What does artificial intelligence actually do for a retailer or brand?

In this series, retail experts at Shoptalk discuss the most pressing issues facing their industries today. Write your own #Shoptalk17 article here.

“We will move from mobile first to an AI first world” Sundar Pichai, Google CEO, April 2016

It’s hard to think of a technology that has more buzz (or dare I say hype) right now than Artificial Intelligence. With good reason, Artificial Intelligence is a game changing capability that is almost certain to drive a new era of business disruption and transformation. 

But what does Artificial Intelligence actual do for a retailer or brand? The challenge is that Artificial Intelligence is not a single capability or technology, that will manifest itself as a single use-case for retailers and brand. It’s a foundational approach that enables a wide variety of different capabilities. At Publicis.Sapient, to help think about this we use a framework with three major categories of capabilities:

Insight Generation – Using AI techniques to process large volumes of data (often unstructured data) to develop actionable insights that would not have been possible with previous approaches. This is about scale, giving business the ability to generate insights at a scale that was not previously possible.

Business Acceleration – Using AI to optimize business processes, often speeding up processes and eliminating friction for core business functions. This is usually about cost reduction.

Customer Engagement – Leveraging the power to AI to engage with customers in new and more effective ways than previous customer experiences. This enables entirely new customers experiences and touch-points, as well as dramatically improving existing ones.

Here are some examples of common retailer and brand use-cases applied to this framework:

Insight Generation –

·     Improved Analytics - Imagine being able to merge all the marketing data from your CRM and DMP, combine it with behavioral data from your web-analytics and transactional systems, and then generate insights about the segments of customers with the most untapped buying potential, the segments most likely to defect, and the segments that aren’t returning favorable value for your current marketing investment. Retailers are doing this now using tools like IBM Watson BlueMix APIs (such as the Cognitive Insights service).

·     Improved Targeting – Imagine being able to ingest the complete firehose of social media and identify your best prospects and the most effective message for each of them? Retailers are already using tools Google Cloud Platform and their open source Tensor Flow Tool or Lumino’s Analytics tool to find customer insights more quickly.

Business Acceleration –

·     Tagging Unstructured Data – Have users uploaded photos of themselves using your product to Pinterest, Instagram, or your own reviews platform but you don’t have enough details about the subject of every photo to use them as social proof in your own marketing? Products like Clarifai can auto-tag your images with relevant attributes, moderate inappropriate ones, and make all those user submitted images suitable for your marketing.

·     Inventory Optimization – Machine learning tools from IBM Watson or Google Cloud Platform allow retailers to leverage enterprise data to more accurately anticipate product demand. 

·     Merchandising Compliance – Retailers can now use computer vision applied to images of their shelves taken by robots or cameras mounted to shopping carts to determine if the in-store visual merchandising matches the store planogram, and identify out of stocks. In some cases these images can even be used to generate an inventory for the store. Simbe Robotics and Focal Systems are two vendors with current solutions.

Customer Engagement

·     Visual Search – Customers select photos of products they like (perhaps even that they’ve seen in the real world) and you are able to recommend the most relevant suggestions from your own product catalog. Retailers are doing this now with Sentient or Clarifai Visual Search products.

·     Recommendations & Personalization– Improve typical e-commerce product recommendations, by leveraging implicit signals shoppers are giving via their browsing behaviors. Even personalize product detail pages for each shopper. Tools like Sentient are doing this now

·     Improved Search – Imagine search results being more contextually aware based on shoppers behavior. IBM Watson and Twiggle both offer AI tools to super-charge your current e-commerce search.

·     Virtual Agents – Millennial and Generation Z shoppers want to engage with your brand via SMS and chat. Natural language processing from IBM and Google make it easy to build your own bots, and vendors like Makesnaps make it easy to deploy turn-key customer service or sales bots across all text and chat platforms.

While these are only a sample of kinds of retail experiences AI enables for retailers it’s easy to see how disruptive this be. Artificial intelligence doesn’t just allow us to do the same thing faster. It lets us do more, which we’re certainly going to need as we all struggle to meet the expectations of the digital enabled consumers.

I'll be moderating a panel on Artificial Intelligence at Retail at ShopTalk on March 21st in Las Vegas. You can hear from many of vendors mentioned in this article there.