Between 2019 and 2025, ad-supported revenue for VOD is projected to increase by 120%, from $24.3 billion to $53.5 billion, according to Digital TV Research. A major driver for ad revenue growth is the expectation to target ads to specific audiences. OTT delivery of video programming gives advertisers and their agents valuable access to specific audience demographics. In turn, OTT service providers can demand higher prices for ad spots. So, while the future looks bright for OTT services, there is a dark spot on the horizon in terms of ad-related failures or low fill rates that can negatively impact not only revenue, but even brand perception. Left unaddressed, these issues could threaten the business viability of OTT services.

The promise of targeted advertising has been talked about for years, yet it is still in its infancy. This is partly due to the sheer complexity of the workflow and a dependance on multiple components all working together perfectly to deliver a smooth and individual experience. The technology behind the curtain is called Dynamic Ad Insertion (DAI). It is a client- or server-side video ad technology that allows content owners and distributors to serve OTT video ads into live or VOD content. DAI stitches the video and ads into a single stream and produces a seamless TV-like experience without latency or buffering between content and ads.

Getting OTT Ads Right

There are many points of failure for ads delivered OTT. The challenge of getting SCTE-35 ad signaling markers to propagate throughout an OTT distribution network are non-trivial particularly when the underlying network infrastructure changes and evolves. It’s also become more complicated because many broadcasters have both a traditional Linear TV path plus an OTT path to contend with. Getting SCTE-35 into Adaptive Bit Rate (ABR) streams for DAI requires yet another workflow.

Conditioning

OTT ads need to be correctly formatted or conditioned for the devices they will be seen on before they get trafficked. Maybe the viewer is watching an HD program and the ad is a smaller internet video at 320x240. When it gets blown up by the players, it looks terrible. The program content itself must also be properly conditioned for ad insertion. Encoder boundary points must be aligned with ad signaling for a seamless transition from program to ad and back again. Without it, an ad might cut into the program early, or the viewer might see the beginning of the next scene and then the ad cuts in. Detecting these attributes and conditioning the assets themselves is critical to enabling a quality experience for viewers.

Ad Policies

Ad policies dictate the business rules about ad frequency and uniqueness as well as where the ad is sourced. It’s important to know if the ad policy is being violated by trafficking the same ad repeatedly. If there hasn’t been an ad break in 30 minutes, we need to know that too. Maybe the ads are being served from an ad network that is unfamiliar? It’s vital when synthesizing the end user experience to send alerts in real time when ad policy is not being obeyed. Ad policies are the “rules of engagement” with the audience, and they must be adhered to.

The workflow complexity required for ad personalization introduces additional handoffs within the ad system. It’s during these handoffs where problems typically occur. When a viewer clicks “play” on a piece of content, the video player will request a manifest from the DAI vendor. The manifest is a set of instructions that inform the player on tasks like where it can locate video files and how long to play them for. Assuming that the manifest is accurately conditioned for the downstream manifest manipulator, ad customization for a specific viewer is possible.  The manifest manipulator is tasked with exchanging information with ad decisioning servers, and if all goes well a customized commercial gets sent back enabling every viewer to have a personalized experience. When it goes wrong, a user might miss the ad completely, get stuck watching a provider slate, or experience poor-quality streaming caused by slow ad networks or low-quality audio and video content. In worse scenarios, the media player can completely reset due to poor handoffs between content and advertisement.

Visibility Is The Key To Successful Monetization Of DAI-based Ads

Having visibility into ad behavior is a key part of ensuring a monetized, healthy advertising distribution workflow at a time when many ad insertion workflows, especially for live OTT, frequently fail. One of the biggest pain points is simply missed opportunities—there was an opportunity to traffic an ad, but for some reason, the viewer didn’t see it. The other big pain point is that the ad that got trafficked was somehow disruptive. It could have been a format conditioning problem, an ad that repeats over and over or blown out audio levels. Broken ads are a big part of the problem too.

To protect and advance ad revenue, it’s paramount to have as much visibility of the workflow as possible. Providers must be proactively alerted to issues and have the means to easily find and fix them before viewers go elsewhere. Monitoring must be deployed at each stage of the OTT advertisement lifecycle whether it’s on the ground, in the cloud or a hybrid combination of both. Only by providing visibility from contribution feeds and following SCTE-35 ad event messaging through the transcoding and packaging process, to manifest conditioning and ultimately into the CDN delivery post-manifest manipulator can we ensure that problems don’t slip through the cracks. It’s important to not only track the ad OUT and IN events, but also ensure that ad insertion is properly filling the entire opportunity duration. Ad delivery must also be sufficient to handle the timing constraints of a live streaming service.

Some organizations have tried to do this with multiple systems from different vendors. They soon realize that correlating information across disparate systems introduces yet another set of problems when trying to pinpoint a fault. To find and fix faults as fast as possible requires a unified “single pane-of-glass” approach—A centralized video monitoring and management system that correlates very complex and diverse sets of data from across each point in the ad workflow. A capable solution should present operators with a simple graphical view of ad opportunities, health status, and proactive alerts to any issues that may arise. Having a dedicated, dynamic advertisement report, enables users to quickly identify large scale issues that are significantly affecting ad opportunities. When problems are more nuanced, engineering should have the ability to quickly drill down and start investigating specific events with correlated audio and video content analysis as well as manifest conditioning and streaming impairments that can cause advertisement failure.

The Future Looks Bright For OTT Ads

With so many moving parts, it’s true that OTT advertising and DAI technology have challenges to live up to the promise of successfully providing targeted ads to consumers. The first step in making the system work for both content owners/providers and advertisers is to have a comprehensive monitoring solution in place so everyone can ensure that ads are playing how and when they are supposed to play. That is exactly what the iQ DAI monitoring solution is built for. Our video management system, iVMS ASM, collects monitoring data from its probing technologies; Inspector LIVE, Sentry and Surveyor ABR family. The technology performs ad-insert data correlation between source, origin and CDN locations enabling operations teams to dynamically visualize performance of the ad workflow and identify issues quickly. The iQ ad-insertion monitoring solution not only monitors and provides real-time health status, it also includes analytical data such as expected vs actual avail times, avail minutes per hour, and ad-ID information that can be used for analyzing the OTT advertising workflow and making better business decisions. These solutions provide entire workflow analytics of advertisement opportunities including source messaging, manifest conditioning, manifest manipulation, as well as ad audio and video quality monitoring from contribution to CDN edge cache.

For pure cloud-based SAS workflows, Cloud Stream Monitor allows users to deploy this capability on their cloud vendor of choice, providing a global monitoring solution for content from over 90 unique locations across the world.

A comprehensive ad-insertion monitoring solution supporting Dynamic Ad Insertion technology must offer end-to-end visibility into this ever-changing network. It should isolate video, audio, and streaming metrics for ad breaks, visually validate ad breaks with thumbnails and effectively close the information gap between a service provider and an ad buyer who believes there was a failure to fill. Having reports that prove ads played across multiple regions is also an invaluable tool in this new world of OTT delivery. With an end-to-end view of the entire ad delivery chain, it is easy to identify and mitigate issues thereby increasing the opportunity an OTT service provider has to monetize content.