VOD- Server-Guided Ad Insertion: The future of viewer engagement and monetisation

CSAI vs. SSAI: We discussed the advantages and limitations of Client-Side Ad Insertion (CSAI) and Server-Side Ad Insertion (SSAI). CSAI offers personalized ads but struggles with cross-platform implementation and seamless ad-break transitions, while SSAI is resilient to ad blockers but is more complex & expensive to scale.
Introducing SGAI: We unveiled Server-Guided Ad Insertion (SGAI), a hybrid approach combining the personalization of CSAI with the seamless transitions of SSAI. It additionally brings higher fill and rendering rates. THEO and Google are collaborating to advance this technology.
Future of Monetization with SGAI: We explored how SGAI is set to shape the future of ad monetization, delivering more personalized and relevant ads, and also new innovative formats, ultimately driving greater user engagement and revenue.

Webinar Transcript

Michel: Excited, right?

Lemu: Yes, sir.

Michel: We're talking about Server-guided Ad Insertion (SGAI) today. And, I have two cool panelists with me. I'm Michel, I work for THEO Technologies, responsible for the engineering team and the customer success team. So that makes a very interesting mix from customers to technology, understanding a bit what the customers are asking for and what our tech can do. Today, talking with two good colleagues from Google.

Lemu: I'll jump in here, I'm Lemu Coker. I'm a CTV Ecosystems Partnerships Manager. It's a lot of words, but effectively, I just work with publishers, third party vendors, and our internal Google teams to ensure that however publishers want to work with us, they can. And we will work with all the different partners that our publishing partners work with to make sure that monetization flows seamlessly and any integration works seamlessly.

Sourya: Hi everyone, I'm Sourya Roy, I'm a product manager at Google. And I'm a part of our Google Ad Manager team. I mainly look after, among other things, our ad insertion efforts, so anything which has to do with ad insertion, Server-Side Ad Insertion (SSAI), Client-Side Ad Insertion (CSAI), as well as our third-party integrations. So yeah, thank you for inviting us. Super excited to be here.

Michel: Absolutely! It's already interesting where you start - as a video player company, we have a lot of integrations with Client-Side Ad Insertion (CSAI) and Server-Side Ad Insertion (SSAI).

Maybe a first question, Sourya - In your experience, where are we currently with Client-Side Ad Insertion (CSAI) technology, Server-Side Ad Insertion (SSAI), the benefits, maybe also the limitations? What's your view on that?

Sourya: Yeah, I think right now we're in a very interesting position. So most of our large customers, they all use SSAI, some sorts of SSAI. Server-Side Ad Insertion (SSAI) is amazing because it has an amazing watch experience. It's super seamless, there's no buffering.

Also, you have a lot of device coverage: you have CTVs, mobile, desktop, everything supports Server-Side Ad Insertion (SSAI). So I think right now we're seeing a lot of our customers are using Server-Side Ad Insertion (SSAI). And people are moving from Client-Side Ad Insertion (CSAI) to Server-Side Ad Insertion (SSAI). So that's great. But we are seeing now with Server-Side Ad Insertion that some challenges are coming up. The big challenge we find is - when you're implementing SSAI, it takes time to implement, because you have multiple vendors - you multiple SSAI providers, you have multiple CDN providers, encoders, and to make everything work together will take some time and make sure everything is working fine.

The second thing also what we're finding right now - is many of our customers are trying to stream low latency, as well as doing things like streaming to 10 million concurrent users, 20 million concurrent users. And let's say if you have one is to one addressable streams for like 20 million concurrent users, and let's say you have maybe two seconds segment duration, a client is making one to two requests every second, multiplied by 30 million requests. So we're seeing sometimes the network is getting tied, the bandwidth limitations that coming into the play. So those are some of the challenges which we're seeing right now on Server-Side Ad Insertion (SSAI).

