Want smarter insights, faster? 5 AI rules for marketers

Marketers are drowning in data but starved of insight. AI offers a way to spot patterns, predict behaviour, and surface answers fast. But speed is meaningless without direction.

Here are five tips on how to make the most of the potential of AI, and how marketers can position themselves to take advantage:

1 Understand the potential
With such a wide range of possible applications, marketers must know what they want to achieve and where AI can help.

“Essentially, AI is trying to observe patterns and predict future behaviour,” says Patrick Affleck, CEO, Havas UK & Ireland. “In theory, you can start to build interesting customer segments and play out how they might respond to different stimuli in the future based on this customer profile. It won’t replace planners and buyers, because we still need to work alongside it. But it accelerates the ability to extract meaningful insights quicker.”

He adds that awareness of potential pitfalls – such as built-in bias – is just as important. “If we don’t organise our data sets appropriately and AI tries to interpret that, it could be completely wrong. There are always natural biases in a customer base, so if companies try to grow their customers into new areas, that will be more difficult. It’s for us as marketers to understand the right questions to ask of it.”

2 Identify your priorities
With a growing number of providers and solutions, marketers must focus on specific challenges. “There is more and more information available, but is it really bringing something new to the table?” asks Mathias Chaillou, chief media officer at L’Oréal. “Are we using it to solve questions we have, or is it just good to know more information? The question is always, ‘What purpose does it serve?’”

When reviewing providers, marketers must also stay focused on what they’re trying to solve, says Affleck. “Is there something in there that might genuinely solve a problem or a challenge that will enable us to switch our attention to do other things that can move our business forward?” he asks. “That requires measurement frameworks and approaches to validate and assess it properly, rather than just getting caught up in the hype.”

3 Build solid data foundations
Lara Koenig, VP global strategy and partnerships at programmatic media partner MiQ, says the technical starting point is building clean, connected, and harmonised data assets.

“There are technologies such as MiQ Sigma, which can solve specific challenges from planning and activation to automation of bidding, efficiencies and then reporting. But there’s also technology that is coming from data warehouses, which can help partnership managers get valuable insights out of data lakes in minutes, which would normally take an analytics department two weeks.”

The challenge for marketers is making that information relevant to their own business, she adds.

Currently, there’s a fragmented picture when it comes to the usefulness of data that organisations hold, says Annette Male, CEO of Dentsu UK & Ireland. “We did a report recently from Merkle, our tech and data brand,” she says. “One of the points that came out of that was that 51% of the brands we spoke to said that their data is rich and accessible. But that also means that 49% may have a lot of data that isn’t accessible. Whether they have the tools and technology in place that helps them organise that is a different question.”

4 Strike the right balance
Finding the right extent when it comes to using AI is a delicate balance. “We’re very happy to leverage AI; it’s much more efficient and delivers better results,” says Chaillou, “but what keeps me awake at night is how do we leverage AI solutions while still giving them a sense of direction? We need to make sure that whatever AI solution our team uses, they can direct it strategically in terms of what’s right for our organisation.”

Sarah Treliving, chief data, digital & technology officer, at Goodstuff, stresses that AI doesn’t remove the need for marketers to develop strategy. “People still have to create new output from all the new insight with the new time everyone’s been given,” she points out. “AI is just a curator of things that humans created. If humans stop creating things, AI is going to have nothing to refer to.”

5 Make your use of AI stand out
Koenig says that once marketers have the tools in place, they need to apply AI creatively and purposefully.

“That means forcing AI to make your brand stand out because otherwise it’s going to recommend the same thing to you that it does your competitor, and you’re all racing to the same price point and the same consumer with the same message,” she says. “Forcing originality and creativity into AI is the new role for agency planners and buyers.”

That means asking tough questions of providers. “Ask if they are talking about machine learning or just predictive modelling,” she says. “Are they talking about generative AI? The goal is to work with partners to understand what they have built from scratch, what they're repackaging and how it applies to their business.”

That means building close relationships with providers, says Zuzanna Gierlinska, CCO, UK, at MiQ. “A critical takeaway is to remember what the business problem that you’re solving is and, as partners, how are we helping you and ensuring that we really deeply understand that,” she says. “We all need to think about how this is actually going to solve the challenges you face.”