One of the big strengths of AI is its ability to translate between different mediums. You can feed any of the large LLMs a written description, and it can generate an image for you. You can upload a video file and it can create a soundtrack.
It makes sense, then, that there’s plenty of excitement in the advertising industry around AI’s ability to break down barriers between the various channels a marketer chooses for a campaign. Instead of having to handle each different medium in silo, AI can read across an entire campaign, compare results, and optimise accordingly.
That’s the premise behind COOL AI, an AI-powered omnichannel advertising platform run by US ad tech umbrella group The COOL Company. And CEO Zach Dugow says that using AI to break down media silos changes the way agencies and advertisers plan their media buys. “Instead of saying ‘I’m going to put X budget into search, X budget into social, and X budget into CTV’, you can have all channels competing against the same goal, and the AI makes budget adjustments based on what’s performing best,” he told VideoWeek.
Doing the work agency teams can’t (or don’t want to) do
The COOL Company itself is an amalgamation of various buy-side and sell-side tech businesses brought together over the past few years: publisher engagement and monetisation tool Insticator, media and attribution solution COOL Media, franchise and multilocation marketing company Balihoo, supply-side platform OKO Digital and dynamic creative optimisation specialist ADventori.
These various technologies feed into COOL AI, which covers all stages of a digital ad campaign. “With two clicks, we build the ad creative utilising AI, which means it’s constantly testing and updating, and we buy all of the media across every channel,” said Dugow. “It does all the attribution as well, which means there’s a flywheel effect across campaigns.” At present, COOL AI operates across search, social, and programmatic thanks to close integrations with Google, Meta, and various programmatic partners, and CTV will be added in the near future.
But the aim of the platform isn’t a Mark Zuckerberg-style “give us your objective and your credit card and we’ll do the rest” takeover. Rather, it handles some of the most manual, time-consuming work which an agency or brand advertiser either can’t, or doesn’t really want to, do themselves. “It’s really a tool meant to empower humans but take away a vast majority of the work and supercharge them,” said Dugow.
Charlie Raper, UK regional sales director for The COOL Company, gave an example of an agency running campaigns across Europe for a large multinational auto brand. Not all of those markets are equally valuable, and some of the smaller, unprofitable markets can end up taking up a disproportionate amount of time and energy. That agency could choose to use COOL AI to run those less important markets, freeing up resource where it’s most needed. Alternatively, an indie agency could use AI to run retention campaigns for their largest clients, which may have less budget compared to acquisition campaigns but still take up a lot of staff time.
Sometimes, it’s less about reallocating resources, and more about doing work which agency workers simply aren’t able to. “A starting point for clients has primarily been optimisation across search, because of AI’s ability to make such a high volume of changes in all hours of the day,” said Raper.
In the same vein, The COOL Company is seeing agency teams using AI tools like COOL AI to expand beyond their traditional capabilities, breaking down some of the silos which exist within agency structures. “We met with an independent agency recently which has historically used that siloed model, with a search team, a social team, a programmatic team, and then account planning teams and AV planning teams,” said Raper. “That structure is changing. Now they’re siloing the whole workflow, especially in AV. They’re planning everything from a video perspective, and integrating programmatic traders into those AV teams to get visibility across top-of-funnel activity. Tools like COOL AI are helping to break down those agency silos.”
Finding CTV’s place on performance plans
Of course, integrating different channels — each with their own strengths and measurement methodologies — into one universal AI tool isn’t straightforward. There’s nuance in comparing performance between CTV and social, for example. Can an AI tool really take this into account, and optimise a campaign accordingly? Or does CTV risk losing out as AI tools allocate budget based on KPIs designed for (and sometimes by) the tech giants?
Avoiding those pitfalls is part of the job of the agency. “Ultimately it comes down to what the client and the agency are using for an attribution model and as their source of truth for the end conversion,” said Raper. “The nature of the tool is that it depends on the objective that you set. If you’re looking for reach, for video completion rates, or an ABC1 UK demographic, then CTV is going to rank higher than display. And with COOL AI you can have multiple campaigns running at the same time, for example, one from an awareness perspective and one from a performance perspective. The distribution of spend will then skew between various platforms, and you can either set your own percentages that you feel are accurate, or let the AI optimise over time.”
That’s not to say that CTV will only receive budget for awareness campaigns. Attribution is a big part of The COOL Company’s CTV offering, and Raper says its models are able to demonstrate CTV’s effectiveness when it comes to performance. “With our ability to then place pixels onto the client’s container tags to then run across premium supply transparently, we can then show on a performance plan how CTV contributed by driving that initial exposure. We can show how CTV plays its part, and I think that’s what the industry is really looking for.”
“I think that’s one of the key reasons people like our dashboard,” added Dugow. “Our reporting is really clear on how a conversion happened, and how CTV impacted your other channels. We can show that a user saw a CTV ad on Wednesday, then on Friday they went to Google Search and bought the product. Breaking out what happened and being clear about where you’re giving credit for a conversion is really important.”
Demonstrating CTV’s ability to deliver on performance KPIs isn’t the only area where The COOL Company sees room for growth. Further progress is still needed on show-level transparency for CTV buys, a frequent headache for agencies. “You can’t really get the breakdown of programmes at the level that clients are looking for,” said Raper. “A lot of the AVOD, FAST and longtail pushers don’t allow programme-level information to be passed through the SSPs because it protects their USP. If you go to them direct and do a nice chunky partnership, they’ll give you those programme-level reports, but if you’re allocating a percentage through an SSP to their platform, they won’t share.”
This lack of transparency may be feeding into continued buy-side uncertainty around some of the less well-known CTV platforms. The COOL Company buckets the CTV world into different tiers and types of inventory, so clients can choose which sorts of sellers they’re comfortable with. But some of the smaller players still have a job on their hands winning over the buy-side.
“There are mid-tail companies which have great inventory and plenty of inventory, but buyers just aren’t on board with them yet,” said Raper. “It’s just an education piece, and that’s part of the job, getting brands and advertisers comfortable with CTV’s mid-tail.”
Follow VideoWeek on LinkedIn.
