The retail industry is on the whole a voracious adopter of modern data-centric technology, aping the manufacturing world by using big data analytics to streamline supply chains, and using smartphone apps and wireless beacons to harvest customer data to deliver better service.
However, Robert Hetu, retail research director at analyst house Gartner, noted that, despite having the technology to collect and access large amounts of digital data, retailers fail to put it to effective use.
"With the rise of digital media came explosive growth in data volumes, and inexpensive storage encouraged retention of data with no present value because of the hope that it would be fruitful in the future," he said.
"Today, retail BI [business intelligence] analysts are so busy gathering information, often in reaction to an opportunity or challenge perceived by the business, that there is no time to analyse the data and glean useful information."
Hetu explained that this situation has led to a disconnect between analysing data and putting it to use in real time to quickly benefit a business.
Luca Bonacina, a retail research analyst at IDC, explained that the retail world has the opportunity to adopt machine learning to improve big data use.
"The retail industry is well positioned to take advantage of machine learning developments as very large volumes of data (structured or unstructured) are being created every second. There is a need to understand the hidden patterns in that data to make the most use of it," he said.
Machine learning systems can analyse data automatically and in real time to present recommendations to retail workers, or take action based on the results of such analysis.
Such systems then allow insights to be gleaned in real time from data being collected across a retailer's entire supply chain and business, meaning that action can be taken to maximise selling opportunities or ensure that a busy shop floor does not run out of stock.
For example, data collected from a customer's smartphone through an iBeacon in a retail outlet or shopping centre can be linked to sales and stock data and analysed by machine learning software. By rapidly crunching the data the system can push a discount code or send alerts to a customer's mobile device specifically tailored to their tastes, location or buying habits.
Machine learning algorithms, cloud computing and multiple data sources can further expand such a system, enabling customers to be presented with offers based on weather data from the Met Office, for example, or the latest topic of discussion on Twitter.
Such systems can, in effect, make strategic decisions well ahead of even the smartest of BI analysts, which in turn can help retailers gain an edge over less technologically advanced rivals.
Furthermore, automating such processes can even bypass the need for in-store retail workers in some situations.
Some retailers run complex businesses that require human interaction, but machine learning can still support retail operations.
Enterprise software firm Blue Yonder uses cloud-powered machine learning software called Forward Pricing that analyses online sales data and dynamically balances product pricing so that it gains the most profit yet sells well.
This form of automated analytics bypasses the need for analysts to sift through data and adjust prices manually, and allows on-the-fly adjustments that humans simply cannot deliver to the same effect.
Things get particularly interesting in the retail world when machine learning and data analytics are applied to the IoT.
Adding a layer of smart systems to a network of sensors, beacons and automated machinery allows more data to be collected and worked with in ways than can define a retailer's entire business model, not just streamline its operations.
This level of automation can be taken by retailers and expanded across their entire supply chain, moving from automated factories to delivery vehicles with telematics systems feeding back location data to a central system.
After a completed delivery, feedback from a customer can be gathered from a company's website or social media page and, when analysed with sales, supply and manufacturing data, can be used to influence the development and provision of new products.
In effect, such technologies create a situation where data is collected across a retailer's entire operation in a continuous cycle that helps optimise everyday work and capitalises on opportunities to boost revenue. This level of automation and insight also aids the working lives of retail staff from the headquarters to the shop floor.
The gamut of data-driven and cloud-powered technology is now available for the retail sector to take advantage of, and some retailers are already doing so, creating a perfect storm of innovation and real-world use cases. Others should follow this lead or risk falling behind in the market.