Author: William Vorhies
Summary: Especially in consumer goods and retail the value of AI/ML is only part of the story. AI/ML will increasingly need to integrate with helper technologies to deliver maximum value. Up your game in IoT, 5G, and robotics to ensure you’re giving your operating team all the best options for their investment.
When I come across interviews with business leaders talking about technology trends I always try to pay attention to how much they seem to know about AI/ML. For example, very few have yet to pick up on the importance of Platform Strategy. Still, many times I get a lesson about thinking a little beyond the narrow confines of what AI/ML can do. It’s not the only tool in the toolbox.
So I was interested in CBINSIGHTS recent report encapsulating what 11 really smart senior execs in consumer goods and retail had to say about what technologies and trends would change their sector over the next 3 to 5 years.
Before we get to the meat of it, a little about the 11 folks interviewed. All are execs or investors in senior roles focused on innovation in their brands. Three are VC investors, 8 are operating execs. Five of the CPGs are in the food & beverage segment, one in fashion, two in electronics.
You would expect all of these folks to focus on the retail consumer and how to reach them. Specifically how to stand out and differentiate from competitors on the shelf or the ecommerce site. Also, since these are all manufacturers, I expected to see some focus on supply chain.
Two Broad Categories
Their observations fall into two broad categories seen from the data science side of the fence. There are a few needs which AI/ML can traditionally fill on its own. However the bigger lesson for data scientists is that most of what these individuals foresee involves AI/ML blended with an additional technology. Here I’m talking about 5G, robotics, and specialty devices.
When asked by our operating execs how as data scientists we can help, I think we’re all more than a little prone to think narrowly about standalone AI/ML, especially since our personal knowledge of these other technologies may be thin. Here were some consistent themes among the interviewees to help you think more broadly.
Identifying Underserved Segments
Finding important outliers is a core strength of AI/ML. Whether it’s from web logs or register data blended with external data we just need to be aware that these outliers can be important. Victoria Treyger of Felicis Ventures observes:
“Traditional CPG products are designed with a one-size fits all approach that assumes the 15-year-old girl and her 39-year-old mother will buy the same products and shop in the same way….Expect dramatic growth in clean products, brands for underserved segments, and new distribution approaches.
Clean products will become mainstream in categories from deodorant to beauty and vitamins and supplements. Brands built around the needs of underserved customer segments including men, Gen Z, and people of color will define previously sleepy categories.”
Ms. Treyger’s observation also reminds us that segments aren’t necessarily demographic (though they may be). For example clean products, or adapting brands for other users or countries. We’ll find hints in the data that can be fed back into the product and marketing loop to create these brand extensions.
From here however, every major trend reported involved adding a helper technology.
Transparency, Clarity, Purity
Several folks in the food and beverage area talked about the desire for consumers to have a better understanding of where their food comes from, what’s in it, and how it benefits their bodies. Giovanni Battistini of Ferrero observes:
“Wearable personal health devices and personal nutrition services are capitalizing on the scientific progress in omics and biosensors to deliver meaningful clarity about how food impacts their bodies.”
This is the first of many comments about sensors and IoT data flows that data science must integrate into information platforms directed to the individual consumer.
IoT Devices Directly Interacting with the Consumer
All three groups of innovators, food & beverage, fashion, and consumer electronics mention the importance of specialty and enhanced devices to increase the attractiveness of the product and/or to enhance the experience of the user. Mentions included:
- Virtual try-on in fashion, in-store and on-line.
- Expanded image recognition. Facial recognition is just the start. Applications using surface recognition and 3D sensing are next.
Antonio Avitabile of SONY says “A plethora of applications from teleportation to interactive volumetric content streamed live to any consumer gadget will change the way we consume and create content.”
- Explosion of smart wearables in new form factors including hearables, eyewear, and smart watches.
In addition to our native capabilities in streaming data and edge compute, you should immediately see that almost none of this will come about without 5G.
More on 5G and Robotics
Several interviews mentioned the need to transform the retail space. One trend that’s been in the press as a recent experiment is ‘unattended’ or ‘unmanned’ retail space. This goes well beyond simple cashier-less pick-and-leave retail.
This is also the basis for transforming the physical retail space to be more seamless with the ecommerce space, and to create a physical draw by creating ‘entertainment and brand education’ into the brick-and-mortar location.
People will still be needed, just not for the routine and repetitive. Robots will be needed for inventory replenishment or even in-store product delivery. The people will be interacting with customers in a new higher-value role.
Again, the helper technologies of robotics and 5G make this possible.
The Message for Data Scientists
Especially in consumer goods and retail, the value of AI/ML is only part of the story. AI/ML will increasingly need to integrate with helper technologies to deliver maximum value. Up your game in IoT, 5G, and robotics to ensure you’re giving your operating team all the best options for their investment.
About the author: Bill is Contributing Editor for Data Science Central. Bill is also President & Chief Data Scientist at Data-Magnum and has practiced as a data scientist since 2001. His articles have been read more than 1.5 million times.
He can be reached at: