How can retailers compete against Amazon if they can’t compete on price?
Customer experience can easily outperform price. More retailers are focused on localisation – optimising their supply chains to answer customer demand. True customer service moves beyond friendly clerks. It’s offering customers the products they want, when they want them, and where they want them.
Can you compare the differences between predictive analytics, prescriptive analytics, and machine learning? How can these terms tossed about help a retailer increase profit in its omni-retail supply chain?
Descriptive analytics are the most commonly used form of analytics, typified by dashboards and reports that summarise data and perform simple calculations to provide historical insights. In retail inventory management, the focus is on tracking key performance indicators (KPIs) such as in-stock rate, turns, and GMROII (Gross Margin Return on Inventory Investment).
Diagnostic analytics go a step further to explain why something happened. Business Intelligence (BI) tools provide the ability to visualise and compare different streams of data for this purpose; you can use them to lay up a timeline of weather patterns against lost sales on flashlights, for example, to explain why current store orders are so high.
Descriptive and diagnostic analytics are concerned with explaining what has already happened. Predictive analytics aim to tell us what will happen.
In retail inventory management, the most important use of predictive analytics is in demand forecasting. Traditional forecasting uses some variation of time-series analysis, in which past demand is used to predict future demand. Advanced forecasting uses machine learning (ML) techniques to mine the available data for additional information. That data is then used to adjust for factors not obvious from the demand data alone. Machine learning eliminates the need for merchants to suggest potential predictors in order to guide the forecasting process.
Prescriptive analytics answers, “what should I do?”, while machine learning techniques are revolutionising predictive analytics, retailers still need systems that can use those results to take action; that’s the job of prescriptive analytics.
Prescriptive analytics are based on optimisation algorithms, which provide an answer that will minimise or maximise the value of some variable. For retailers, these algorithms optimise the economics of the omni-retail supply chain. Such algorithms are being used today by some retailers to set retail prices and maximise revenue on markdowns.
Predictive analytics can tell a retailer to carry more or less of a certain SKU. How can analytics address products that are not currently in a retailer’s assortment that might be beneficial?
Predictive analytics considers all attributes of SKUs with high customer demand. For example, a tyres and auto centre client carried a premium brand of car wax. It was offered in various packages and sizes. Predictive analytics revealed sales would increase if the brand offered its product in paste form and was packaged differently. Once offered to customers, sales for that SKU skyrocketed.
What advice can you give a retailer who wants to increase profit in its supply chain, encompassing stores, online, distribution centres and other channels?
Retailers face numerous challenges in the quest to become omni retail. In the past, brick and mortar, online and distribution centers operated in distinct silos. When choosing a supply chain solution, retailers should:
Be sure your solution does not only consider historical sales. It should include both predictive and prescriptive analytics. This is the most profitable way to improve customer experience and maximise profit.
Be sure the solution addresses customer demand versus fulfillment location. These are often different. Weak solutions tend to overstock, but that only results in excessive markdown. Inventory levels should be accurate starting from the initial buy, through the distribution center, and following through to the fulfillment location, whether in stores or online.
Look at ROI: SaaS solutions like 4R are implemented quickly, and retailers see ROI in weeks or months rather than years. Paying exorbitant amounts of money with little assurance of ROI is a thing of the past. Look for a profit guarantee.
Visit 4R at RBTE 8 & 9 May on stand #510