experience requires deeper, real-time insight into how video performs at the session level. That means moving beyond averages and aggregates to understand the specific conditions that impact playback quality as they happen.

Forward thinking streaming teams are beginning to adopt this model of granular, actionable visibility. With the ability to detect anomalies early, isolate root causes, and correlate quality issues across environments, they’re shifting from firefighting to prevention, reducing churn and improving satisfaction. Analytics platforms are helping make this shift possible by embedding observability into the heart of the video workflow to help providers scale diagnostics and accelerate resolution.

Enhanced Visibility = Higher Quality Streams
To deliver video in the best possible quality, service providers also need to optimize video encoding. And to do this, providers need to understand how content is performing in the real world, across devices, networks and regions. Relying on fixed bitrate ladders can lead to quality issues, especially when viewing conditions vary as much as they do today. If video providers can access metrics such as download speed, resolution changes, and the bitrates actually being delivered, they’re much better equipped to work out where the viewing experience is falling short.

Another key aspect of delivering a quality viewing experience is ad playback. Issues such as stalled playback, buffering and blank screens can quickly ruin the viewing experience. However, ad performance can be difficult to monitor, especially with server-side ad insertion (SSAI), where these issues can often go undetected. To address this, providers need tools that make it possible to track how ads are really performing, so that ad playback issues can be spotted and resolved quickly.

Raising the Bar
Viewers today have high expectations and little patience for playback issues. Most will abandon a stream within seconds if it buffers, crashes, or the quality drops, and many won’t return after a bad experience. To meet these expectations, video providers must account for a wide range of variables, including device type, screen size, connection quality, geographic location, and user behavior.

Observability enables teams to manage this complexity by monitoring key playback metrics as they occur. This gives providers the clarity and speed needed to resolve problems before they affect the viewer. The services that succeed won’t just be the ones delivering popular content, they’ll be the ones delivering it flawlessly, every time, on every device, without users ever noticing the work going on behind the scenes.

BARCELONA—Viewers worldwide viewed more on demand and linear TV than in the previous two years, according to NPAW’s 2023 Video Streaming Industry Report released this week. 

The amount of time the average user spends watching content per day increased globally in 2023 by 12% for VoD and 4% for Linear TV content. This pivot might signal an increase in consumption or a shift in the consumers' attention toward a more selective number of platforms, according to the researcher.

Previously, NPAW had blamed declines in VOD and linear TV on fragmentation in the streaming market but growth in Connected TV sales and “Quality of Experience” helped revive the sector in 2023, the researcher said. 

(NPAW’s 2024 Predictions: Role Of AI In Video Streaming To Climb Next Year)

The most notable finding of the report is the consolidation, in the second semester of the year, of the upward trend in daily user engagement for VoD and Linear TV, at 4% and 3% respectively, NPAW noted. Episodic content continued to dominate as the primary type of VoD content for another year, capturing 67% of the total global playtime. However, movies saw a slight increase, accounting for 26% of all VoD minutes streamed in 2023.

Additionally, 2023 was also a year of significant improvements in the Quality of Experience front, both globally, with a global decrease of 38% in the buffering ratio, and regionally, with regions like Asia experiencing an 18% increase in the average bitrate. These boosts signal that there is still room for optimization of the viewing experience.

At the device level, the growth in Smart TV sales helped drive up total playtime by up to 8 percentage points. NPAW concludes that this reaffirms that consumers overwhelmingly prefer big-screen experiences when streaming video content, reserving smaller-screen devices like smartphones for casual streaming or while on the move.

"These findings underscore a significant year of renewal and growth for the streaming sector. The increase of up to 12% in daily playtime per user, coupled with tangible improvements in the Quality of Experience, highlights how the industry is consolidating and evolving to meet and exceed viewer expectations," said Ferran G. Vilaró, CEO and Co-Founder of NPAW.

"As providers navigate this journey of optimization, leveraging deep end-user behavior and experience data will be crucial in capturing the attention and loyalty of consumers. Understanding the insights derived from NPAW’s 2023 Video Streaming Industry Report will play a pivotal role in shaping successful streaming strategies moving forward." Vilaró added.

The data analyzed in the report were extracted from the NPAW Suite from January to December 2023 and compared to data from 2022. They represent real-time data from over 190 global NPAW clients, including leading OTT providers, broadcasters, and telecom operators. 

