Why Ad Monitoring on Real Devices Is Essential for OTT Performance
For decades, traditional broadcast television was supported by advertising. The equation was simple: broadcasters made original content, which aired on television to viewers at home. Advertisers, eager to reach those viewers, would pay the broadcasters to run their advertisements during the programming. Ad slots during more popular content, which allowed for more “ad impressions” (views), would cost advertisers more than less popular content. This model lives on: 2023’s Super Bowl brought in a record-breaking 200 million viewers, and each ad shown to them cost $7 million dollars to run.
OTT streaming video entered the mainstream with different revenue models. YouTube launched in 2005 without any ads or admission fees. iTunes offered transactional video on demand (TVOD), a familiar pay-per-view model in which users could purchase or rent videos they wanted to watch. Netflix’s 2007 launch popularised subscription video on demand (SVOD), revolutionizing the industry by earning revenue through monthly subscribers instead of advertisers.
Eventually, everything comes back around. In 2007, YouTube began running ads before and during their online videos (and, in 2015, they launched an SVOD option that allowed users to skip the ads entirely). Netflix, along with top streamers Max, Hulu, and Disney+, has added a cheaper subscription option for viewers who are all willing to sit through ads. Advertising-based video on demand (AVOD) services like Freevee and Pluto TV are entirely free to watch but include ads, as do the recently-booming free advertising-based streaming television (FAST) channels that show programming on a set schedule. Even short-form videos on social media platforms like TikTok and Instagram, some of the most viewed streaming videos online, regularly run ads in the middle of their users’ feeds.
So, it’s clear that advertisements are important to streaming video today. A 2022 Comscore study concluded that American households were adopting AVOD services at a higher rate than SVOD. Unfortunately, inserting ads into streaming video isn’t as simple as it was with broadcast television – and it can cause many problems.
How Ads Are Dynamically Inserted in Streaming Content
While old-school cable TV providers would often package ads with their content and send them out together, streaming ads are typically “dynamically inserted” into the content. This means that they come from a different origin server than the rest of the content and are being added in on the fly. The content is pulled from its original source to an online stream and then to a video player, which inserts the ads in real time.
The most popular form of dynamic ad insertion utilises SCTE 35 ad cues. These cues are inserted into the content and trigger “ad pods” when the video player reaches them. The cues will specify how long an ad pod is and how many ads are needed to fill it. From there, the ads are pulled from different sources, often matched to the location or interests of the viewer watching. The technical path to dynamic ad insertion (DAI) is complex, which means there are plenty of opportunities for issues to arise.
Common Issues with Dynamic Ad Insertion
Issues with streaming video ads are so common that any reader has almost certainly encountered them before. Significant buffering can occur before and during ad pods, and ads often play at a completely different volume or picture quality than the content they’re interrupting. Sometimes ads are triggered too early or too late, or run the wrong duration for the ad pod they’re filling. One single ad can appear multiple times in an ad pod or repeat in every following pod for the duration of the content, aggravating the viewer instead of selling the intended product.
Ad issues are a hot topic for FAST channels as well as AVOD. Since FAST airs on a set schedule to simulate the feeling of cable TV, ad pods are longer and often filled with slates instead of ads. These slates can be countdown timers to the content returning or confusing blank screens that give the impression of the stream crashing; and that’s if the stream didn’t actually crash. In Witbe’s independent research, we have observed up to 30% of FAST channel viewing sessions are affected by an ad insertion issue.
The Importance of Reliable Ad Monitoring
So how can these issues be fixed? The question is of vital importance to video service providers, who depend on selling ad impressions to advertisers. If viewers are encountering issues from ad insertion, they are likely to stop watching the video, costing providers current and future ad impressions. If advertisers claim that their ads are not being aired properly, they may withhold payment or avoid working with a provider in the future. In the end, poor ad performance reflects back poorly on the video service, justifiably or not.
Unfortunately, there isn’t one single common mistake that triggers all these issues, which could have begun at any point in the complex video delivery pathway. Instead, there is one single method to discovering the issues and beginning the process of fixing them: ad monitoring.
Like all OTT video, verification is key for ads. Performance cannot be improved until it can be reliably verified. Ad monitoring on real devices allows providers to measure the quality of experience (QoE) being delivered to viewers at home. For catching dynamic ad insertion issues, monitoring on the same apps, devices, and networks that viewers use is essential. Furthermore, automation can help monitor at the large scale that video service providers require.
Specific criteria must be set to properly measure performance. For dynamically inserted ads, it is essential to monitor the audio level and video quality of the ad compared to the prior content, the duration of the ad, if any slates appear within an ad pod, and, of course, if the ad itself even plays. Recording ad performance is helpful for recognizing common issues over time and for verification with advertisers asking for proof of delivery.
Ad Monitoring Is Essential for Service Performance
The ultimate goal of every OTT video provider is to deliver a smooth viewing experience to those watching at home and on the go. While dynamically inserted ads are essential to providers, they often threaten the quality of that viewing experience. The only way to verify, and in turn improve, streaming ad performance is by monitoring on real devices and networks. With the help of reliable ad monitoring, providers can deliver a superior viewing experience, increasing ad revenue and building positive word of mouth within the ever-expanding industry. For streaming video providers, ad monitoring isn’t just helpful – it’s essential.
[Editor's note: This is a contributed article from Witbe. Streaming Media accepts vendor bylines based solely on their value to our readers.]
Related Articles
Vijay Sajja of Evergent explains how OTT and FAST market consolidation will spark a new wave of alternative monetization strategies.
26 Sep 2023
There are two broad categories of CTV ad insertion: Server-side (SSAI) and Content-side. We've had the SSAI vs. CSAI debate many times before, with SSAI often coming out on top in recent years, but the uptick in FAST viewing is changing the conversation for many stakeholders in the current media landscape. Effective ad delivery is far from a one-side-fits-all proposition these days.
28 Jun 2023
Wilichinsky offers a look inside Paramount's ad-tech stacks and strategy.
19 Dec 2022
As last week brought the long-awaited, and arguably overhyped debut of Netflix's new ad tier, I had the chance to sit down with Ed Laczynski, CEO of enterprise OTT app and platform builder Zype—a perennial Streaming Media 50 honoree—to talk about what Netflix's move means in the context of the ever-evolving OTT monetization landscape, and get some perspective on longtime, current, and emerging trends. We also talked a bit about how will impact the current and coming ad inventory ecosystem.
11 Nov 2022