Predicting Consumer Trends: How AI Forecasts Shopping Behaviors

From chatbots to inventory management, the use of AI in retail continues to grow, but how can brands use AI to understand the desires of shoppers? In a recent guest post with Consumer Goods Technology, Gopal Tadiparthi, Head of ML/AI at Insite AI discusses how AI-driven insights can uncover what consumers want today and what they will be buying tomorrow.

CPGs (Consumer Packaged Goods companies) have access to extensive shopper data from their own sources, third-party panels, macroeconomic reports, social listening insights, and retailer purchasing and loyalty data. AI can analyze this wealth of information to deliver strategic findings on shopper behavior. By leveraging AI, brands can enhance customer segmentation, forecast future shopping behaviors, personalize shopping experiences, conduct sentiment analysis, predict customer churn, and map customer journeys. These capabilities allow brands to optimize their marketing, sales strategies, and product assortments, ultimately leading to improved performance and stronger collaborations with retailer partners.

Key Points:

• AI and platforms like Walmart Luminate are revolutionizing shopper research by providing deep insights into consumer behaviors.
• Brands can uncover data on BOPIS usage, brand-switching, pricing, and promotions to predict future shopping trends.
• CPGs have access to extensive data sources, which AI can analyze to deliver strategic findings on shopper behavior.
• AI helps brands enhance customer segmentation, personalize shopping experiences, and conduct sentiment analysis.
• Leveraging AI in shopper research allows brands to optimize marketing strategies, improve product performance, and strengthen retailer collaborations.

Making Private Label and Consumer Brands Work Together On-Shelf

Private-label sales had a banner year in 2023, but looking forward, retailers also need to strike a balance between private label and national brands on shelves. In this guest column with Food Processing, Kristine Joji, EVP of strategy consulting at Insite AI, discusses how predictive analytics can help retailers develop optimized assortments between private label and national brands.

Private-label sales had a banner year in 2023, as retailer-owned brands continued to gain share across categories, especially in food. Yet, looking forward, retailers also need to strike a balance between private labels and national brands on shelves, putting a greater emphasis on developing fully optimized assortments.

For this reason, predictive analytics will have a huge impact as retailers and consumer goods companies leverage technology to create the most localized and optimized mix of private-label and national-brand items. The goal is to get both sides working together.

According to research by Circana and the Private Label Manufacturers Assn., private label dollar sales grew by more than $10 billion in 2023. For the full year, refrigerated products generated the highest dollar sales — $46.7 billion. General food items grew 10%, year over year, the highest growth category in private label behind beauty (+10.5%).

The research showed private labels grew in every category except tobacco and outpaced the growth of national brands, growing by 4.7% versus 3.6% for consumer brands. All told, private labels had a very good year, as consumers looked to the products to save money and the items found their way into more shopping carts.

For example, Kroger’s budget-conscious brand Smart Way was the fastest-growing private label last year, up 4.5% in household penetration. Adding to the trend, Target recently announced its low-price brand deal worthy.

Yet, with all the trends and momentum circling private-label products, how can retailers derive an optimal balance of private label and national brands? The answer’s in the data.

To be clear, there is no right answer in the debate of private labels and name brands. First and foremost, retailers should focus on providing consumers with a hyper-local assortment with the products their shoppers want. Using predictive analytics, retailers and brands can localize product mixes that sell best at an individual store level.

Advancements in how AI can parse data and recommend assortments that generate the most volume and revenue can bring brands and retailers together to find a mix of private labels and national brands. Together, the companies can identify how to best fit an assortment to every store’s varied demographic and buying habits.

It’s no longer about how private brands and national brands are competing against one another but how they join forces to build a product mix and assortment that best meets shoppers’ needs.

AI-powered assortment optimization

AI models read large amounts of data (sales numbers, retailer loyalty data, shopper data from programs like Walmart Luminate, macroeconomic data, third-party panel research, etc.). From there, category and brand managers coach the AI to learn and predict how a store-level assortment can be shaped.

