It seems like every day, there’s a new study or report about how AI, GenAI and machine learning will transform industries.
For Georgia-Pacific, this isn’t forward-looking speculation. It’s our reality. When we look at the innovation investments Georgia-Pacific has in play, we’re predicting a nine-figure return. We see $100 million worth of opportunity just in terms of how AI is enabling us to transform our manufacturing facilities.
A key lesson we’ve learned along the way – one that I hope will be helpful to other business leaders – is that when it comes to integrating AI and GenAI into your business, it is so important to retire a “one-and-done” mentality. It is tempting for a business to think, “I'll develop a killer model, and it will revolutionize my organization’s future.” If only it were that simple. Based on what we have learned, it’s about embracing a long-term innovation journey.
Let's say you want to build a pricing model. You’re going to look at your historical data to build it. And because that model worked effectively, your sales conversion rate goes from 10% to 40%. Now, you have a whole lot of new data about customer behavior. To keep momentum, you have to build a version two. Then, down the road, a version three. And so on. New data can always lead to new opportunities to refresh and retrain your AI models – and to look at new modeling possibilities.
What informs the AI strategy?
The secret to successfully evolving AI over time does not rest on the shoulders of an omnipotent data scientist behind a curtain. It’s the people using technology as part of their daily workflows who drive the progress. At Georgia-Pacific, this includes all of us – from folks in the C-suite to the teams working in our manufacturing plants.
Right now, we’re deploying SAS AI technology to empower our floor operators. It’s a huge leap forward. The technology not only flags when something may go wrong; it also provides suggested corrective next steps. It truly augments the operators’ abilities and helps upskill them at the same time.
In this case, humans are an important part of unlocking the value of AI and GenAI. There is no replacement for a human’s experience and judgment. Only a human can answer a simple but critical question: Are AI outputs meeting expectations?
This is why establishing robust feedback loops is so important. A stream of daily and weekly feedback is necessary for continual optimization over time. As we pursue deployments of GenAI technology like computer vision to improve safety and quality in manufacturing facilities, we 100% rely on continuous input from operators and managers to ensure our models are optimized.
The other key variable is ensuring you have the right platforms and tools in place to scale horizontally. We’ve learned that access to low-code or no-code capabilities provides a huge advantage in our ability to scale. Writing new code takes time – and that’s going to slow you down as you look to pivot and evolve.
Culture is critical
Successful transformation is about evolving company culture as much as it is about deploying new tech. As Georgia-Pacific has been an early mover on AI and GenAI trial and deployment, we know the many efforts and resources that go into any new use case. When new technologies are rolled out, it’s human nature to interpret zero user feedback as a good sign. But if you’re hearing crickets, it suggests people aren’t embracing the technology. If people are not adopting it, you’re not unlocking value.
Change is difficult. When people flag issues or frustrations, it’s a signal that they are adopting the technology and adapting their ways of working. As one of my early managers once told me, feedback is a gift. The tougher bit from a change management perspective is ensuring people (especially leadership) embrace the idea that continual feedback is an absolute must if you want to maximize the value of GenAI.
When people truly adopt a new technology, it gets woven into the fabric of their daily working lives. And when issues arise, you are going to hear about it. I think about the advanced AI analytics tools in our manufacturing facilities. They’ve enabled our teams to reduce unplanned downtime by 30%. It’s remarkable, but it’s quickly become our new normal and what teams expect.
Change is difficult. When people flag issues or frustrations, it’s a signal that they are adopting the technology and adapting their ways of working. As one of my early managers once told me, feedback is a gift.
I think of it like this: You deploy a new AI use case. After the “test drive” phase, it’s like people have a fancy new sports car. It offers them amazing performance (and generally makes life more fun). But just like with any car, the performance isn’t going to stay the same over time. In deploying a new AI use case, it’s important to be preparing for the next model – and you want your drivers to tell you when it’s a good time for that trade-in.
The bottom line is that fully utilizing AI and becoming data-led is not just about algorithms. It’s about fostering an innovative culture. For us, it’s been so important to ensure that every team member recognizes the role they play in helping keep the organization’s innovation trajectory on track.
In parallel, we have seen the power of leadership changing their mindsets – embracing the idea that our success will be underpinned by continual innovation fueled by new data and feedback loops. In other words, the “one-and-done" mentality is gone.
