The Drum Awards Festival - Extended Deadline

-d -h -min -sec

Artificial Intelligence Advertising

The AI ad tech revolution is on the way

By Jamie Evans-Parker, founder and CEO

October 7, 2016 | 7 min read

Apple chief executive Tim Cook is among the leading industry experts touting artificial intelligence as a tech on the cusp of world domination. Talking about the evolution of the smartphone, Cook told the Washington Post: "AI will make this product even more essential to you. It will become even a better assistant […] where you probably aren’t leaving home without it today – you’re really going to be connected to it in the future."


But how so – will we soon be permanently wired to the web? Not quite, but we will have something that is, and is set to be very smart indeed.

It all lies in AI’s prediction capabilities. As Cook went on to explain, AI will be weaved into Siri and Mail, enhancing their ability to anticipate a consumer’s next sentence. Siri, or ‘SuperSiri’ as it’s been dubbed, is set to become users’ very own virtual PA – and by opening it up to third-party developers, Apple could soon enable consumers to order things like takeaways and taxis just by speaking to Siri.

It sounds rather futuristic, but Apple is by no means the first to exploit AI tech in this way. Ad tech innovators have also been utilising AI’s ability to predict for some time; employing it to help shape first-rate ad campaigns.

So how exactly is AI refashioning advertising, and what does the future hold for this high-tech revolution?

Streamlined set-up and fast production

Just as ‘SuperSiri’ is destined to become a smart assistant for consumers, AI is set to become an advertiser’s ‘best friend’. AI’s capabilities are such that a creative agency could upload static design assets, including images and text, and AI will instantly read the target market – luxury or mainstream for instance – as well as the ad campaign vertical ie sport, automotive or travel, and animate the ad accordingly. It can identify what a campaign is about from a detail as minute as the shoes a model is wearing in an image – so if the model was wearing football boots, AI would recognise it as a sports brand, as opposed to a model wearing formal brogues, which would be likely to indicate a high-end fashion brand.

And in the same way that an iPhone knows a user’s typical pattern of behaviour and might send an update on relevant travel news at the end of a working day, AI can ‘learn’ how a particular brand should be presented through animation duration and style, making every subsequent campaign quicker and easier to develop.

But recognition and learning is only part of AI’s ability, it can also convert a brand’s static files into rich media ads for all screens in minutes – streamlining a production process that would previously have taken days or weeks to complete. Given the fast pace of the digital advertising world, anything that speeds up the method of campaign creation is viewed as highly beneficial and it’s therefore probable we will see many more brands producing AI-driven cross-screen campaigns going forwards.

Broader programmatic scope and flexibility

There is no question that programmatic has already done a lot for the efficiency of digital ad trading – replacing lengthy manual processes with algorithms that rapidly manage deals at scale. But now it’s time to start addressing the creative that’s used in the programmatic landscape, such a tendency to overlook valuable inventory. At present, inventory is traded based on deal IDs and identifying format sizes, so brands are often closed off from inventory that could provide a better experience for users and a greater return on their investment.

AI has the ability to detect where an ad is being served, from billboard to full-page. This element of the technology is very new but potentially huge for the industry, as instead of being limited by a variety of tags, buyers can access all inventory with a single tag. What’s more, by deploying AI tech they will also be able to identify the best ad to use, how it should be displayed, and set it to instantly upload.

Deeper data-mining and greater personalisation

Personalisation in advertising is key, but while the majority of Data Management Platforms (DMPs) can tailor ads based on the weather, location and time of day, AI can delve much deeper into individual user data. It has the ability to plug into a publisher’s data sets to gain unique insights into their specific audience, which it can break data down in terms of gender, age, profession and interests, and thereby pinpoint the best message to serve to an individual. Combining the best ad based on the publisher data set and layering that with dynamic data sets — time, location, weather – is the closest the industry’s come to bespoke advertising yet.

A complete and accurate view of performance

With its advanced algorithms and detailed reporting, AI can give advertisers a more precise picture of campaign performance than ever before – establishing both viewability and interaction metrics. For instance, if an advertiser buys multi-purpose ad units (MPUs) across mobile, tablet and desktop for a particular campaign, it is likely that the way each ad performs will differ due to the variations in the way content is viewed across screens. On mobile the average time an ad is in view might be four seconds, as users scroll through shorter pages at a faster rate, whereas on desktop – where content real estate is greater — an ad might be in view for an average of twelve seconds.

AI is not only able to collate these insights to determine overall ad performance, but can also automatically optimise delivery so that the full creative message is seen for the maximum length of time on any screen and, of course, reaches the right user. Moreover, this knowledge will also be stored to inform the construction of future campaigns, creating a constantly rising arc of engagement and impact.

In one key respect AI in ad tech is a mirror image of AI in consumer tech: a PA that makes life easier. It’s a virtual designer, an ad executive and an analyst all rolled into one all-powerful assistant that can build, serve, analyse, and optimise ad campaigns in progressively less time, while constantly acquiring ever more valuable knowledge.

Right now the AI ad revolution is in its infancy, but it won’t be long before its potential to improve production, expand programmatic, refine targeting and clarify performance sees it become more essential to those in the advertising industry than the trusty smartphone, even if it can order you a margherita with extra cheese.

Jamie Evans-Parker is founder and chief executive at wayve

Artificial Intelligence Advertising

More from Artificial Intelligence

View all


Industry insights

View all
Add your own content +