KHWS Blog: Marketing and the machine learning revolution

Marketing and the machine learning revolution


What is machine learning?

Machine learning is an AI technique that allows computers to take large amounts of data, apply complex algorithms and identify patterns to predict outputs, without being explicitly programmed. It uses a set of features from the data to either predict a specific outcome or group similar trends. Crucially, as the trends change the algorithm learns and adapts to these variations accordingly.

Where is it most prevalent?

We see machine learning in many areas such as autonomous driving, speech recognition, medicine, finance and marketing. In marketing, it’s used to identify customer preferences and tastes to power online personalisation. This comes in the forms of websites, email campaigns, product recommendations, to name a few. Machine learning has greatly improved over the past decade due to the large amount of accumulated user data and we now have the tools to process them. In short, machine learning has become mainstream, and is here to stay.

How do brands use machine learning?

Facebook uses machine learning to personalise news feeds based on user interaction. If a user stops scrolling to read or like a friend’s post, or interacts with a particular advertisement, then their feed will slowly customise to show similar stories and adverts. Netflix, Amazon Prime and Spotify use machine learning to continually improve movie and music suggestions based on their personal browsed and played history as well as similar profiles, allowing users to make the most of their subscription.

Similarly, Under Armour and Nike use machine learning to personalise their websites and apps to recommend the latest products based on previous purchases and searches. If a user tends to search for a specific fitness regime, or a product tailored to a sport or activity, these sites will customise the user’s online experience to that focus. By being able to predict what a customer wants to purchase, we can create a richer and more user friendly interface to personalise their shopping journey with relevant content. And the more data we collect, the smarter the algorithm becomes.

The opportunity for marketers

A survey by Demandbase and Wakefield Research on marketing companies with 250+ employees revealed that 80% of marketing executives believe that AI and machine learning will revolutionise marketing, but only 26% of those are confident in their understanding. In a world where machine learning becomes the norm, how does one get ahead in such a competitive market?

Perhaps the real opportunity is in adapting these algorithms to emotionally connect with the customer. As we leave a larger online footprint not only can we use machine learning to predict what a customer will want to buy, but we can also predict the customers’ emotional and behavioural triggers, allowing us to share relevant marketing content at the right time, to help drive purchase.