Accelerating AI Value in CPG: Closing the Gap Between Potential and Performance

The consumer packaged goods (CPG) industry is evolving faster than ever – and the pace of change is only accelerating. As markets shift faster than strategies, AI is becoming the critical differentiator. Take food trends, for instance, which can turn on a dime.

When consumers are suddenly looking for high-protein options or low-sugar alternatives or hot new flavors, they want to see choices on the shelves – and they want to see them now. That puts CPG companies in a tough spot.

Given the many moving parts of the CPG supply chain, rushing new products to market to accommodate changing customer behaviors can be a tall order. Predictive analytics and rapid product development could help – both of which sound like great openings for AI to step in and save the day, right?

Despite these and other significant opportunities, however, CPG often lags other industries when it comes to taking advantage of AI innovations. This despite the fact that one study showed double-digit gains in productivity for CPG teams working with AI.

Of course, there are very real issues confronting CPG companies looking to implement AI, but there also is a world of potential for those that get it right. Here’s how CPG firms can overcome their AI challenges, seize the opportunities and better keep pace with the fast-changing world around them.

Understanding the Lag: CPG’s AI Problems It’s not that CPG companies don’t want to take advantage of AI’s potential. But every industry has its own unique challenges related to AI implementation, and those facing CPG can be particularly difficult to crack. They include: Data Fragmentation: When it comes to AI, good outcomes require good data.

And while most companies have to deal with the challenge of getting their hands on clean, usable and trustworthy data for their AI experiments, CPGs have an even tougher task in this regard. Why? Because the CPG value chain is bigger and more complex than those in most other industries, comprising a host of suppliers, distributors, retailers and consumers.

With massive amounts of data coming from so many channels, it can be nearly impossible to get the comprehensive view needed to use AI for things like personalization, forecasting and supply chain resiliency. Risk Aversion: Like their peers in some more conservative industries, CPG companies often move at the speed of compliance rather than curiosity.

Uncertainty is the enemy, and staying the course is the prudent move. Food production and distribution companies, for example, tend to focus more on meeting applicable standards and regulations than they do seeking out different ways to innovate their products or processes. Prudence has its place, of course – regulators and public trust do matter, after all.

But progress requires motion. Eschewing experimentation and slow-walking innovation doesn’t work when it comes to AI. And sitting on the sidelines is definitely not a strategy for long-term success.

Legacy Reliance: CPG companies are often operating on old infrastructure and outdated tech platforms when it comes to things like their ERPs and supply chain systems. Extensive customization of these platforms over decades, coupled with duplication across different countries or regions, has complicated data consolidation and the effective application of AI.

Even if they’re experimenting with a variety of AI tools and solutions, these companies are likely running into trouble trying to integrate this technology with existing legacy systems. This leads to a significant gap between what they want to do and what they can do with AI.

Overcoming the Hurdles: Making AI Work for CPG Can CPG companies overcome these challenges? The short answer is that they have to. There’s simply too much potential – from productivity gains to cost savings to new product development – to risk sitting this one out.

Instead, they need to start confronting the challenges head-on, which means: Creating a safe place to experiment: AI is all about trying things. Not every experiment will work out, but it’s only by taking some risks that CPG companies will be able to determine which tools and solutions can truly make a difference for their business.

Prioritizing legacy modernization: If it’s been on the to-do list for a while, AI should be the reason to move it up that list.

Ideally encompassing both the application and data layers, modernization can allow a CPG company to better understand its existing infrastructure and engineer it into new architectures that support the kind of risk-free experimentation that AI demands.

For example, a major beverage company might focus on unifying its fragmented ecosystems into a more modern and cohesive global platform. In doing so, it can enable quick experiments and heavy AI use, such as tailoring sales representatives’ task lists. It’s a heavy lift upfront that can lead to countless opportunities down the road.

Fostering a culture of innovation: AI is often delegated to IT, but it isn’t strictly an IT proje