Data to Dollars
Two AI experts explain how exactly you can maximize your margins by integrating, adopting, and executing AI-powered solutions.
*This post was co-authored by Alex Barnes, Senior Director of Product Management, antuit.ai, part of Zebra.
###
As a member of the Forbes Tech Council recently pointed out, AI has the potential to benefit demand forecasting and inventory planning in five ways for retailers and consumer packaged goods (CPG) companies:
1. Improved accuracy. AI can analyze vast amounts of data to identify patterns and predict future demand more accurately than traditional methods.
2. Faster analysis. AI can process data more quickly than humans, enabling businesses to make decisions in real-time.
3. More comprehensive analysis. AI can analyze multiple data sources simultaneously, including structured and unstructured data, to provide a more complete view of demand and inventory.
4. Reduced human error and bias. AI automates mundane tasks, which helps minimize the risk of data entry errors and mitigate individual biases that can skew product demand predictions, resulting in a more balanced perspective based on objective data and actual demand trends.
5. Cost-effectiveness and improved revenue. AI-optimized inventory effectively works on two fronts: it lowers inventory costs by curtailing the risk of excess stock and boosts potential revenue by reducing the chance of lost sales, thereby circumventing the ironic predicament businesses often face of simultaneous out-of-stock and surplus inventory scenarios.
To fully realize the benefits of AI, though, you need to ensure you have the right integration, adoption, and execution strategies in place.
Integration. AI systems need to be integrated with existing systems and data sources to ensure seamless data flow and enable accurate predictions. This may involve developing APIs and other integration tools to connect disparate systems.
Adoption. To ensure successful adoption of AI systems, you’ll need to educate your workforce on how to use them effectively. There is a direct correlation between training investments and return on investment, as shown in this article. Just be prepared to provide ongoing training and support to help employees understand how AI can benefit their workflows and improve their decision-making processes. This isn’t just a go-live readiness exercise.
Execution. Successful execution of AI systems requires ongoing monitoring and fine-tuning to ensure optimal performance. You must be prepared to continuously evaluate your AI systems and adjust as necessary to ensure accurate predictions and maximize margins – then educate your team on how the adjustments affect them.
Perceived Business Concerns with AI Adoption
Despite the widely reported benefits of AI, particularly adaptive AI, you (or someone on your team) may still be hesitant to adopt AI or machine learning (ML) technologies due to perceived concerns, such as fear of job loss, privacy concerns, and uncertainty about the reliability of AI predictions.
One common fear you may hear among employees is that AI will replace their jobs. However, research shows that while AI may automate certain tasks, it is unlikely to replace entire jobs, especially when it comes to supply chain planning, inventory management and more. According to a study by the World Economic Forum (WEF), while AI is expected to displace some jobs, it is also expected to create new jobs and transform existing ones. The WEF’s Future of Jobs Report states that by 2025, AI and automation will lead to a net increase of 12 million jobs globally. Rather than replacing humans, AI is expected to augment human capabilities and improve productivity, allowing employees to focus on higher-level tasks that require creativity and critical thinking – which are absolutely required when actioning the demand forecasts and inventory plans.
You and other business leaders can assuage these fears by providing employees with the necessary training and support to effectively integrate AI into their workflows. By involving employees in the AI adoption process and demonstrating the benefits of these technologies, you can help employees feel more comfortable with AI and view it as a tool that can enhance their work rather than a threat to their job security.
We know another common concern with AI adoption is privacy. As AI systems analyze vast amounts of data, you must ensure that you are protecting customer and employee privacy – as you would with the use of any other technology. This may involve developing strong data security policies and protocols and obtaining the necessary consent from customers and employees for data collection and use – which, again, shouldn’t be much different than what you’re already doing today. If you’re not sure you’re taking the proper precautions with other technology systems, you may want to check out what these experts advise.
Finally, we know you may be hesitant to adopt AI due to uncertainty about the reliability of AI predictions. However, as noted earlier, AI has been shown to provide more accurate predictions than traditional methods, particularly when analyzing large datasets. By carefully selecting AI models and continuously monitoring their performance, you can ensure the accuracy and reliability of their AI predictions.
