John has walked into his local sports shop with his son, Billy. He’s a regular, has one of your loyalty cards, and you recognise him. Billy is looking for some fancy new football boots for the new season.

But you don’t know that.

They start looking at the football boots in the kids’ section. Billy likes the new blue Nike boots, because that’s what his favourite player wears.

But you don’t know that.

You wander over and suggest the new black and orange Nike, because there’s a stock surplus and you’ve been asked to push them, and you also remember that John has bought orange football boots before. Billy hates orange.

But you don’t know that.

You suggest the blue Adidas that are on offer. Billy doesn’t like Adidas boots, because they’re quite narrow and hurt his feet.

But you don’t know that.

Imagine this scenario on your website. In this day and age, when we believe we have so much customer information at our disposal to make accurate, personalised recommendations to encourage a sale, do we ever really know what’s going on in someone else’s mind?

Do we ever really know what a potential buyer is thinking and what they actually want?

No, we don’t.

Sure, you can infer what they might want from what they look at, or even what they’ve clicked on to arrive at your website, but you can never really tell why they’ve come to you today and what they really want.

Scientists call this the cold start problem - you cannot draw any useful inferences about someone before gathering some information about them.

Personalisation today is inference based on data from previous purchases, previous visits, ongoing campaigns, pages landed on and persona paths, and it requires you to tag content and pages, and manually set up a myriad of rules accordingly. But this still doesn’t help you know exactly what your customer wants and why he’s come to your website today.

But what if you could remove the need for all this manual work, remove the guesswork around personalisation, and make recommendations more accurate?

Marketers need to start holding on to their view of the customer lightly, and be able to quickly adapt to their need in the moment – whether they’ve bought from you before, whether they have an account with you, or whether they are indeed visiting for the first time.

Machine learning can help you to understand your visitor in the moment, based on your existing data about that visitor and other visitors, to contextually personalise everyone’s experience at each visit.

Knowing your customer’s wishes better is possible, but it needs marketers to get closer to technology, bring data scientists on-board, and realise they need to personalise for the unknown.

Want to know more?

We’ve been working with the leading technologies in the personalisation and machine learning space for a long time.

Whether you want advice on how to deliver personalisation to your customers online, help to build a personalisation roadmap, or you're interested in a proof of concept for AI / machine learning to make a significant digital transformation step-change in your organisation, then we’d love to hear from you.

To talk personalisation and machine learning, just get in touch.