The report can be downloaded here.

The video streaming market continues to grow rapidly. A recent report from Allied Market Research found that the value of the global video streaming market will approach $150 billion by 2026. While the rise in video streaming represents an opportunity for revenue growth, it’s also a challenge. Today, quality of experience (QoE) is the difference between having a successful streaming business or not. Operators need more control over streaming quality and bandwidth in order to optimize network usage and deliver a fairer, more consistent QoE for all users.  

The network has a critical role to play in the stream selection process and in regulating the amount of bitrate distributed to each connection individually. Ultimately, the network can help video service providers achieve control over the quality of their services and secure rational usage of their infrastructure. Performing stream selection on the server side—as opposed to the client side—is greatly beneficial in terms of delivering low-latency streaming and enabling fair distribution of the available bitrate between the devices of a same-home network. Server-side stream selection reduces transport network expenses and significantly improves QoE,  ultimately impacting the loyalty of end-users.

IMPLEMENTING SERVER-SIDE STREAM SELECTION

There are two different processes that make up stream control: The first consists of measuring the bandwidth between the server and the clients for each and every connection. The second component involves selecting a video quality stream and often overriding the client’s decision. These two functions work together, as the server can use its bandwidth assessment as the main criteria for selecting the appropriate quality, replacing or complementing the ABR client’s selection. 

However, the components can also be used separately. It is possible to use bandwidth measurement just for analytics or apply streams selection based on other criteria, typically using preconfigured logical rules or commands sent from the service provider’s management system. Both processes have a very limited impact on the server’s hardware resources. Therefore, no redimensioning is required to activate server-side stream selection.

Once the system has a way to seamlessly impose a different quality stream than the one requested by the client, it is possible to act on these decisions from a centralized operation point. The available bandwidth on the link is no longer the only criterion for stream selection, as it is for the traditional client-side approach. Commands can drive or influence decision-making, for instance, to impose a reduction of bitrate on all the connections of a network segment getting near saturation, possibly taking into account that some users, content or devices have to be preserved as much as possible.

Server-side stream selection is simple to implement. The technology can be implemented at one particular point of the distribution chain or at all levels, depending on the targeted use case. It is fully compliant with ABR streaming standards, including Apple HLS and MPEG-DASH, and doesn’t require any adaptation to existing ecosystems. 

BENEFITS OF SERVER-SIDE STREAM SELECTION

A key advantage of using server-side stream selection for video streaming is that it enables service providers to deliver a better, more consistent QoE with fewer degradations in quality.

Server-side stream selection takes into account network conditions when determining which quality stream to deliver to each connection. This allows service providers to focus on the available resource where it matters most, leading to an improved average QoE for all viewers. 

In standard ABR distribution systems, when monitoring indicates a probable video degradation, there’s little that network operators can do since they don’t manage the different end-devices streaming from their network. Placing the network bitrate distribution on the server side is an opportunity to act on the issue and resolve it before it has any significant impact on QoE. The servers on the network can dynamically measure the available bandwidth, ensuring very accurate and valuable information on the state of the network. Through advanced analytics, service providers can determine how network investments will translate into an improvement in video QoE, prevent events that can cause QoE degradation, anticipate streaming limits and manage bandwidth intelligently.

Another advantage of server-side streaming is low latency. Typically, streaming live content without adding latency requires breaking up the several-second video segments into smaller chunks. This causes issues on the players deployed today, as they rely on the existence of these segments to choose a quality stream. Server-side bandwidth measurements rely on the use of lower-layer TCP information provided by the operating system and, consequently, remain immune to that limitation. 

When all video sessions cannot be served at the best available quality—for example during peak hours—the best way service providers can secure optimized overall quality is to maintain control over which sessions may be affected by this resource limitation. Server-side stream selection can gracefully and dynamically help by preserving the bitrate of higher value content such as live events, streams serving the main screen rather than mobile devices or of premium subscribers. 

Through home network orchestration, server-side video streaming allows service providers to dynamically distribute bandwidth resources among different devices and monitor bandwidth at the home level. For example, let’s say a household has 10 Mbps internet access and the children are consuming 8 Mbps on their smartphones. If the parents come home and want to watch an important event on the main TV screen, then the orchestration process can reserve 4 Mbps to the main screen, ensuring a minimum QoE, while seamlessly lowering the bitrate delivered to the smartphones to a level that is still very acceptable for such screens.

