We live in an increasingly automated world. Every month, machines take on more optimisation decisions once made by humans. This transformation has many people—marketers and consumers alike—wondering whether we’ll reach a point where we simply set the machines free to do their job. Given that many see this transformation as inevitable, what will the role of the digital marketer be in five years’ time?
In truth, many machines are already being set free. Recently we’ve seen major breakthroughs in speech recognition from systems like Amazon’s Alexa voice service. We’ve seen rapid growth in the power of machine translation in systems like Google Translate. And technologies like automated image recognition—not to mention self-driving cars—wouldn’t be possible without major advances in the way computers see the world.
Learning from feedback
What all these technologies have in common is that they’re applications of artificial intelligence (AI)—more specifically, neural networks and deep learning. We’ve had these technologies at least since the 1990s, but computer scientists have finally harnessed both the vast computational power and the enormous storehouses of data—images, video, audio, and text files strewn across the Internet—that, it turns out, are essential to making neural nets work well.
But neural nets, like many modern AI technologies, don’t work in a vacuum. As projects like Google Translate and self-driving cars have shown, AI can only do its job effectively if it receives ongoing feedback from the real world.
In 2007, many high street banks in the United Kingdom were using TouchClarity (a company I used to work for, now part of Adobe)—an automated behavioural targeting platform—to determine which products to show to which customers. The results of machine learning at the time indicated that a hero banner, along with three smaller creative slots, was the best way to use the homepage real estate, and most banks’ homepages really did end up looking like this fairly rapidly.
Over time though, banks realised that the homepage had to be seen as more than a sales opportunity; it’s an effective tool for signposting specific customer journeys. Banks arrived at this realisation through a creative process and by consulting with customers—not as the result of a machine learning exercise, which is why today their homepages look very different from that formulaic hero banner-led approach.
Even so, machines are having an impact on optimisation in a number of ways.
Advances in analytics
Generative design uses computerised algorithms to explore entire solution sets. In other words, you state your goals and constraints, then allow the computer to generate designs and iterations for you that you might never have thought of—a kind of accelerated artificial evolution, as some have called it.
We often use the concept of generative design in digital marketing, under the names audience segmentation, cluster analysis, and predictive analytics. These technologies analyse data in an automated way to discover insights about our audiences more quickly, and even predict future behaviour, by understanding when and why a customer might get in touch. This computerised “foreknowledge” enables us to deliver enhanced personalisation and allocate our resources more efficiently.
Now that generative design can predict customer behaviour more accurately than human marketers can, should the marketing industry prepare for a massive wave of redundancy notices? After all, as the Governor of the Bank of England, Mark Carney famously said, “Every technological revolution mercilessly destroys jobs well before the new ones emerge.”
The truth, as usual, is a bit more nuanced and complex.
Transformations in marketing
Marketers’ jobs will indeed undergo massive changes over the next few years, but it’s not all gloom and doom. The changes will, on the whole, be positive.
Over the last few years, we’ve seen marketing become more accountable for a broader range of disciplines, including direct revenue contribution, and, in many cases, even customer experience. This set of roles, skills, and requirements is too big for us to master with manual processes, which means we need to think about how machines can help.
We need to start by fostering conversations about how we can bring data science into our marketing activities, and, ideally, how we can work with data scientists within our own organisations to provide more of that all-important real-world feedback to the machines that are becoming central to our work.
But even more than this, as Econsultancy’s CEO has previously explained, the marketing industry needs to start replacing its “T‑shaped people”—those with a specific, deep expertise—with “pi-shaped people,” who have a broader knowledge base and possess skills that span both left and right brain disciplines. In other words, we need marketers who can be analytical and data driven, while also understanding brand image, storytelling, and experiential marketing.
As marketing undergoes these shifts in skills and thinking, our relationship with machines will also change. In many cases, we’ll move from being merely operators of digital marketing tools to being curators—choosing the best possible solution and working alongside the computer to co-create the ideal design. And when machines advance to the point that they’re offering their own opinions, which, strange as it sounds, may be as little as five years away, we’ll move into the role of mentors, providing input in the form of human-defined goals, values, and parameters.
In short, the idea is not to replace people with robots, but to “remove the robot from the person,” as Aviva’s CFO, Tom Stoddard superbly expressed it in a recent interview. In other words, AI will help free marketers from mechanical tasks and empower them to use their creativity to tell powerful, creative stories about their brands. When we set the machines free to do their jobs, we’re likely to find that marketers, too, are more free to do theirs.