Coming back to Client-Side Ad Insertion (CSAI), the great thing about client side, it's super easy to troubleshoot, it's super easy to implement. But the problem with Client-Side Ad Insertion (CSAI) is the watch experience. With the traditional CSAI, which we have, you see a video, then you have the buffering, because the client device has to download the ad. It has to process the ad one by one. So yeah, it's an interesting situation right now.

Michel: Of course! Today we're also talking about Server-Guided Ad Insertion (SGAI), a bit of a new kid on the block where we redistribute the responsibilities a bit. So server still indicating the breaks but the client doing the splicing, server still offering the fully pre-transcoded ad manifest ready to play basically. When did the concept first get on your radar?

Sourya: Yes, Server-Guided Ad Insertion (SGAI) is a very cool concept. To give you a high level what SGAI is, it's Server-Guided Ad Insertion. So it takes all the benefits of Server-Side Ad Insertion (SSAI), all the pre-processing, ad unwrapping, falling through all the VAST tags and everything. And it provides a manifest to the player, so the stitching happens in the player. So it's a very simplified way where you mix the best of Client-Side Ad Insertion (CSAI) and the Server-Side Ad Insertion (SSAI).

So at Google, we have been experimenting with Server-Guided Ad Insertion (SGAI) for quite a while. I think a big confliction point came was in 2021. So Apple announced HLS Interstitials. So HLS Interstitials is the standard of doing Server-Guided Ad Insertion (SGAI) and Apple launched that with HLS. HLS is one of the popular streaming specs along with DASH. And before even that, we have had Disney and Hulu. They had been championing the use of SGAI for a long time. So a lot of the Hulu apps have been running Server-Guided Ad Insertion (SGAI). So they have been to many of the conferences talking about the benefits of Server-Guided Ad Insertion (SGAI). So we found that to be very exciting. Basically, we were very interested in the flexibility and the ease of use that Server-Guided Ad Insertion (SGAI) does. So that's why we were like - okay, let's try to work this out and see how we can make this simpler so we can expand it to a lot more customers, a lot more partners.

Michel: I like how you already use the word 'simpler'. You redistribute the responsibilities and actually everything becomes simpler. The way we were doing things, Server-Side Ad Insertion (SSAI), Client-Side Ad Insertion (CSAI), and making the hybrid of it right now actually makes it very simple to operate but also to manage ecosystem, I would say.

Michel: From a product perspective - how is Google Ad Manager supporting Server-Guided Ad Insertion (SGAI) or getting ready for this new innovation.

Sourya: Yeah, so one of our biggest goal in Google is - how we make the ad insertion experience as easy as possible? How do we make the ad insertion the best and easy possible, but at the same time, how do we maximize our revenue? Of course, we want to make money. So make money, and at the same time, have a great watch experience. Have a great implementation experience. So we will have customers who want Service-Side Ad Insertion (SSAI) and for that, we have open APIs, we partner with many SSAI providers, see if they like to do that, we have a full suite of products. But now what we are hearing is that a lot of our customers are testing out Server-Guided Ad Insertion (SGAI) and they're interested in Server-Guided Ad Insertion (SGAI). They're doing POCs and so for that what we have done is we have launched a lot of documentation, a lot of samples, lot of guides, as well as we're partnering with player partners like THEO and others, to make sure we're able to build something which is super straightforward, which people can use from the get-go. One thing I do want to mention, this is super early stage, we're still iterating out, we're still testing things out, as we worked with THEO closely here. But our goal is to make it as easy as possible and at the same time provide some really good monetization experience.

Michel: Very cool! A question that I always wanted to ask you, Lemu, because we ventured on this path a bit naturally - started working together at some brainstorm sessions, I still believe, thought this could work, let's try it out. I always wondered - how do you select partners? Or what do you value in the partnerships? And why, for example, choose us to together built this journey?

