In conversation with Gavin Buxton, Managing Director, Asia, Magnite
Over the past few years, the digital advertising industry has witnessed a significant shift towards automation and data-driven decision-making, spurred by the proliferation of AI and machine learning technologies. This shift has been driven by the growing complexity of digital advertising ecosystems, with advertisers and publishers grappling with an ever-expanding array of channels, formats, and audience segments. Against this backdrop, we interviewed Gavin Buxton, Managing Director, Aisa, Magnite, to understand the role of AI in reshaping advertising. Gavin highlighted AI’s impact on automation, efficiency, and personalization across various facets of the industry, from campaign optimization to content recommendation and forecasting
Agency Reporter- In your view, what are some of the key ways in which artificial intelligence is opening up opportunities for brands, publishers, and operational processes?
Gavin Buxton- The ad tech industry has leveraged AI for some time but over the past year, the interest around AI has grown considerably. While much of the excitement surrounding these developments in AI is related to generative AI, there has also been a focus on the benefits of machine learning, a subset of AI.
Machine learning can contribute to the automation of a range of processes, including media campaign optimization, personalized comms based on predicted behaviours, workflow automation, and intelligent audience curation and targeting. As ad spend continues to grow, leveraging automation where possible can free up resources and help enhance the overall ad planning and delivery process for brands while helping publishers manage inventory to improve yield. The result is greater efficiency across both sides of the aisle, ensuring publishers can deliver great ad experiences and brands drive performance improvements.
Agency Reporter- How has Magnite been leveraging artificial intelligence in its operations and offerings?
Gavin Buxton- One example of Magnite leveraging AI is that in 2023, we announced machine learning-driven recommendations for A/B testing within our Demand Manager product. This feature is an additional tool for publishers to manage their Prebid stack that uses machine learning to provide automated optimisation recommendations based on data. Initial tests showed that 80% of wrappers that ran a machine-generated experiment saw an increase in revenue compared to the existing setting.
A second example is Bingewatcher, a tool our clients can use that helps automate creative reviews within streaming environments. It scans video creatives using “computer vision” to evaluate whether the field that’s passed accurately describes the creative, so publishers can then set their guidelines for allowing the creative to run. Bingewatcher brings ease and efficiency to CTV and OTT work streams by providing an automated, scalable and flexible technological alternative to the manual process of creative review.
Agency Reporter- Can you discuss how AI helps sell-side platforms navigate the complexities of open ecosystems, such as data sharing, interoperability, and standardization?
Gavin Buxton- AI and machine learning, specifically, can help address challenges related to data sharing, interoperability, and standardization. Machine learning algorithms can analyze a large number of datasets from disparate sources to extract important insights without compromising data privacy. They can then help SSPs to consolidate this data to facilitate more collaborative decision-making and improve transparency.
Machine learning can also drive greater efficiency around interactions with various partners and platforms, helping to ensure consistency across different SSPs and data sources with regard to data formats. AI-powered governance frameworks can also enable SSPs to enforce industry standards, regulatory compliance, and best practices, fostering trust and collaboration within the ecosystem.
Agency Reporter- Could you elaborate on how AI-powered technologies facilitate more immersive and interactive experiences for users, enabling a deeper level of engagement with content?
Gavin Buxton- On streaming platforms specifically, content recommendation is a primary driver of viewer engagement and one of the ways this is executed is on the smart TV ‘home screen.’ The primary function of this interface is to help users quickly find content to watch, which is where machine learning comes into play and can be used to serve recommendations of new content based on what users have previously watched. For instance, if someone watches season 1 of a particular show, they’d expect season 2 of that show to be recommended. This practice has evolved into helping people find content that they may not have considered watching previously. These machine-driven recommendations can create a more immersive and interactive experience for users, which in turn fosters deeper levels of engagement and a better overall user experience.
Another example is ad podding. Ad breaks are dynamic opportunities filled with a variety of ads of different lengths (15, 30, 60, 90 seconds). These blocks of time have to be considered from a yield management perspective, and price floors altered based on the length of the ads to ensure the pod is properly filled and avoid running house ads or slate. Machines are well placed to create the perfect ad break as they can take into consideration demand diversity (no duplication, different ad lengths, genres, etc.) while delivering quality user experiences that yield better results.
Agency Reporter- Could you discuss the role of AI in predictive data and forecasting within the advertising ecosystem?
Gavin Buxton- Within advertising, machine learning can strengthen the accuracy of predictive data and forecasting. Machines are better equipped than humans at forecasting efficiently as they can take into consideration past viewership numbers to predict future viewership as well as the strength of the content to determine avails and pricing. A huge opportunity exists across the ecosystem, particularly in streaming, to tap into the power of machine learning across streaming services, applications, and home screens to help improve performance and yield as well as user experience.
Agency Reporter- How does AI enable more accurate forecasting and pacing of campaigns, particularly in the CTV landscape?
Gavin Buxton- AI has a unique ability to analyze vast amounts of data quickly and identify patterns, a task that wouldn’t be possible for a human to accomplish in a short timeframe. Algorithms powered by machine learning can ingest information and will learn and improve over time as they receive more data and feedback on campaign performance. This iterative process helps refine forecasting accuracy and pacing strategies.
Advertisers can effectively leverage AI to personalize and target ad campaigns to specific audiences to maximize efficiencies. Automated content recognition (ACR) data, which tracks viewing across opted-in smart TVs, can be a resource in addition to AI. ACR can help deliver a better understanding of audiences and allow for greater insights into reach and frequency across platforms to help inform advertising decisions.
Operationally, AI can automate the historically manual process of campaign setup and create more efficiency across campaign delivery elements, such as flighting, to ensure they’re pacing correctly towards the goal.
Agency Reporter- How does combining AI with streamlined offerings maximize value in terms of driving cost efficiencies and profitability?
Gavin Buxton- Combining AI with streamlined offerings provides an opportunity to improve ad monetization and efficiency in various ways. For example, forecasting inside of an ad server connected to an SSP allows Magnite to set, analyze and adjust the clear price in real-time with the goal of improved performance.
AI can enable real-time floor price setting that is automated, seamlessly making changes to positively impact yield across the time of day, day of week, and month of year, while also freeing up resources. AI tech can also be applied to video ad workflows and monetization in several ways, including creative review automation and audience curation.
Agency Reporter- What are some emerging trends or developments related to AI that you believe will shape the future of advertising?
Gavin Buxton- I believe AI’s role in the industry will become much more important over the next few years as machine learning, in particular, has the potential to revolutionize facets of the streaming TV landscape. Over the next year, we’ll see increased adoption of machine learning to automate things like creative review, ultimately contributing to an even more enriching and immersive streaming TV viewer experience.