Within that learning, brands and retailers can work together to focus on a few issues such as:

• Differentiating SKUs. By region and each individual store, retailers can leverage AI to uncover unique assortments that differentiate SKUs to drive incrementality. Retailers identify assortments that meet each local community, reacting to a sense of how a region shops. The assortments discover what products, private labels and national brands, resonate most with each area.
• Innovating categories. AI also combs through data to identify potential innovation gaps in product categories. A retailer can spot a potential product trend to introduce within its private-label portfolio to lift an assortment and category. At the same time, a CPG can see where there are innovation opportunities to present to retailers. Together, brands and retailers can create assortments that distinguish and excite assortments.
• Addressing the price gap. Shoppers no doubt buy private brands to save money, as there tends to be a price gap between private labels and national brands. AI can optimize pricing with price elasticity models to identify ideal price points for all products in the assortment, striking an optimal balance of what shoppers are comfortable spending on private brands and national brands.

Consumer goods brands don’t need to view rising private-label sales as a threat. Instead, companies should look at increased sales of both consumer brands and private labels as an opportunity to enhance collaboration.

AI and predictive analytics provide both companies with an unbiased look at how products are performing. It’s up to both companies to use the data and find ways that lift all brands and deliver an assortment that feels local and special to each store in the network.

Kristine Joji is EVP, strategy consulting at Insite AI, and formerly VP of Merchandising at Walmart. Kristine led strategic initiatives that resulted in substantial revenue growth for the company across Grocery and prior to that Personal Care. Widely recognized as a visionary leader, she played a pivotal role in optimizing Walmart’s merchandising with large CPGs.

Decoding Complexity: Navigating the Intricacies of the Wine Aisle

By Capri Brixey

Inside the four walls of the retail store, few categories are more complex than wine. For consumers, there’s an array of varietals from all over the world, each sold at wide-ranging — and to the consumer — seemingly random price points. For brands, there’s getting those bottles to the shelf, working through state-by-state regulations, distributor needs, and promotional limitations. For retailers, it’s knowing what to stock.

Using emerging technology, wine brands can simplify the process through AI-powered, data-driven optimization and decision-making. Producers can leverage predictive analytics to harmonize data from retailers, distributors, third-party sources, POS data and more. This includes, but is not limited to, macroeconomic and other influencing data that tee up strategic recommendations as granular as each store in their network.

As a result, wine brands can gain more control over where their wine goes and develop stronger relationships with their distributor and retailer partners, in addition to better managing supply and demand for future planning.

Understanding wine’s journey to the store

Currently, wine brands, big and small, need to jump through plenty of hoops to get products inside retailer doors throughout the country. To start, there’s a litany of governmental regulations to comply with, particularly regarding advertising and labeling wines, and how it’s imported and sold in certain states.

Brands must adjust their strategies on a state-by-state basis, factoring in unique rules before selling into retailers through a distributor network. Wine companies often work with several distributors, each with their own regional expertise and connections. However, distributors don’t just buy wine from a producer and sell it to the retailer. Distributors build a relationship with manufacturers to leverage sales materials, displays, and most of all, data and insights to help tell a story to the retailers.

Combined, regulations and distributor requirements can complicate how a wine company moves bottles of wine. Historically, brands have leveraged a matrix for each target state, distributor and retailer, and the data can get quite dizzying. Additionally, remaining inventory and pull-through rates have been used as performance indicators, even though those often do not reflect real preferences. Rather, they could reflect default performance through pushes to reduce latent inventory, or an incentive for display. This is where AI and predictive analytics can infuse levity and strategy. 

Cutting through the complexity of the category 

With so many entities at play and data sources from retailers, distributors, third-party providers and much more, brands can leverage AI to harmonize the data and deliver recommended strategies on what wines will sell best in what regions. In addition, AI can be modeled by brand teams to factor in state regulations and insights such as product availability within the distributor networks. 

Machine learning models can look at data to see how one varietal is performing in a region in Ohio and how another varietal is performing in Texas. The AI can go even more granular to see how tastes change at individual stores or clusters in one region, too. This is tremendously important with store locations reflecting distinct differences in preferences depending on the location’s customer profiles. Retailers want to know distributors and wine brands will stock products that fit the tastes and demographics of the shopper attributes at each store. 

Another advantage AI brings to the wine category is the ability to narrow in where choice may seem endless. Machine learning recommendations can help distill assortments down to the most essential for each store, presenting a measurable value to retail and distributor partners.

This pertains to pricing, too. AI can assist brands with recommendations on optimal price points for their wines by retailer. Value brands can fall under $10, for example, while premium brands can easily be $50 and up. AI can find and explain the price in that range that will deliver the best profit gains for a retailer and a brand.

Serving brands precise recommendations

To be sure, some industries fear LLM (large language model) AI outputs, questioning the accuracy and seeing a risk in its lack of explainability; however, AI-powered insights (different from LLM AI) should be looked at purely as an asset for wine brands. 

With AI/ML, brand managers can input complex data and work with it to serve reliable recommendations to explore. Brand teams can plan endless scenarios for varietals, stores and regions and make strategic decisions in near real time that are not entirely reliant on historical data.

For wine brands, predictive analytics can enhance how the category delivers the right bottles to the right shelves, meeting regulatory compliance. Brands, retailers and distributors can work better together, using a better foundation as a starting point and further refine through execution.

Consumers will likewise respond to assortments aligned to their preferences, as retailers create efficiencies and consumer delight within their existing spaces. Simply put, machine learning and predictive analytics uncorks opportunities for wine brands that they’ve never seen before. 

__

Capri Brixey is EVP of strategy consulting at Insite AI, bringing extensive strategic leadership experience from both retail and supplier roles in the consumer goods industry. Most recently with The Coca-Cola Company, Capri has led small and large-store teams, across multiple routes to market, channels, categories, and segments in the industry. Recognized in 2022 as a Senior Level Leader for Top Women in Convenience, Capri has also been recognized for her leadership in collaborative/joint business planning with top retailers across multiple channels/formats. 

ck unprecedented opportunities.

Decoding AI for CPGs: A Path to Category Management Success

Hosted by the Category Management Association

Curious about integrating AI into your category management practices? Join us for this panel discussion with retail industry veterans and former category and sales leaders at Coca-Cola, Walmart and Nike as they discuss AI adoption in the CPG world.

Our panelists will explore critical topics such as generative AI, strategic starting points on your AI journey, and the nuances of outsourcing AI solutions. Equip yourself with the knowledge to thrive in an AI-driven marketplace and stay ahead of the curve.

• Identify the best opportunities for AI integration in your category management practices
• What to look for in an AI partner and how to identify AI white washing
• Receive expert guidance on where and how to initiate your AI journey, tailored specifically for CPG companies.
• Benefits and challenges of outsourcing AI talent.
• Explore the potential of generative AI for CPGs

Get actionable steps and practical advice on how to execute an AI project, both with partners and gain alignment and support internally. Gain clarity and confidence in embracing AI to outpace your competitors in the dynamic CPG landscape.

Presented by:

• Capri Brixey, EVP, Strategy Consulting at Insite AI
• Kristine Joji, EVP, Strategy Consulting at Insite AI
• Marsha Shapiro, SVP of Client Solutions at Insite AI

How Retailers Can Work With Consumer Brands To Fit Their Private Label Strategies

In this StoreBrands guest post, we explore how retailers can collaborate with consumer brands using predictive analytics to customize assortments, optimize shelf sets, and predict sales trends. Private label sales are growing, and by leveraging AI-driven insights, retailers and consumer brands can work together to build assortments that align with shopper preferences, enhancing the overall shopping experience and boosting category sales.

Decoding Walmart Luminate: Leveraging Predictive Analytics for Success

Anything new takes some getting used to, and a new data engine is no different — especially one as robust as Walmart Luminate.

In March, Walmart will officially transition consumer goods partners into its tiered Walmart Luminate data platform, and supplier partners will need to adjust to this next-generation program. As I wrote earlier, Walmart Luminate is truly game-changing, providing an unprecedented look into the shopping behaviors of more than 140 million households, plus never-before-shared online pickup and delivery data. With Walmart Luminate, brands have access to a level of reporting to grow their businesses in ways they never could before.

Of course, managing and understanding data as complex and revolutionary as what’s inside Walmart Luminate takes time to grasp. AI and predictive analytics can support brands by harmonizing the data and turning findings into actionable outcomes.

Here’s a look at how predictive analytics and our CPG-tailored AI expertise can assist brands using Walmart Luminate.

Supporting brands using Walmart Luminate

Walmart began rolling out Walmart Luminate to brands this fall, providing some CPGs a trial in how they might to use the data, particularly the Charter version of the program (see below summary of the data plans).

Walmart Luminate Charter is the paid tier. It gives brands complete access to shopper behavior data from millions of households, including loyalty data and custom reports on brand-switching behaviors. The full package also debuts behavioral insights on pickup and delivery, and shares robust scoring and recommendations from the Supplier Quality Excellence Program (SQEP). As part of SQEP, all Walmart suppliers are scored based on the condition of packages, pallets and products shipped to their distribution centers, for example. Charter delivers in-depth metrics on how the supplier is performing in those areas. It also has historical data and custom reporting capabilities that’s not available in the basic versions.

