As the mandate is increasingly on retailers to understand and serve their customer in more varied, broad and transparent ways, predictive analytics has emerged as a powerful, rapidly developing platform that offers strong value across the spectrum of work flows and operations for many different businesses.

Analytics has long been a key fixture of effective planning and review, allowing companies to better understand the wide-ranging consequences of the many decisions they make on a regular basis. In the highly competitive and diverse retail industry, predictive analytics can offer varied opportunities for improvement. Businesses can use this powerful platform to target specific goals, both addressing pain points and seizing opportunities for growth.

The broad benefits

Here are just a few examples of the innumerable uses of predictive analytics:

• Revenue and profit forecasting — Both on the companywide level and for individual stores, a predictive retail analytics platform can draw on past information as well as modeling and machine learning functions to provide effective predictions about income in the coming quarters and throughout the next year. Along with directly informing changes to inventory levels, promotions and other considerations, this information can serve a role in other developments, such as selecting potential sites for a new store location.

• Customer relationship management and improvement — Predictive analytics draws on both broad demographic data and the actions of individual customers to help retailers formulate more effective engagement and marketing strategies. Increasing positive feelings and developing consumer loyalty are universal goals for the retail ­industry.

• Identifying leading and lagging products — Large retailers with a diverse mix of offerings may not have total visibility into the value of individual items. Predictive analytics allows businesses to drill down and make individual determinations that boost overall performance and meet customer expectations.

Ways predictive analytics can help

Michael Kim

One common issue for retailers with many product offerings is correctly determining the value of each SKU offered and finding opportunities to eliminate underperforming items. Taking SKU rationalization as the overall goal of predictive analytics can lead to clearer insight into their products. While eliminating low-selling items is one objective of this type of analysis, it also helps retailers to find opportunities to increase inventory levels of high-performing offerings and craft an ideal product mix.

By using market basket analysis and margin mix modeling, and drawing on existing sales and inventory data as well as other informational sources, retailers can generate a list of underperforming SKUs that could easily be eliminated going forward without any negative consequences. This approach ensures that overall performance will continue at the same level and revenue will remain stable, with the potential to grow going ­forward.

Retailers taking this approach can expect major, quantifiable benefit from a well-executed retail analytics implementation, to the tune of a 1% to 2% increase in gross margin overall. That kind of improvement in insight combined with the strong financial gains make predictive analytics a “must have” for certain types of retailers. From our vantage point, one of the greatest values of predictive analytics we see is its adaptability. Whether a retailer wants to eliminate underperforming SKUs or more effectively reach its customers, an advanced data analytics implementation can effectively address such issues and provide a path toward success.

Mike Kim is a director at AArete, a global consultancy specializing in data-informed performance improvement, and heads its Center of Data Excellence. He can be reached at [email protected].