CONCLUSION

One of the main reasons for ABR streaming’s success has been its adaptability to any kind of device and to different types and qualities of connections. Yet, having the client choose which resolution and bitrate to use for every video streaming session leads to a certain number of limitations on how network resources are used and how the different devices share these resources. These limitations can be alleviated with a server-side approach, such as Broadpeak’s S4Streaming solution. S4Streaming gives back control to the service provider so that they can better optimize resources and ensure that they are used wisely to deliver a fair and consistent quality of experience to all users.

Damien Sterkers is product manager for Broadpeak.

As OTT (Over-The-Top) technology has gotten more mature and established robust standards over the years, the concept of OTT monitoring is gaining popularity. With customer expectations soaring, it’s vital for OTT providers to deliver superior-quality content. To deliver Quality of Experience (QoE) on par with linear TV broadcast, the entire system, starting from ingest to multi-bitrate encoding to delivery to CDN must be monitored continuously.

Streaming service providers and broadcasters understand the complexities involved in OTT delivery and the media distribution chain—from content acquisition to actual transmission. OTT monitoring primarily involves monitoring the status of all key elements in the pipeline starting with the media source, encoders, decoders, output of Content Distribution Network (CDN), etc. The amount of media consumed over OTT is growing at a staggering pace, and people’s viewing habits are rapidly moving towards a unified multiscreen experience. These rapid changes demand a strong comprehensive monitoring solution that takes care of all the aspects in the OTT chain and makes it easy for people at various levels in broadcast operations to monitor and control the entire system.

The primary goals of monitoring OTT are to guarantee Quality of Service (QoS), Quality of Experience (QoE) and compliance monitoring. These quality metrics include media quality for confidence monitoring and delivery of captions, both for primary and secondary languages, to increase user engagement. In addition to QoS and QoE monitoring, broadcasters have another challenge to track viewership using ID3 Tag technology.

A comprehensive monitoring system includes the following:

• Encoder status, start/stop streaming, alarms, etc. It is essential to ensure proper encoding of content that comes in from multiple sources. As opposed to traditional broadcasts, live cloud OTT workflows are much more complex as it involves delivery to individual devices with multiple profiles. It is critical to monitor for all possible issues like frame misalignment, syntax errors, over-compression, absence of markers & metadata, dropped packets, and compliance slips as these issues can negatively impact a positive viewer experience. A monitoring system should be able to monitor end-to-end workflow, raise alarms, report streaming status, and notify issues that occur anywhere in the end-to-end system. This helps to quickly resolve issues before the end customers get affected.

• CDN status (concurrent connections, origin, visitors’ statistics, etc.) The CDN plays a crucial part in delivering content to end customers. It is critical to monitor the origin of the content, viewers’ statistics, and CDN edge.

• RTMP Surveillance and analysis of local encoders' output

• External report on streaming services availability at CDN outputs, ping, etc.

• Monitoring of website accessibility, CDN status, network status, etc.

• Video-on-Demand accessibility and Quality Control Analysis. In addition to live streaming, VOD clips must be monitored for quality.

• End-user experience status report

• End-user experience report for developers

• Server-Side Ad Insertion

• Captions

OTT Monitoring Advances

OTT monitoring is evolving at a rapid pace. Early on, OTT monitoring had largely been limited to IT infrastructure monitoring, and to some extent, monitoring streams at edge locations as a proof of delivery.

With the market growing, more and more OTT broadcasters are starting to see the value of monitoring. Efforts are on to converge OTT platforms and optimize streaming protocols.

OTT providers use various streaming protocols for media delivery, including Real Time Streaming Protocol (RTSP), Real-time Transport Protocol, Smoothing Streaming, etc. Often, the receiving devices support only one or two protocols requiring service providers to stream their content in multiple protocols. Nowadays, the media streaming process is largely being performed in the cloud, making use of the CDN. The data that goes in and is delivered out from the CDN needs to be monitored to guarantee QoE.

Also, transmission errors can occur while transporting media data using the IP network; errors could pertain to packet latency, jitter, packet loss, etc. This makes it important for service providers to introduce complete quality monitoring solutions, including video monitoring at the CDN data centers and at the edge locations.

