TOTAL BUSINESS VISIBILITY PREDICTIVE ANALYTICS IN RETAIL: The next stage of the retail analytics journey The next step in this journey is predictive analytics. INTRODUCTION Analytics is already a common feature of ecommerce, allowing online retailers the ability to target customers with special offers and advertised products, tailored directly towards them. The simple act of looking at a customer s past purchases and then presenting offers based on these insights creates a shopping experience that immediately feels more personal. For a long time, personalized interaction between retailer and customer seemed extremely difficult to achieve and inefficient for anyone beyond luxury retailers who could afford to dedicate the time, effort and resources required. Today, implementing a similar strategy on the high-street for personalizing the shopping experience for customers is an increasing opportunity for high-street retailers. The advances in technology mean that customers provide digital footprints and create valuable data which retailers can utilize to improve the shopping experience. The next step in this journey is predictive analytics: taking proactive action based on real-time data and predictions about future trends. This whitepaper presents current uses of predictive analytics in the retail industry, the potential benefits of utilizing predictive analytics, and the expected impact on the future of retail. TODAY S PREDICTIVE ANALYTICS LANDSCAPE Many retailers are already collecting and analysing data, but this information is often only used in a reactionary capacity, such as correcting previous mistakes or taking action based on historical information which could be months or even years old. The next step is to take this analysis and use it in a predictive manner: one that can be used to optimize and streamline every step of the retail supply chain. This involves not just reacting to previous problems, but tackling potential issues before they occur. W H I T E P A P E R Written by David Ricketts and Carrie Morgan 2016 C24 Ltd For example, predictive data analysis can inform retailers of the exact supply and demand for certain products, which can then be used to inform stock purchases and even internal store marketing in real-time. This can be tailored based on behaviour within individual outlets, stores, and customer demographics. For example, if a clothing retailer can see that certain types, sizes, and styles of clothing are popular in certain store locations, they can ensure that there is enough of that particular stock in those stores to meet demand, while reducing the amount of other products ordered (1). This enables the retailer to operate a more lean operation ordering only what they need and reducing what they don t need.
Predictive analytics can also be used to reduce waste and optimize the supply chain. INCREASING EFFICIENCY IN STORE The American retailer, Stage, instead placed store managers directly against analytics software in order to determine the effectiveness of one party over the other in determining price markdowns. The results were surprisingly conclusive. Rather than relying on instinct, predictive analytics were able to utilize past customer purchase patterns in order to accurately predict when the optimal time would be to introduce a sale or price markdown in order to best engage with customers. Steve Hunter, the CIO of Stage who oversaw the test, commented that 90% of the time, analytics won (1). Hunter points out that most store managers often place sales at the end of a season, in order to get rid of excess stock. However, the analytics approach suggested introducing a sale earlier, just when the purchases of products were starting to lull, which proved far more effective (1). Predictive analytics can also be used to optimize the current usage of loyalty schemes in retail. Having been an accepted part of UK retail for over three decades (7), loyalty schemes are an effective way to acquire data on regular customers, and offer promotions. However, rather than simply offering blanket promotions based on mass analytics, as most companies currently do, predictive analytics can be used to make offers based on an individual loyalty scheme member s personal shopping history, with rewards and promotions based on more accurate forecasts of their purchasing demands (8). These kinds of tailored, personal offers are more likely to endear a customer to a brand, as it appears they are offering specifically what they want, making them more likely to become long term customers. Retailers, however, now have to balance this with a growing awareness within their consumer bases about how their data is tracked and utilised. Not all consumers are happy with their data being collected and the saturation of loyalty schemes within the market may actually initiate a change in how customer data is collected in the future. ENHANCING THE CUSTOMER EXPERIENCE At its most basic level, all of these improvements build to one greater whole: improving the customer experience. If a customer finds that stock to their tastes is available in a store, and receives recommendations that are made specific to their sense of style, then it all adds up to providing a heightened in-store experience. Improving in-store experiences has been identified as an essential way to ensure the future of high-street retail against the growing online trend in shopping. In a world where customers can order anything from electronics to groceries with just a few mouse clicks and have their products delivered the same day, highstreet retailers are under pressure to identify how they can continue to attract shoppers. As a means to this end, there have been active steps towards integrating digital and predictive analytics into the store experience by many top retailers. Several retailers are using data collected from in-store digital experiences to provide seemingly personal experiences to each customer, with a focus on building a long term relationship between brand and user (9). Building this form of relationship should be the ultimate goal of a retail brand and predictive analytics is increasingly becoming a useful tool for achieving this, as online and offline activities are becoming increasingly integrated. Suggestions and recommendations have been cited as key factors in enhancing shopping experiences (1), something that online retailers such as Amazon have utilized to great effect for many years. Simple things, such as remembering where a customer s last order was delivered to, or making suggested additions to a basket based on current purchases, are all intended to make the customer s shopping experience as simple and as straight-forward as possible. In short, it provides what customers want, when they need it - which is a service philosophy that can be applied to high-street retail through predictive analytics.
