Retail IoT: the hyper-relevant, competitive retailer

By Prem Couture, CEO, ShareMyInsight,

In a previous posting, I discussed how Retail IoT enabled stores are able to combine live shopper journey and product data with POS, loyalty, social media and other data sets. Also, how applying machine learning and Retail IoT enables real time insights that can transform the customer experience, enable customer centric merchandising and streamline operations.

In this posting, I would like share my thoughts and experience on how Retail IoT can power bricks and mortar stores to compete in an omni-channel world by becoming hyper-relevant across all customer touchpoints.

A Surging Wave of Disruption and Opportunity

As previously noted, classic retail strategies and methodologies for discovering and engaging customers are increasingly unmanageable, due to rapidly evolving customer interests and behavior patterns and as evidenced by:

  • the exponential growth in the amount of shopper journeys: from research to purchase to fulfillment and customer support, the number of possible journeys has grown from 40 to a maze of more than 800 (Cisco) and further increases over time.
  • the expanding number of data points (beyond spend and demographics) and the rapid change in consumer interests is making the traditional rules approach to data mining customers to be less and less meaningful.
  • the increasing demand for seamless shopping with greater choices and lower prices across online, in-store, and mobile platforms, is creating a ‘digital divide’ between consumer expectations and retailers’ ability to deliver.

Retail IoT: Innovation at the Heart of the New Retail Reality

If a sensor network represents our nervous system and a Deep Learning platform is our brain, then the part that manages retail processes from supply chain to merchandising and customer communications is similar to the way we engage and learn from the environment around us.

Enabling customers to make easy and cost efficient decisions from a wide array of choices is what a connected retailer precisely because it continuously learns and adapts to new information. Some of the key technology advances that make the above possible include:

  1. Retail sensor devices that act in a sensor fusion mode and live stream shopper and product data to cloud platforms.
  2. AI and Deep Learning: advances in GPU accelerated computing power enables Deep Learning algorithms to find patterns in large and disparate data sets and to transform data into insight.
  3. Store diagnostics can detect how product placement, brands, range assortment, pricing, personnel and store location affect shopper behavior and purchasing decisions.
  4. Dynamic, automated processes can trigger at key moments on the purchase decision path and engage customers on the preferred communication channel.
  5. A new evolution in CRM manages hyper-relevant and contextual customer interactions, delivers more efficient engagements and offers immediate customer savings.

Productivity Savings for both Retailer and Customer

A connected retailer can realize productivity gains across a number business areas, from supply chain to merchandising to marketing activities; further, help resolve issues which retailers have been struggling with for a number of years.

Here a few key areas where productivity gains are most visible in a connected environment:

Store Inventory Efficiencies

Retailers and FMCG partners have long known that incorrect product placements, poor shelf maintenance and out of stock conditions all contribute to significant losses in revenues.

Retailers and FMCG tackle the problem by utilizing field marketing agencies to periodically check for compliance with the agreed range, shelf share in the product category, share of competitors’ shelf and price for each item. Noteworthy is that a typical category audit samples only 2% to 5% of all store locations at a frequency of 1 time per week or every other week.

According to ECR when buyers can’t find the product they are looking for in its usual place, 9% of clients choose an alternative product, or do not make a purchase.  Out of stock is estimated to cost a retailer approximately 4% of sales in lost revenues.

In contrast, an IoT powered store with efficient, battery powered cameras that sends product images to the cloud for product recognition, can provide ongoing information on conditions and predict when shelves need replenishment.

Product and Inventory management is one of the key areas where IoT and Deep Learning can make a big difference by monitoring products and signaling when errors occur and replenishment actions need to be taken, resulting in achievable gains:

  • 2 monthly visits per store by a field marketing representative at a yearly cost of approximately $1,500 per category/store can be saved
  • merchandise placement errors across all IoT connected stores can be reduced by 50% or more
  • timely stock replenishment can reduce lost sales from out of stock products by 1%-2%

 Customer Centric Merchandising

The ‘one size fits all’ planogram deployed across all stores fails to consider that consumers and their shopping behavior differs by point of sale and many other factors.

