Many people believe we’ve already lived through a data explosion in the past decade or so. But this isn’t entirely true – because the real data explosion is likely to happen over the next few years. So the big question for Performance Management & Controlling functions is, which innovative ways of using data could improve business decisions in the future? Here are three use cases which – in my view – are particularly interesting.
Traditional business management uses key indicators and dashboards based on traffic-light logic. Things become innovative when automatic comparisons are performed across large datasets (benchmarking), automatically triggering alerts in the event of deviations. Last Halloween, for example, algorithms alerted Walmart sales controllers to the fact that a new brand of cookie – selling very successfully at most outlets – was generating zero revenue at two stores in particular. A quick investigation showed that they had forgotten to display the cookies on their shelves – an oversight that was quickly corrected. The same idea could be applied to many other situations and thousands of KPIs in what could be described as “KPI-based Management 2.0”.
In 2019, McDonalds acquired AI specialist Dynamic Yield. And many people undoubtedly wondered what artificial intelligence has to do with flipping burgers. As it happens, there are lots of possible applications. One example is drive-through displays. Using smartphone location tracking, algorithms can tell who is currently in front of the display and then, based on the individual’s customer history, offer – for example – high-margin products. Or if the algorithm detects a long queue, it can instead display products that need minimal time to prepare. This reminds me of the traditional “planning production schedules to eliminate bottlenecks” approach which many readers will remember from lectures on financial controlling or operations research – but in this case digitised and optimised in real time.
DIY retailer Lowe’s used big data to identify the best locations for new stores. Using smartphone geolocation data (“grant this app access to your location…”), they worked out where the customers of every single Lowe’s store lived, as well as the customers of their major competitor Home Depot. Based on this analysis, they were able to plot trade catchment areas and relative competitive intensity by geographical area. They were also able to rank competing Home Depot stores by number of visitors (information they also obtained from smartphone geolocation data). This enabled Lowe’s to open stores precisely where they could best cannibalise sales from their competitor’s stores. This type of analysis can be used to make investment decisions for setting up stores of all kinds, from banks to gas stations to fashion boutiques.
For anyone interested in subjects like digital disruption or the digitisation of controlling, performance management, transformation of existing business models, investment controlling, AI, forecasting or predictive analytics, Frankfurt School offers continuing education courses and seminars aimed at people in full-time work, including specialists and executives, senior managers and CFOs.