Learn How Gartner's Algorithmic Approach to Merchandising is Changing Retail


“Algorithmic retailing is the application of big data through advanced analytics across an increasingly complex and detailed multichannel retail structure to provide for customer expectations that are driven higher by consumerization of retail.” – Gartner, Inc.

Complex algorithms that were once the domain of scientists and academics are now tools anyone from any role can leverage thanks to the rise of artificial intelligence and machine learning (AI/ML) technology platforms. Industries across the spectrum are using AI/ML to enhance the customer experience, reduce costs, or increase sales, all with the underlying motive to improve bottom line business results.

The same is true for retailers, where the use of algorithms isn’t necessarily “new.” In fact, algorithms have always been a part of the merchandising process, as planners rely on calculations in Excel or existing applications to determine the right inventory flow based on forecasted turn or potential margin. While this approach worked for a period of time, the new challenges and scale of complexities retailers face to meet demand anywhere while providing the perception of an “endless aisle” makes old forecasting methods and decision-making impossible with traditional algorithms alone.

"Leveraging algorithms is the only way that merchandising can meet the customer centricity challenges of digital business and the digitalization of retail." - Robert Hetu, Gartner, Inc.

First off, the volumes of data retailers have is too much to handle. Secondly, the depth of analysis is limited. Because of these challenges, retailers must evolve their current processes to handle massive amounts of data and make sense of it all. As a result, this new algorithmic approach to retailing, according to Gartner, encourages retailers to leverage the new technology available (like AI/ML) to support merchandising decisions.

Unlike traditional forecasting models, smart machines and advanced analytics are built to handle massive amounts of data, at scale. For example, at Celect, our platform utilizes patented AI/ML specifically trained for retail to predict demand and systematically:

  • Identify the most relevant structured and unstructured data sets to create models
  • Surface relevant attributes that drive predicted demand from messy or missing data
  • Address sparse data and short-life products
  • Identify the most statistically significant attributes that drive demand
  • Continuously improve predictions from supervised and unsupervised machine learning
  • Consider relevant context, such as the assortment choice available to the consumer during a point-of-purchase
  • Quickly pick up trends from existing and new data sets

In addition to using AI/ML to predict demand for retailers, we also use AI/ML to help optimize these predictions with real-life business constraints (i.e., store capacity constraints, assortment mix, inventory turn, in-store stock availability, etc.). Not only is AI/ML able to detect relevant insights from all the “noise” in the data, but it’s also very effective when it comes to considering the entire pool of inventory, objectives, and constraints to understand all possible trade-offs (and the potential to maximize sales) for every decision a merchandiser makes.

The power of AI/ML to a merchandise strategy cannot be understated. Access the full Gartner report below to learn more about achieving success from algorithmic retailing.

Gartner, Algorithmic Retailing: Merchandising Leads the Way, 5 February 2018, Robert Hetu.

Download your complimentary Gartner report! 

Topics: artificial intelligence, inventory optimization, brick-and-mortar retail, physical stores, machine learning

Ready for more?

Request a Demo