Did moving the bakery section to the front of the store result in customers spending more time in the store? Did moving the wine section next to the cheese counter create more cross shopping between those 2 categories? Do we have just the right amount of sales people in the shoe department at peak shopping times and, if not, are we loosing sales?

Sensor fusion and Deep Learning can provide a level of diagnostics and insights that uncover which variables are working together to influence how shoppers make purchasing decisions. Further, suggest planograms and product assortments that target shopper preferences during their shopping journey, as well as optimizing pricing strategies and forecasting demand for better customer service.

By continuously detecting shopper journeys across merchandise zones and applying learning algorithms, analytics can pinpoint areas of assortment optimization, range localization and better product visibility, resulting in a shopper journey based store layout with improved shopping metrics and return on every square meter of shopping area.

Based on live store examples, here are some of the capabilities and efficiency gains obtained from implementing tracking sensors in shopping areas:

  • Monitoring of key metrics in every shopping zone, with clear visibility into over/under performing zones
  • Measuring the effects on shopping behavior before and after merchandise changes are put into effect, resulting in engineered store layout plans that increase traffic to poorly visited zones by up to 3%
  • Reducing time friction in service zones by detecting congestion and alerting the need for additional personnel, resulting in increased sales conversions of 1-3%
  • Increasing Return on Space in specific store zones/categories by more than 2% by flagging the need for space re-allocation and range planning
  • Faster reaction time to changes in shopping conditions and identifying probable causes e.g. Promo area traffic decreased by 15% because of low inventory conditions and the need to replenish stock

Hyper-Relevant Engagements, by Design

Influencing customers by getting inside their minds during the purchase journey represents an ongoing challenge for marketers.

Within current means, marketing department personnel supervise opportunities for engaging customers and create targeted marketing campaigns based on their best judgment. In addition, use communication channels that are unable to reach the customer at the moment of making a purchase decision. These types of limitations mean, for example, that a wine offer may reach a customer only after a shopping trip and when home drinking wine at dinner.

However, Retail IoT stores are powered by Deep Learning analytics and are in a position to deliver real time savings during the shopping lifecycle.  Deep Learning is the engine that provides hyper contextual and relevant interactions exactly at the right moment, thereby adding a layer of efficiency that is highly valued by the consumer.

With an ability to know what customers are looking for and need to know at a given moment during the shopping journey, marketers can achieve an unparalleled level of ‘response to conversion’ metrics:

  • Increased cross-shopping between zones by 8-15%
  • Increased basket size by 1.25% in targeted customer groups
  • Increased visit repeat rate, shopping frequency by 1.5%
  • Increased sales on promo items up to 4%
  • Increased sales conversions by floor personnel up to 35%
  • More interactions in service areas between store personnel and customers by 15%

Retail IoT Conclusion

By testing and deploying Retail IoT and Deep Learning for bricks and mortar stores, retailers are able to evolve their business in a challenging new environment.

Getting it right entails knowing your customers in a much different way than ever before, meeting their expectations as they change over time and becoming hyper-relevant across all touch points.

Intrinsic to success is becoming more cost efficient on all operations and finding the right balance between pricing, product assortment and customer services – all of which depends on a digitalized physical environment capable of detecting and adapting to conditions as they change.

Go to SMI’s IoT for retail platform

A few words about myself

As the CEO and principal architect at ShareMyInsight (SMI), I have been involved over the past 10 years in developing proprietary technologies and applications for big data analytics and statistical models on consumer behavior. In the last few years I have seen retailers increasingly struggle to create meaningful and relevant customer engagements, largely due to traditional statistical methods that are becoming obsolete. I believe that sensor fusion and Deep Learning technologies are now ready to replace traditional rule based models, enabling a new type of shopping experience that will benefit consumers, brands and retailers. My current focus is on the design to production cycle of a variety of in-store sensors that live stream data to the SMI machine learning platform for detecting, identifying and putting into action information for store operations, merchandising, marketing and customer communications.

I work with a range of partners, from consultants to market research, trade marketing and ad agencies, to solution providers and integrators. Feel free to contact me at




Retail IoT
Tagged on:         

Leave a Reply

Your email address will not be published. Required fields are marked *