How to turn retail customer data analytics into revenue

Based on wins from DFS, Boden, and other leading retailers

Today’s retailers have access to a massive amount of customer data – from structured data (like transactions) to unstructured (think social media reviews and testimonials). But here’s the thing: being able to collect data – even being able to learn from data – is not the same as getting value from data.  

When we helped furniture retailer DFS with its digital transformation (Digital Transformation at DFS | Beyond: PDTW) we didn't just analyse historic customer trends. We also developed a machine learning (ML) model – the DFS Growth Engine – that could use advanced analytics to identify which questions we needed to answer next. Which customer segments should DFS focus its marketing on? Where should they open another store?  

What we’ve learnt from success stories like this is that being decision-driven, not just data-driven, is what really makes the difference. Being data-driven can sometimes mean getting stuck in analysis or just looking backwards. The real value comes when you use data to make confident, forward-looking decisions that drive the business ahead.  

And, with people in the UK continuing to spend more cautiously in some areas, using customer data to guide smarter decisions is clearly a powerful way to drive long-term revenue.   

5 steps to becoming a decision-driven retailer   

From reducing customer churn with attrition modelling, to matching products to customer segments with propensity modelling, retailers can use customer data analytics to become truly proactive with their decision-making.  

Yet one of the biggest issues we see is retailers trying to use analytics tools and techniques without the proper data strategy, capabilities, or infrastructure. With this in mind, here are five steps that you shouldn’t ignore if you want to get more revenue from your data.

1. Create a data strategy   

To create the conditions needed for data-driven decision-making, you first need to create a data strategy. Effectively, this is your roadmap for using data to achieve your strategic business goals.   

You shouldn’t expect your data strategy to lead you down a single path. Of course, it should have specific use cases and business goals at the core, but it should ultimately be iterative and agile. It should allow you to experiment and keep learning from your data, so that you can start uncovering revenue opportunities that you didn’t know existed.  

To help you get started, we’ve developed a practical guide: Why Developing a Data Strategy is Mission-Critical. Here, you’ll find six steps to developing a data strategy that drives informed decision-making, now and into the future.  

2. Increase your organisational data literacy

Spreadsheets and gut instinct have their place – but when it comes to making sense of today’s vast and complex customer data, they just don’t cut it. Relying on manual analysis and experience alone eats up time and limits how much data you can realistically review. And that means missing the bigger picture – or worse, missing critical context.  

To be truly decision-driven, you need a team that a) trusts what the data is telling them and b) knows how to turn those insights into actions that move the needle on revenue. In short, you need an organisation with strong data literacy.  

According to Gartner, almost 30% of Chief Data and Analytics Officers (CDAOs) say ‘poor data literacy’ is one of the biggest roadblocks to success. If that rings true, now’s the time to invest in a data literacy programme that builds confidence and drives smarter decisions across the business.  

3. Create a single source of truth

We find that a lot of retailers have multiple data environments. This often means that not all teams have access to the same customer data and therefore aren’t working towards the same goals.  

For instance, due to the nature of their role, marketing teams often have customer data sets that other departments don’t. As a result, they often understand customer behaviours, price preferences, emotional trigger points, and so on, in ways that other teams don’t. Without this data, the product team, for example, can only rely on historical sales and trends to drive merchandising decisions.   

To overcome these data silos, consider creating shared key performance indicators (KPIs) across departments. You should also update any disconnected legacy technology that might be preventing teams from building a unified customer view. The right data infrastructure will help you to streamline data flows across your organisation, and scale up as your datasets and goals do. 

4. Prioritise the customers who present a genuine opportunity

A core part of becoming decision-driven is being able to understand which customer segments present real opportunities for long-term revenue growth and employing the right strategies to seize them.  

This looks different across every retailer, depending on how, where, when, and why your customers shop. Take these two examples from our greatest hits list. 

For fashion retailer Boden, whose customers shop frequently and engage across multiple touchpoints, we identified that reducing churn would have a direct impact on their bottom line. By using machine learning to segment at-risk customers and guide tailored retention strategies, we helped drive a revenue increase of £9.6 million. 

Customer data highlighted a different opportunity for optician retailer OPSM, however.

With customers shopping for eyewear less frequently than clothing, we decided to focus on delivering highly personalised communications to OPSM’s highest value customers. This saw a 17.6% increase in revenue from the targeted groups.  

5. Choose the right metrics to assess your strategies

It’s easy to track metrics that make things look good on the surface – so-called “vanity metrics” – but they don’t always lead to better decisions.  

Take conversion rates, for example. An increase might seem like progress, but what if those conversions came from customers who were already on the verge of buying? A more meaningful indicator might be the number of new prospects you've successfully moved closer to purchase. That’s what builds a strong future pipeline.  

When evaluating your data and AI strategies, focus on metrics that reflect real movement toward your goals – the ones that influence decisions, not just tell a flattering story.  

Start becoming decision-driven with Beyond: Putting Data to Work

Becoming decision-driven requires a whole organisation transformation that won’t happen overnight. However, the payoffs can be huge – as our success stories show.   

These five steps will help you get started, but we’d also recommend taking our Data Assessment. This short questionnaire will help you identify whether you have the culture, leadership alignment, and operational readiness to make strategic decisions and unlock revenue.