Where Is Artificial Intelligence Taking Marketing?

Unless you’ve been hanging out under a rock for the past 10 years or so (and I can’t say I blame you if that is the case) you’ve probably heard an awful lot about artificial intelligence already. Unfortunately, as is often the case with any ‘hype’ technology a lot of that coverage generates more heat than light. Although we talk about artificial intelligence all the time, we don’t necessarily know what we mean by the term.

That’s doubly an issue for marketers. As bears of little brain, with some honorable exceptions we don’t really understand topics like these in the way a computer scientist might. To some extent our information is second-hand, and that only adds to the potential for confusion.

That’s a shame, because artificial intelligence will completely transform the way brands and consumers talk to each other. Note I don’t use the word ‘might’ or ‘may’. There is no doubt about what is about to happen, partly because it is already happening. So with that in mind, what follows is my own humble attempt to explain what artificial intelligence really is, and show how it will impact the world of marketing - with a focus on the short and medium term.

What Is Artificial Intelligence?

It probably goes without saying that it is hard to define the term, but it helps that we’re thinking of the concept in terms of its implications for marketing. On that basis, we might fairly describe artificial intelligence as ‘software with the ability to learn’ - a definition that also neatly covers the fact that many people use the phrase ‘machine learning’ interchangeably with artificial intelligence.

This definition is helpful in the sense that it removes from our minds visions of HAL 9000 or robots passing the Turing Test (although it may well be how we get there). Instead it allows us to think in concrete terms about how AI is applied or could be applied to the here and now. Let’s talk about what that means in marketing terms.

Knowing The Person, Not The “Audience”

Good marketing has always been about learning, but learning has been a painstaking process. Back in the good old days a TV ad would be created, media would be bought on appropriate channels at appropriate times, and we would commission expensive studies to establish whether they worked or not. We’d pore over what data we had and then make adjustments for next time around. But we would always think in terms of ‘segments’ or ‘audiences’.

In the world of direct marketing we’d send competing variants of our creative to randomised lists and study which had a greater impact. Then we’d adjust copy for the next time around. But there was never any sense that - even if we had the ability - we’d be able to learn at the individual level. So although marketers have always loved data, using it has been laborious, and data has tended to be consumed and acted on at an aggregate level.

Artificial intelligence changes all that. Today we are surrounded by more data than ever before - far too much for any one human or team of human’s to process or learn from. That data tells us almost everything about every one of our customers and prospects. Where they browse and for how long, what they respond to and when, how they behave across multiple channels - plus a huge raft of demographic and device data if we want it.

Artificial intelligence (or machine learning) enables us to use and act on that information. By learning from every action and interaction, AI-powered marketing is able to optimize in real-time and - most importantly - do so at the individual level.

What AI Marketing Will Look Like

At this point let me paint a picture of the future.

As a marketer, I arrive in work one morning and remember that I need to tell my customers that I have a flight sale starting tomorrow. I open up some form of marketing automation platform. I fill in a few fields to communicate the nature of the sale and any other pertinent information. And then I sit back.

What happens next is that what I wish to say, is communicated to those people who will probably be interested in it, in the most appropriate channel for them, at the right time, and with creative most likely to be effective.

In the background, artificial intelligence will make a number of key decisions, that I do not need to consider at all. These decisions will largely revolve around the behaviors of our customer, let’s call her Mary.

The AI engine within my chosen platform will know that Mary is one of a select few who is most likely to engage with communication relating to your offer, based on her previous usage and response data. It will also know exactly how Mary would most like to receive that communication - whether by mobile push, web push, email, etc. Not only that, but AI will figure out when Mary is most receptive to communication, and what she would like to see in it. For example, if she is an early riser, AI will know that a 6am push notification is the prime way to reach her, and what kind of creative copy she would like to see in it to optimize her engagement and response.

These granular pieces of data will create marketing campaigns that are truly millions segments-of-one, optimizing brand messaging for engagement, retention, and most importantly, revenue.

Hidden within my description above is one final point worth discussing: propensity. Marketers have long wished to have a sense for what any individual is about to do, or likely to do. That is the ‘decides who is likely to to want to receive a communication’ in the bullet points above. To date, however, it has been extremely hard to deliver reliable results in this area.

Artificial intelligence will change that. Handling vast datasets at the individual level will enable us to be more accurate than ever in anticipating needs and wants. That in turn gives us the ability to be there in the moment - without any human intervention - with the right message or offer, delivered in the right channel at the right time.

As a result we’ll see interactions that benefit both brand and consumer alike. New levels of relevance, more ‘helpful’ marketing than ever before, and greater loyalty and revenues as a result.

Artificial intelligence will empower marketers to ‘win the moment’ like never before.