AI in Retail
McKinsey and Company reports that as much as 35% of Amazon’s onsite revenue is generated by its recommendation engine. Retailers know that recommended products can generate increased cross-sell and up-sell opportunities, but few have actually implemented truly personalized product recommendations on their sites, in emails or other channels. If they contain recommended products at all, many websites and emails still contain static recommendations where everyone who visits that page or receives the email sees the same products. While these recommendations are better than nothing, they are far from the individualized items that consumers now expect.
Every retailer wants to have high quality recommended products everywhere from emails to landing pages, their homepage to categories, product pages and cart to drive increased cross-sell and upsell. AI can be used to find the patterns in customer behavior from clickstream data, prior purchases, demographics and preferences that lead to the best product recommendations for each individual consumer. For example, when a customer visits the homepage of a website, the products can even categories displayed can be based on their known preferences and prior purchases such that the items displayed are highly relevant to them. For emails, product recommendations can be generated as an email is being opened so that the recommended products are relevant based on the customer’s most recent browsing behavior and purchases. With AI, highly relevant and personalized product recommendations, websites, email campaigns, call center agents, and mobile applications can provide personalized experiences for consumers that drive increased conversion rates, basket size and customer loyalty.
AI is ideal for optimizing assortment for retailers. AI models can look at a variety of factors including past sales, store display space, local trends, online behavior, predicted weather patterns, and more to determine which products would be the best fit for a given store location. This AI based optimization prevents stockouts by sending more inventory to stores where products are most needed and minimizes markdowns by making sure that products are on display where they can be sold at full price. AI models can even reroute inventory between stores to ensure that retailers can take advantage of local trends.
With retail assortments growing, increasing turnover and with many cases with decreasing store footprints, retailers need new ways to generate profits. Pricing for retailers has typically been driven from the corporate level by established pricing guidelines and competition. Markdowns for many retailers are driven by tried and true techniques with x% at 6 weeks, y% at 8 weeks etc. These traditional methods are insufficient to compete with new online or omni-channel competitors who are better positioned to capture profits through careful price management.
AI is ideal for situations where a retailer needs to optimize across a wide assortment of items based on a variety of factors. AI models can be used to determine the best price for each item using data on seasonality and price elasticity along with real-time inputs on inventory levels and competitive products and prices. The result is more careful markdowns on specific colors or versions to a very specific price to increase demand and maximize profits. Marginal price increases are also possible on some items to capture demand from trends. AI can also be used to provide reasons for pricing suggestions that indicate the key factors when making the pricing suggestion. This is helpful to retailers who want to know why particular items are being suggested for markdowns.