Cloud-based monitoring is extremely efficient as it works like a virtual machine (VM) that can be moved to any place within a particular network. This makes it easier to store, manage and process data, thereby saving a lot of time, effort, and money. Cloud-based monitoring allows monitoring in geo-restricted streams and regions as well as scaling up monitoring, particularly with server-side ad insertions.

The State of Logging and Compliance

Logging and compliance of traditional broadcasts are being done by a lot of companies. However, these types of loggings neither cover OTT content nor do they cover content that has advertisements inserted at the MSO. Relying on logs to understand what happens to the content could turn out to be wrong.

OTT operations have changed the way that advertisements are delivered; they can specify ads for a specific market or a community. However, these broadcasts generally are poorly monitored, or at times not monitored at all compared to traditional broadcasts.

Thus far, there has been no major push from the industry to monitor OTT due to minimal FCC compliance requirements and a lack of an industry standard on OTT. Rapid innovations are being made in OTT and broadcast video delivery. Content owners are under constant pressure to respond to the surge in OTT content consumption.

OTT video service delivery technique is making use of dynamic adaptive video streaming over HTTP (DASH). Over the past year, MPEG-DASH has seen rapid growth and has also gained strong ground in OTT technologies.

The ATSC 3.0 specification is making use of DASH for Hybrid Broadcast-Broadband TV. ATSC 3.0 offers broadcasters the opportunity to deliver broadcast and broadband services to provide better a television experience.

Statistical multiplexing (Statmux) helps in providing high-quality ATSC 3.0 streams and facilitates efficient use of available bandwidth and supports to optimize bitrates and video quality of channels that share the same physical layer pipes. OTT content travels via a broadband connection in the receiver and DASH acts as the enabler in this scenario.

OTT Monitoring Best Practices

According to the image above, the entire OTT video delivery process is carried out in three different layers. The three OTT Layers: Operations Layer, Media Layer, and Distribution Layer are monitored by three groups of API Integration. These API integrations monitor content throughout the workflow right from content acquisition to content consumption to content delivery. Monitoring includes minute observations for all touch points of the workflow and provides all the technical error data, which works as an alarm, and helps provide users uninterrupted video.

The key to improving operational efficiency is to create a fault-tolerant OTT streaming infrastructure. In case of an eventual fault, reducing the resolution time to minimum is fundamental. As highlighted in the diagram above, a comprehensive and easy- to-use dashboard that captures and controls the health of the entire infrastructure real-time can add value to the overall operation.

Current Cloud-based Challenges

• Multiple protocols being used to support a wide range of devices OTT providers use multiple streaming protocols for media delivery, and the receiving devices support only one or two protocols, which prompts OTT service providers to stream their content in multiple protocols. The data that goes in and comes out from the CDN needs to be monitored to guarantee the quality of experience (QoE).

• Last mile monitoring cloud-based OTT monitoring must handle the problem of last mile monitoring. Cloud-based applications must be monitored at various levels including user experience. Physically, there is a considerable distance between where the application is running and where the users are. Conventional monitoring, being part of the application environment, makes it technically difficult to monitor how an application behaves at the user’s end—be it a mobile app or a web browser. Internet issues specific to each region are to be monitored. Bad internet connectivity could impact user experience. 

Multiple multi-bitrate results in an increased resource requirement. Another major challenge of OTT content streaming is that the devices and networks are far more diverse than those assembled in controlled environments like satellite, cable, etc. An adaptable architecture is necessary to address the varied needs of network conditions and device requirements.

Streaming videos are being delivered to a range of devices, including computers, smartphones, and tablets, making the monitoring process more complex. Bitrates and profiles with which each variant is encoded are different. So, from a monitoring and quality control perspective, operators need to put in more efforts as they have many versions for each piece of content.

Networks used by some of the devices need to tolerate dynamically fluctuating characteristics. Also, the underlying technology used in adaptive bitrate streaming is not standardized.

The aforementioned are not specific to all cloud-based applications, but to OTT monitoring in general.

Future OTT Monitoring Fixes:

• Simulating last mile bandwidth variations
• Convergence of protocols
• Limiting the number of manifests and reducing error rates while switching from one manifest to another

Future Advances in Cloud-based Monitoring of OTT Content

• Converging of streaming protocols, a widely adapted industry standard; MPEG-DASH for instance
• A protocol based on segmented delivery with near zero latency
• Advances in video compression, partly promised by H265 and efficient utilization of available bandwidth for best QoS

Hiren Hindocha is the CEO of Digital Nirvana. 