To illustrate the benefits of AI in retail and CPG, let's look at a few real-world examples of how retailers and CPG companies are already leveraging AI for stronger business outcomes:
Intelligent Order Promising - During the pandemic, a multi-billion-dollar consumer goods company was required to react and adjust more quickly to inventory planning and order fulfillment to counteract skyrocketing consumer demand for food and consumable products. To optimize business performance in these market conditions, this global company launched a strategy focusing on a competitive advantage by investing in data and analytics. One critical area that promised significant business benefits was order processing and available-to-promise (ATP). The system helped them achieve a 4-5% improvement in case fill rate for strategic customers, and 10x ROI (Return on Investment) from increased revenue and reduced OTIF penalties.
Demand Forecasting - PacSun, a leading retailer of California lifestyle clothing, used AI-powered demand forecasting for allocation and fulfillment, to improve inventory accuracy and reduce stockouts. The system helped the company double its ship completes, forecast and allocate omnichannel demand, and balance inventory between stores, distribution center (DC), and web-depot locations for in-store and online sales.
“Antuit.ai’s solutions and expertise ensure we have the right inventory in the right stores, which significantly improves our ship completes.”
Shirley Gao, CIO, PacSun
In another example, Bimbo Bakeries USA worked with the antuit.ai team to collaboratively tailor an AI-powered demand forecasting and predictive ordering platform to support different front-line workers via custom user interfaces (UIs). Everyone from operations managers to DSD drivers can now open their respective UI to right-size production and localized delivery plans down to a SKU/store/week level factoring seasonality, local events, promotions, and other outside influences that may not be considered with human-led demand forecasting and inventory planning models.
While so many CPG companies—and competing bakery companies—struggled with supply chains and logistics for months on end during the pandemic, the Bimbo team was able to adapt its forecasting and production in less than a month to meet the heightened demand for their baked goods as more consumers started to eat at home amid restaurant closures. That’s right. In just a few weeks, AI enabled them to right-size their production volumes, adjust delivery routes to avoid out-of-stocks, and properly staff production lines, loading docks and trucks to meet the skyrocketing demand.
When a colleague asked Morgan Smith, Vice President of Direct Store Delivery (DSD) Center of Excellence at Bimbo, to evaluate the quality of the AI data modeling, he offered perhaps the most succinct answer we’ve ever heard:
“Nine times out of ten, it knows what my wife wants to buy on a Tuesday with a coupon in her pocket and she’s hungry. That’s unbelievable accuracy.”
You can read more about why Morgan, and the whole Bimbo team, have leaned heavily into AI in recent years in this blog post and accompanying success story. I also recommend you take some time to browse these other AI success stories, as they all illustrate how AI can drive significant improvements in demand forecasting, inventory management, and overall profitability for retail and CPG businesses.
Ready to Improve Your Margins?
We’ve said this before but feel it’s important to re-emphasize: AI offers the potential to revolutionize the retail and CPG industries by optimizing demand forecasting, inventory planning, and pricing and promotions. However, you must ensure you have the right integration, adoption, and execution strategies in place to fully realize the benefits of AI. By addressing perceived concerns, involving employees in the adoption process, and highlighting various successes, you can overcome barriers and unlock the full potential of AI like companies like Bimbo and PacSun have to optimize margins and gain a competitive edge in the market.
Thanks for reading our three-part series. If you would like to learn more about the power of AI for your business, contact us.
###
Missed the first two posts in this series? Catch up now:
Nicholas Wegman, Ph.D.
Nicholas Wegman leads the AI product development group at antuit.ai, a Zebra Technologies company, where he is oversees building of the algorithms behind their data-driven forecasting and optimization solutions. He brings extensive experience leveraging machine learning and data science across a wide range of retailers and CPG companies. Prior to joining antuit.ai in June 2013, Nicolas worked as a mathematics professor at a liberal arts college in the Midwest. He holds a doctorate in mathematics from Purdue University.