IMPROVING THE SUPPLY CHAIN The benefits of predicative analytics to customers are just the beginning, as there are many ways these tools can be applied to improve the end to end operations of a retail organisation. The shop floor is just one part of the retail system. Retail success is reliant upon an efficient supply chain and the more efficient distribution of stock and materials is made possible through predictive analytics, such as EAT s 14% reduction in waste food through better stock analysis (2), and Stage s demographic study ensuring that the right amount of clothing styles and sizes go to the correct stores (1). While this allows for consumer supply and demand to be met, it also provides an obvious improvement to profit margins: less time and resources are devoted towards dealing with waste, therefore cutting losses and increasing the potential for profit. Predictive analytics can have an impact at every stage of the retail supply chain. By making analyticsempowered predictions based on supply and demand, manufacturers can order the correct amount of raw materials to meet demand (5). During production, accurate visualisations of the factory floor and processes can be used to streamline production, increase efficiency, and enhance final product quality (5). The potential for real time tracking and monitoring of products, particularly perishables, can be used to cut losses on spoiled or damaged goods by identifying particular trouble spots when goods are in transit (4). Predictive analytics software can even be used to identify stages where the supply chain slows down or is impacted, making predictive suggestions that could target these trouble spots, leading to increased efficiency (5). This increased efficiency is only made possible by the rate at which new technology is brought to market. Steve Hunter from Stage has stated that the degree of data analysis needed for their retail requirements would previously require large amounts of staff and manpower, but can now be handled by just one fast system in a relatively short amount of time (1). The business intelligence tool, Qlik, has shown that it can calculate a day s costs for a supply chain to 99% accuracy in one day, whereas previously it could take between seven and ten days (10), allowing companies to be far more efficient in their responses to the data. The whole system of predictive analytics is based around increasing efficiency, and it is likely to become increasingly so: as predictive analytics and its associated systems become more efficient, supply chains will become more efficient, which will cut costs, and enhance the retail experience. This leads to an increase in sales, resulting in improved profit margins, which will increase demand for efficiency focused analytics activities in a mid to make more savings. The entire process is cyclical and likely to continue to develop at a rapid rate. This rapid pace requires organizations to be flexible enough to adapt, make changes, and take action once they have collected and collated the data, in order avoid being rich in insight and poor in action. Ensuring that online and offline services are integrated as a means of data collection will be an important step for high-street stores as retailers look for ways to link the entire shopping experience (8). For example, a customer could browse clothing on a mobile device, purchase it online, and then go into the store where their chosen product could be waiting for them at the register to pick up. On the infrastructure side, analytics software could be applied at every step of the process to deliver the greatest benefit, so as to ensure that each phase of a supply chain is running at optimal efficiency (10). Ensuring that online and offline services are integrated as a means of data collection will be an important step for highstreet stores as retailers.
