Retailers are aware of the significant opportunity to leverage vast amounts of customer data to drive in-depth customer analytics that enhance merchandising processes through AI. The goal is creating customer-centric assortments, across every touchpoint, optimally priced and available to consumers when and where they expect to browse, transact and acquire items. Gartner conducted the 2018 Unified Retailer Survey to understand organizations’ multichannel or cross-channel strategies and digital business initiatives. Below are the high level key findings from this research.
Key Findings
- Retailers overwhelmingly have or plan to implement AI solutions in five merchandising processes: product development and selection, planning, buying, demand forecasting and allocation and replenishment.
- For most retailers, AI will be accessed as part of advanced applications that enable merchandising processes rather than generic AI platforms.
- Practicality, transparency and explainability are foundational principles for successful implementations.
- AI implementations will be unsuccessful without significant organizational change.
In November 2015, Gartner published a prediction that, by 2020, merchant leaders will be using algorithms, prompting the top 10 retailers to cut up to one-third of headquarters merchandising staff. AI implementations in merchandising will be unsuccessful without significant organizational change. To facilitate this change, it must be confronted head on and with transparency across business users and senior executives. While Gartner predicted substantial reductions in merchandising staff, only a portion of this will be redeemed as labor cost savings. Many resources will need to be redeployed to improve customer experiences and support new collaborative activities.
Practicality, transparency and explainability will be key tools for successful transition to AI as part of algorithmic retailing in merchandising.
Practicality is a fundamental force in success with a merchandising team. Addressing a known problem, making things quicker and easier for the business, ultimately supporting their existing KPIs, will attract attention and positive responses from the team.
Transparency is a must with this group who will not trust until they verify. Avoid black-box solutions in favor of open algorithms and explainable results. This is a fine line, as they do not want to know the science, but will want to understand how it goes about learning and deciding. They will expect ability to manage parameters, rules and controls that are expressed in business terms.
Explainable AI (XAI) enables a better adoption of AI by increasing the transparency and trustworthiness of AI solutions and outcomes. XAI also reduces the risks associated with regulatory and reputational accountability for safety and fairness. Explainability is perhaps the most challenging to provide but will do the most good for progressing from pilot to production. Using plain business language, familiar visualizations and scenarios, the application “tells a story” about the results it determined. Users must then be able to question through “what-if” analysis and build their comfort level, supported by an all-around test-and-learn, team culture and approach to working.
Gartner clients can read the entire research with explanations, statistics, and examples here: What Retail CIOs Need to Know About AI for Merchandising
The Gartner Blog Network provides an opportunity for Gartner analysts to test ideas and move research forward. Because the content posted by Gartner analysts on this site does not undergo our standard editorial review, all comments or opinions expressed hereunder are those of the individual contributors and do not represent the views of Gartner, Inc. or its management.