Lemu: Absolutely! So when I think about partners that are good to work with, I think I'm looking for partners who have a bias towards action, who are ready to develop POCs and to put in the work, to be able to test these ideas. There're ton of ideas that come up often, but we have partners that come down and they won't say it, but they're looking for more of a press release than the actual work. So if we have a partner that's ready to work, like you when came ready to work, saying - hey, we'll test this, we had an idea. You tested it two days later, that type of feedback, I'll send that to Sourya and the rest of the engineering team. Great! here's some additional feedback. And the quicker and faster we can have that back and forth, for us, that means - all right, they are ready to test with us. So a bias towards action is one way we look at partners who come to us who want to do something and say, all right, how quick can we get this done? In fairness, working with a large company like Google, things can move slowly on our end. But when we are ready to go, having someone ready to go with us is how we value folks.

Michel: I must say the first thing that starting to work with a big company, can be very daunting. But in the relationship we had, it was amazing, it's like a garage build up of the technology.

Lemu: It's exciting ideas with a team that was ready to work. So that's when Sourya and his team said - THEO is an interesting company that's pushing us to do more work. How can we work together more closely for this idea and ideas going forward.

Michel: That, of course, results into something pretty cool that we can show at the THEO booth today with different ad formats, different ad experience: double box, and L-shape, it all becomes at the tip of your hands and becomes much easier. How do you see this shaping the future of monetization? How do you the impact of Server-Guided Ad Insertion (SGAI)?

Lemu: Yeah, I think you know right now for large publishers that are looking for more demand or looking for more fill this is one way that we believe they can support the monetization but also it's new inventory for them. Programmatically creating new ad formats, innovative ad formats that they can take to their advertisers and sell, is something that they're all excited about. I think whenever you create new innovative formats, and this is agnostic of Server-Guided Ad Insertion (SGAI), you also need to educate the buy side - the advertisers, and let them know: hey, this is now available to be able to purchase. So part of what we're doing is - as we build this out on the technical backend, is having education sessions with agency teams, advertisers and tell them - here's a new available unit that we believe can increase your revenue and increase your overall monetization for your organization. Because that's at the end of the day what they're looking to do. So not only do we work closely with the technical team, but the advertisers, to be able to build that demand to make sure that everyone is getting the fulfill the value of what this can do.

Michel: Super cool. It's very nice. And it really elevates also the experience. If you combine it, for example, with live low latency, you really get a super cool engaging experience. It's a broadcast-like experience you get, but even engaging with the content.

Lemu: The ability to engage with content is what a lot of large publishers are looking to do now, because that inherently builds more value than standard sit back watch experience. So adding QR codes or clickable units is something that is what people are asking, they want to know what's the best way to do that. And this is one of the avenues to be able to do that.

Michel: Showing more relevant ads to your viewer, engage with content, that's what it's all about.

Lemu: Exactly!

Michel: What would you advise media companies today when looking at Server-Guided Ad Insertion (SGAI)? What are the things they should consider, in which use cases should they look at it?

Lemu: I think as companies are thinking about this, I think it's a matter of how much are you willing to test and innovate, to play with these new ad formats, to allow a small sandbox to test. Understanding your current audiences: what would work best for them, whether it be an interesting L shaped format with QR codes, whether it be double box format. Knowing your audiences and saying - hey, this is a format that we think could work well for who we are, willing to test there. So it's not in that matter of preparation per se, but more of like - what are the innovative ways you're looking to add additional inventory effectively to your set context?

Michel: Super cool.

Sourya: And one thing I just do want to add here - keep in mind, it's a standard-based thing which has already been launched, but it's super new. People are testing it out, so whoever you're working with, whether it's Google, whether it's THEO, whether anybody else, just try different things and make sure your use case is covered, because right now the standards are being matured, so we would love to make sure the standard takes care of most of the use cases, because later it becomes much harder to change the standard. But now everybody's open - DASH, HLS, Google's, THEO, everybody is really looking forward as we like testing these things out. So it's a really good time to talk about a use case and see how this can work together.

Lemu: Exactly! With one-to-one relationships things are a little easier, but when you want to scale programmatically, that's when we need to start looking at what the standards can be so that we can scale.