Many retailers are also developing omnichannel strategies, which integrate offline, online, and mobile shopping experiences. IOT AND THE FUTURE OF RETAIL Future tech developments are likely to increase the opportunity for retail analytics within the retail sector. A major development is the increase in Internet of Things (IOT) devices across retail outlets. The Internet of Things refers to a network of devices communicating with a wireless connection. For instance, trackers and monitors could be harnessed for the transportation of produce along the production line; just one step in integrating IOT devices into the retail supply chain (3; 4). track in-store traffic patterns or to monitor which displays are looked at, allowing for highly targeted positioning of advertising, produce, and valuable insights needed for changing store layouts, to ultimately maximise sales (12; 13). While some of this technology appears to be from the realm of science-fiction, the market is already offering sophisticated IOT solutions. Sensor devices such as the Aurora sensor (14), allow for smarter and faster analysis of instore traffic, with the ability to exclude staff movement from its analysis. Digital billboards are already being offered by companies such as Offer Moments (15), whose digital signage displays personalized, location and demographic based advertising to customers who approach it. As the success of these tactics and devices becomes more widely known, demand for them will increase, which will in turn push development into even smarter and more efficient devices. On the shop floor, digital signage can be used to push content into the store based on location, trends and current events (3). Many retailers are also developing omni-channel strategies, which integrate offline, online, and mobile shopping experiences (11). This can involve the development of mobile phone apps to enhance the in-store experience with features like floor plans or integrated shopping lists. These apps could also interact with in-store beacons, which can inform staff about frequent visitors upon their arrival (3). Equipping staff with information such as purchasing habits and clothing tastes allows them to provide more effective recommendations and deliver a highly personalised retail experience. Based on the excellence of service, customers are then encouraged to return in the future, producing long term relationships which are essential for the success of retail brands (9). IOT sensors can even be used to
CONCLUSION The potential opportunities for predictive analysis in retail are huge. It can allow retailers to go beyond just reacting to historic internal data and instead utilize an extremely broad range of variables to develop responsive marketing and retail strategies. These variables can range from internal, customer-based and external data points, which predictive analytics software can amalgamate in order to tailor each store experience directly towards individual customers. Almost every step of the retail supply chain, from manufacturing to the shop floor, can be optimized and made more efficient by the integration of predictive analytics, leading to obvious improvements in a retailer s profit margins. Along with this, analytics can potentially have even greater long term effects; ones that are essential to ensuring the retailer s longevity in the market and brand loyalty amongst customers. Return customers are essential to high-street retail s future, which requires establishing a relationship between brand and customer. Retailers that are embracing analytics are gaining in-depth insights into their customers and the retail environment as a whole, data which would have previously been impossible to gain without current technology. These insights are being used to create a new form of retail: one that is efficient, smart, and conducive towards encouraging brand loyalty and better customer experience. REFERENCES AND FURTHER READING (1) Forbes - Big Data In Retail: How To Win With Predictive Analytics (2) Computer Weekly - EAT builds appetite for predictive analytics (3) Accenture - The Internet of Things: Revolutionizing the Retail Industry (4) CIO - 6 ways data is taking over retail (5) Data Science Central - Predictive Analytics in the Supply Chain (6) Business Solutions Big Data Analytics - Real World Benefits for Retailers (7) Guardian - How loyal to your reward cards are you? (8) Vend Retail Trends and Predictions 2016 (9) WARC - Design the retail experience (10) Logistics Viewpoints - Competing on Analytics: A Supply Chain Case Study (11) Computer Weekly - Are retailers using data analysis to their advantage? (12) WWD - POS Data: Retail s Biggest, Untapped Opportunity (13) BizShifts-Trends - War for Retail Shelf Space (14) Retail Next Aurora (15) Offer Moments ABOUT C24 C24 Ltd is one of the UK s leading privately owned specialist managed service and hosting providers, based in the Midlands, UK. Working with businesses all over the globe, the company manages, secures and delivers critical business applications to over 100 countries, with a particular focus on the legal sector. As a strategic partner to Practice Management ISVs, we deliver enterprise hosting platforms legal firms who are looking for more flexible solutions for their core practice management platforms. To discuss how C24 could help you to create actionable intelligence for your organisation, then please get in touch to arrange a demo session of your live data. TOTAL BUSINESS VISIBILITY T: 0121 550 4569 E: marketing@c24.co.uk Find us on: