An algorithmic approach to more personalised customer relationships

Personalisation in retail is not new – this is what retailers have been doing since the days of the little black book. However, machine learning and automation of marketing and loyalty communications are proving to be game changers in terms of delivering individual experiences, more efficiently.

Increasingly we are seeing brands collect customer data accrued in-store and online, to help with targeted promotional communications and offers that are not just more relevant to each individual consumer, but also delivered to them at a time when they are most likely to engage with them.

However, making manual or rules based decisions to deliver personalisation across a collection of data that covers a large number of products and a diverse customer base is not just complex, but it is time consuming and costly.

A more algorithmic approach is required in order to ensure retailers are considering the right data points and enabling cost effective personalisation that evolves and improves over time as feedback from customer interactions is collected.

This is where machine learning can truly help customers to feel as though their favoured brands understand them and are making efforts to fit into their lives, rather than appearing at unwanted times with irrelevant content. If retailers successfully offer their customers this tailored, one-on-one experience, they are more likely to drive engagement and longstanding devotion to their brands.

Although machine learning has been available to retail brands for a while now, it is surprising to see how few customers feel like they are getting a personal service. Our own research found that currently just 30% of UK shoppers get relevant recommendations for products and services, and only 31% said that they were rewarded with offers tailored to them.

It appears that retailers are still struggling to effectively implement the technology and algorithms that will help them to determine what a customer’s data is telling them about the way they shop, including what motivates them to purchase and which products they prefer or could be buying next.

Take the Christmas period as an example. Consumers not only make far more purchases than they would in any other given period of the year, but they also spend a lot of time looking for inspiration and ideas for what to buy.

The savviest retailers out there are already using advanced analytics to mine and analyse the data gathered from loyalty programme members during the increased Christmas footfall, so that they can learn from it over time and use it to make increasingly accurate predictions about what specific customers might be interested in. The personalised marketing experiences that they will deliver as a result will help them to build on Christmas custom and use it to drive longer-term loyalty and brand relationships that will last longer than the average New Year resolution.

Fast forward to next Christmas, and machine learning could be helping those retailers to understand how many presents a customer needs to buy, and who they are shopping for based on the ghosts of past Christmas purchases.

Each time a customer interacts with a brand and has a good experience a new standard is set, and machine learning represents the best possible chance to ensure that retailers keep meeting – and indeed – raising that bar with individual interactions and shopping experiences that make customers feel recognised and valued.

Ultimately, the goal for every retailer this year should be to explore how they can use machine learning and individualisation to build on the successes of busy sales periods, providing each customer with better rewards, content and customer experiences that will keep them coming back throughout 2018.

Jason De Winne, is the UK general manager at ICLP, who help clients on the journey from acquiring customers to transforming them into advocates.

Back to top button