Virtualization Technology News and Information
Article
RSS
Fluent Commerce 2024 Predictions: In 2024, AI in Retail Will Depend the Most on People and Data

vmblog-predictions-2024 

Industry executives and experts share their predictions for 2024.  Read them in this 16th annual VMblog.com series exclusive.

In 2024, AI in Retail Will Depend the Most on People and Data

By Nicola Kinsella, SVP of global marketing at Fluent Commerce

Generative Artificial Intelligence (AI) dominated the headlines in 2023. But will it deliver for retailers in 2024? 

With the introduction and massive overnight adoption of technologies like ChatGPT, millions of people everywhere have discovered new ways to use AI to generate content from email to images to music or even code.

In 2023, ChatGPT amassed 100 million users in just two months but then appeared to decline as more people accessed it through enterprise accounts. Corporate interest in generative AI and Machine Learning (ML) reached a level never seen before. Why? Because the promise is so big and the fear of missing out is even bigger.

Even today, much of what you see of AI is demonstrated in its potential form. That's because AI continues to learn. And it can only learn and grow next year with the help of the sources it depends on - people and the right data, especially in retail.

When it comes to retail, AI will continue to rely on people and up-to-date data to help drive more sales, enhance customer loyalty, and improve inventory management.

Enterprise organizations can achieve much more when they combine humans and generative AI. For example, Harvard Business School research found that consultants using ChatGPT-4 finished 12.2% more tasks, completed them 25.1% faster, and achieved 40% higher quality results than those who did not use the tool.

In a retail context, clean, accurate, and accessible data is vital for AI to build product inventory models correctly, predict fulfillment and labor requirements, and create shipping strategies that ensure a positive customer experience.

Finding and extracting the right retail data can be tough. The information is poorly structured, often incomplete, and resides in multiple systems. Retailers need data to train and test AI models by recreating the exact conditions in which an order was sourced.

That data should include, for example, when the order was placed, where the item is available, how much labor will be needed at a particular store, and how long it takes to process the order from multiple locations. 

The cost of finding all this data can account for 80% of an AI project budget. And in many cases, even after much effort, organizations still don't have the right data. So, the project fails before it's launched.

To successfully use AI/ML, retailers need two things: the right questions to solve and clean data that contains signals that are relevant to those questions.

A modern order and inventory data management system (OMS) helps businesses address these questions to enable AI/ML to improve inventory availability and order management, resulting in faster inventory turns, lower delivery costs, quicker and more accurate deliveries, and more efficient fulfillment operations. 

Modern event-based distributed order management will play a key role next year in helping retail professionals access and leverage the right data for AI/ML to reduce project failures and drive growth. 

This technology can capture time series data and other contextual data, such as general order history, inventory positions at a specific point in time, fulfillment rules, and attributes for locations, products, and customers. All of this contextual data can be stored, not purged or condensed, so it's always available for future analysis.

In addition, composable OMS technology is highly flexible and able to improve inputs to extend AI/ML models. This enables users to capture and tag any additional data when needed. This can make it easier to extract information by tagging data that contains signals.  

OMS can provide relevant and enabling data by adding custom attributes to orders, returns, locations, products, shipments, and inventory positions. And the systems can capture point-of-sale transactions. This provides retailers with a complete picture of offline demand, which is critical to support AI/ML use cases in the future.

With an OMS, AI models digitally become more agile and flexible, allowing them to evolve over time to meet changing demands. What works today may not be the best choice in the future. OMS can also enhance flexibility for workflows and user interfaces to allow for different types of model outputs.

OMS technology helps people find the right data to enable generative AI to achieve its fullest potential in retail, improving profits and customer loyalty.

##

ABOUT THE AUTHOR

nicola kinsella 

Nicola has over 20 years experience in B2B and B2C enterprise commerce, supply chain, and logistics technology. She holds a BA in History and Politics from Macquarie University and is a founding member of the Product Marketing Alliance. Born and raised in Australia, Nicola has worked most of her career in the greater NYC area.

Published Monday, December 11, 2023 7:30 AM by David Marshall
Comments
There are no comments for this post.
To post a comment, you must be a registered user. Registration is free and easy! Sign up now!
Calendar
<December 2023>
SuMoTuWeThFrSa
262728293012
3456789
10111213141516
17181920212223
24252627282930
31123456