Once requiring its own specialized networks, television is rapidly becoming just another cloud-based service being delivered by general-purpose IP networks, both wired and wireless. But streaming video, nonetheless, is still a big challenge for network operators. Sharing a network with other content and applications raises both traffic engineering and security challenges. Customers have high expectations from years of rock-solid cable TV service, so service assurance is critical. One of the key pieces in delivering customers the quality of experience (QoE) they expect is a new breed of network analytics.

In the coming five years, the networking industry will be moving from 4G LTE to 5G, the new wireless standard. Promising +10X download speeds, 5G will open up the world of mobile television. While this is getting a lot of attention in the television industry, one of the other significant shifts that will come with 5G is a move to a cloud-based architecture across both mobile and fixed networks based on virtualization and software-defined networking (SDN).

Virtualized, SDN networks have long been used in cloud data centers, but the technology is swiftly migrating to the wider area network and, with 5G, to both wired and wireless networks. One of the reasons this approach is being embraced is that SDN networks running on virtual platforms are cheaper to run, not simply because they use general purpose computing platforms instead of specialized networking hardware, but they also use the physical resources more efficiently.

This efficiency is the result of them being more dynamic. SDN makes IP networks far more flexible and scalable. They can re-allocate under-utilized hardware resources instantaneously to meet whatever needs arise. Thus, if Ninja invites Drake to play Fortnite with him, instantly bringing 500,000 new viewers to his video stream, the SDN controller can spin up the network processing resources to meet the sudden spike in demand.

A PATCHWORK JUNGLE

Dynamically re-assigning IP resources to meet sudden shifts and spikes starts with good analytics. However, the analytics traditionally used in data centers aren’t enough. The wide area network outside the data center is a jungle made up of a patchwork of telecom operators, ISPs, MSOs, long-haul carriers and content delivery networks (CDNs). Traffic between them traverses a continually shifting constellation of peering points. Video streams can come from anywhere, and demand from viewers can shift in the blink of an eye.

Making matters even more challenging, the traditional network analytics approaches aren’t sufficient. Wide area networks have depended for decades on Simple Network Management Protocol (SNMP) to understand what was going on in the network. But as its name suggests, SNMP isn’t up to the complexity and the real-time demands of today and tomorrow’s networks.

The other key analytics technology is deep packet inspection (DPI). Deployed in routers with the processing power to look inside each packet at line speed and identify the payload, DPI can be expensive. Thus, operators tend to deploy the technology to do spot checks at key points in the network. However, its biggest challenge is that the majority of video streams are now encrypted, so DPI has no way of identifying the payload.

MAP AND CATALOG

The good news is that there is a wealth of network data that can be used to bolster SNMP and DPI. The problem is that this data is currently collected in silos across multiple different systems—making it extremely arduous (or impossible) to use them for better planning and traffic engineering. Thankfully, recent big data techniques, such as streaming vector, column-store databases, are coming to the rescue.

Using only public available data of the same sort that Google collects, it is now possible to literally map out the entire internet and catalog the development, over time, of everything happening on it. For instance, it is possible to map IP flows from source to destination. If you know the destination, for instance, a Netflix cloud server at AWS, then you know what service your customer is accessing. With this technology, operators are able for the first time to understand what over-the-top applications are traversing their network and how they are impacting it.

With this holistic view of the entire network in real-time, operators can adjust their resources for dynamic service assurance. They can also work intelligently with partners upstream, such as content providers and cloud providers, as well as downstream, such as neighboring networks and CDNs, to adjust their own resources. Is peering point A acting as a bottleneck, while peering point B is under-utilized? Would a different cloud provider better serve a content provider, such as Netflix, in a newly served market?

As television moves into the cloud era of 5G, it is leaving the very predictable confines of its purpose-built cable networks to become part of the world wild web. While there are all kinds of good reasons to go this route, it also raises issues around network management and service assurance for increasingly demanding customers. Happily, network operators, content platforms, cloud providers and distributors have new tools to take the guesswork out of network planning and troubleshooting. These techniques are already being deployed in 90 percent of US cable operators. Along with SDN and network virtualization, which are coming with 5G, they will form part of the core technology of this exciting and challenging new world.

Kevin Macaluso is General Manager, for Nokia’s Deepfield group.