The retail sector was the first to recognise that data is the most valuable resource available to businesses today. Tesco’s Clubcard in the 1990s, which gathered data about its members’ shopping habits, paved the way for a flood of rival loyalty cards – and eventually e-commerce sites – whose aim was to understand shoppers’ needs and to anticipate them. As a result, today’s retailers are in possession of vast quantities of data; from transactions, gift cards, or online loyalty schemes, but still have no idea what to do with those customer insights – everyone thinks they have a plan but the reality is, it’s not enough. To simply capture the nature of a transaction – a sale, a call, an upgrade etc. does not tell the whole story. We need to understand the path that led to this transaction and the behavioural responses that caused it. Without the appropriate tools to see correlations, retailers are simply hoarding data, like old antiques rusting away in the attic.
The insight retailers need
More retailers than ever are beginning to think of themselves as data or tech-enabled businesses, armed with huge data resources that can help in all areas of their organisation – from personalised marketing to optimising product discounts. The business need, however, is to understand the entire customer lifecycle, connecting disparate points in order to comprehend the shopper’s route to transaction. For example, if a customer buys a watch, have they first conducted their own research online? Have they sought reviews on social media? However, so far not one system exists that can identify all this customer activity. And the analysis of the data that we possess relies on a mixture of software and human intuition – which is simply not up to the task of coping with these advanced correlations.
The realm of data analytics is massive and multi-faceted and hiring more analysts won’t solve the problem. The way that the industry is beginning to address the challenge is to shift from user-initiated analytics to machine-initiated analytics. A machine’s capacity for data crunching will revolutionise the day-to-day life of category and store managers, releasing them from entire days of analysis every week. Black box AI models and machine learning systems can provide intelligence in real-time, allowing for more time to be dedicated to execution of strategy, rather than the discovery of problems.
A giant leap
It’s not all smooth sailing, however. To get the results retailers want requires a cultural shift in the organisation, prioritising data extraction and analytics as a pillar of company strategy. It’s a question of seeing the bigger picture; this investment is for longer-term ROI, meaning that there will probably be limited returns in the immediate future. Understandably, there is significant hesitancy among many business leaders; it takes considerable vision to understand that this is the future. For retailers who are able to take that leap of faith, however, the path becomes clearer. With the right tools, decision-makers can create a roadmap with short-medium goals in mind, which will help to make the project accessible to the whole company.
The good news
Change is on the horizon as digital transformation continues across the retail industry. This is reflected by the rapid growth of the analytics market – IDC forecasts that revenues from big data and data analytics (BDA) solutions will reach almost £147 billion this year, an increase of 12% from 2018. The more data there is, the more computing power will be required. Once retailers have this computing power in place, AI can do much of the heavy lifting, as well as the analysis. Companies that reach this level of data analysis – known as ubiquitous AI – will be the biggest disruptors of business. Stalwarts like Amazon can move into an infinite number of markets because they are at the forefront of technological innovation. Every retailer is a technology company now and using data wisely will revolutionise the shopping experience.
Credit: By Sy Fahimi, SVP Product Strategy, Symphony RetailAI.