The scale, speed, complexity, and consumer orientation of modern experiential retail create conditions ripe for artificial intelligence (AI) solutions. While retail is already the second-largest consumer of AI application software behind banking, in terms of adoption rates, the industry lags others.
IDC’s Examples of AI in Retail That Extend Human Capabilities” report document takes a deeper look at use cases focused primarily on extending human capabilities; this excerpt will highlight the promotional impact and pattern recognition solution from Periscope by McKinsey. The basic capability when supported with AI, can help retailers accomplish tasks faster and more accurately to deliver insights and customer-facing impacts at scale, which would otherwise not be possible.
For example, taking promotional analysis one step further represents a significant challenge for retailers without an AI/ML-backed solution. Solutions such as Periscope enable retailers to identify previously unknown or unknowable patterns in promotional activity, like the following:
- which past promotions worked best? Are those successful promotions repeatable, or are they outliers?
- what attributes of promotions are most critical to success—discount depth, features (dollars off versus percentage reduction), timing on the calendar, cadence, or region?
- what are the guardrails for designing future events? What are the optimal duration, frequency, and time in between promotions?
- what combination of attributes is the most likely to produce winning promotions? What is the optimal set of individual promotions across the calendar, and what is the expected ROI?
For retailers, leveraging AI/ML-based solutions for such analyses means faster and more accurate insights at scale, accelerated learning, and less subjective bias. This capability represents a significant step forward for most retailers, even those with highly mature analytics capabilities.
It takes a deep understanding of the full impact of promotions across categories—promotion halo effect analytics—to enable the broader promotional pattern recognition. Without a full analysis of the halo impact, any patterns identified would be inaccurate and retailers would miss the opportunity for a game-changing approach to marketing analytics.
*"Examples of AI in Retail That Extend Human Capabilities", by Jon Duke. Copyright 2019 IDC.