Michel: Yeah, absolutely! And that's also, I think, the feedback that we're looking for right now, working with customers. We have a lot of use cases that we can already cover right now in production. But the ideas are endless and talking to everybody you get a new perspective, you get an image overlay combined with a video ad, you get the L shape, you get even highlights to be shown as an ad for example, so pretty cool stuff is possible.

Lemu: And then as I mentioned earlier then building the demand for that is the second part that we always need to remember that we are working with internal teams need to be able to educate the market on what this is so that they can sell it and we can just make it work.

Michel: And eventually set the benchmark for what it should be.

Lemu: You get it.

Michel: Okay cool. Maybe last question Sourya: from a product perspective what are the key considerations for using or looking at Server-Guided Ad Insertion (SGAI)?

Sourya: I think from a product perspective, the biggest thing to understand is - that it's a very new way of doing ad insertion but what this provides is that is unlocks a lot of new use cases, a lot of new formats: new things like skip ads on live, things like picture in picture, L banner, which you can now do dynamically. You can do everything server-side but in client-side it just makes it much easier. So we expect the user watch experience to be much better, because right now you don't have to get out of the content and see an ad but instead of that the content keeps on playing and you have another ad playing on the side. Second thing we expect it to be - you can easily implement and keep on iterating because since a lot of things are at the client side.

Also third thing, I would say like in full transparency, Server-Guided Ad Insertion (SGAI) is great but there are some old devices which might not be supported yet. So as you're thinking or as your partners are thinking about Server-Guided Ad Insertion (SGAI), of course keep Server-Side Ad Insertion (SSAI) as a backup because one thing what we have seen is - for most of the new devices you can put some amazing experience with Server-Guided Ad Insertion (SGAI), but for the older devices you still have Server-Side Ad Insertion (SSAI) as a backup.

Michel: Absolutely true! There is still a lot of fragmentation in different device markets and every device has its limitations, has its specific works. It's of course, something where we from the player side already have quite some experience in, so we are lucky to bring it already across a lot of platforms, but I do agree there are still some challenging things out there, and it's depending on your use case what is the best choice. Where do you do Server-Side Ad Insertion (SSAI), where you do Client-Side Ad Insertion (CSAI), where do you do the hybrid - Server-Guided Ad Insertion (SGAI)? And then certainly if you're looking for that next level of experience combined with low latency, live sports, for example, engage with that content, keep your viewer connected to your experience, that's where I would say is the ideal use case to go for Server-Guided Ad Insertion (SGAI).

Sourya: Absolutely!

Michel: I think we're already at the last slide. We call it the future of advertising because we really feel we have something super powerful at our fingertips without really realizing yet what can it all do and where will it end. So I think we're all very much looking forward to seeing what bright ideas we get.

Don't know if you guys have some amazing key ending words for this presentation.

Lemu: Experimentation! This is about experimenting on ways to bring additional value to the end publishers. And we're excited! My team specifically is out there working with third parties to build POCs on a variety of different use cases in innovative formats. But this is one we think can be deployed relatively quickly that we think we want to see people trying different things with so that we can learn how best to bring the demand, to bring additional value revenue to our partners. So experimentation, testing, innovation is what we're all here for.

Souya: I'll just say go ads. Ads empower free TV, a lot of different things. As long as we make the ad experience better, a lot more people will watch the content.

Michel: That's a very good closing statement. Go ads, go experience! That's what we should do. Okay, thanks everybody for visiting. It was lovely to present for you guys.

Lemu Coker

CTV Ecosystem Partnerships Manager
at Google Ad Manager

Michel Roofthooft

VP of Engineering
at THEO Technologies

Sourya Roy

Senior Product Manager
at Google Ad Manager

The future of viewer engagement and monetisation

Watch the recording of the IBC2024 talk where together with Google Ad Manager, we unveil the future of viewer engagement and monetisation

Webinar Transcript

Speakers

Want to deliver high-quality online video experiences to your viewers, efficiently?

Subscribe to news and product updates