Brands that pay for the full service receive an exhaustive and extraordinary set of data that can seem overwhelming at first. For that reason, some CPGs will need assistance sorting it out. Insite AI’s services range from data integration and harmonization all the way to predictive analytics and scenario modeling meant to support each supplier given where you are on your individual AI journey.

Here are four ways predictive analytics and our team of engineers can help simplify the data and get more out of it:

Linking Walmart Luminate to other data-management programs.

Some CPGs may want to integrate Walmart Luminate into other data tools that they’re already comfortable working with, and our engineers can help connect the solutions. For example, a brand may want to continue using Power BI, the data visualization platform from Microsoft, but not know how to integrate Walmart Luminate into the solution. Our team can export the data and insights from Walmart Luminate and link it to a program like Power BI. Brands can get the most out of Walmart Luminate without needing to upend their data analysis habits.

Harmonizing Walmart Luminate insights with additional data sources.

Walmart Luminate provides unprecedented access to customer segmentation data and e-commerce insights, but brands still want to leverage findings from other sources such as third-party panel data, programs like 84.51, macroeconomic data and other external sources. Insite AI’s AI modeling can take the data and harmonize it into one single version of the truth. No matter the source, it’s all big data that feeds into the AI, and our solution harmonizes the information and delivers insights in a format that’s meaningful and easy to understand.

Recommending strategic actions based on the Walmart Luminate insights.

With Large Language Models (LLMS) steeped in retail knowledge, our CPG-focused AI-powered solutions and services can help explain the data. Predictive analytics and AI models can take Walmart Luminate data, create accurate forecasts on a product’s demand and explain why sales may spike or decline in the coming months. Perhaps a rise in gas prices will increase online sales of a product? AI can add a layer of “explainability” to the robust Walmart Luminate data stream. By uncovering the in-depth meaning behind the insights, you can see the trajectory of your business and discover talking points for sales and merchant teams to help drive the business forward.

Enterprise solutions tools that solve business needs within revenue growth management, assortment planning/modular planning, macro/micro space optimization, pricing, pack architecture, forecasting and strategic business planning. By leveraging AI/ML powered solutions, brands can set and visualize multiple goals and obtain proactive and prescriptive recommendations with data explainability. These solutions are embedded within your cloud so your data never leaves the safety of your environment.  Let’s say you are wanting to run TPAs (Temporary Price Adjustments) across a few brands or items to see if it will generate a lift in volume and revenue and if that lift will be maintained if you decide to make that your new retail. Our enterprise pricing tool integrates Luminate data alongside other internal and external disparate data sources and utilizes our price elasticities model, using forecasted vs historical data, generating the ideal retail price points to drive the goals of increased volume and revenue.

Walmart Luminate will be a force in 2024 as brands take advantage of the data to learn more about their shoppers and product performance online and in stores. Leveraging AI and predictive analytics, consumer goods companies can harmonize the data and get insights explained.

Whether it’s bringing expertise to sync Walmart Luminate to other internal data solutions or tailoring predictive analytics models to meet backend technical needs, we can craft custom solutions that meet a brand’s specific business goals.

Tailored to Perfection: Creating a Hyper-Local Shopping Experience with AI

Nearly 60% of consumers are more likely to shop at a store that has personalized content.

When done right, retailers can reach higher sales, volume and profit margins, as well as build customer loyalty, by localizing assortments to each store’s shopper preferences and needs. AI-powered insights will help retailers forecast demand around what shoppers are buying by each store beyond just a zip code, and identify which items will perform best, when and at which stores. CPGs that bring these insights demonstrate category thought leadership and will set themselves apart from their competition.

About the Author

Kristine Joji serves as EVP of Strategy Consulting for Insite AI. She spent 20 years at Walmart where she was recognized as a visionary leader playing a pivotal role in optimizing Walmart’s merchandising strategies.

Four Trends That Will Drive CPGs To Adopt AI in 2024

While food retailers have been actively utilizing AI to forecast shopper behavior and streamline supply chains, the adoption rate among consumer brands lags behind. In this article, we discuss four key trends that will drive CPGs to adopt AI in 2024 and beyond.

About the Author:

Brooke Hodierne serves as EVP of strategy consulting for Insite AI. She previously worked at 7-Eleven as SVP of merchandising for the leading c-store. Before joining 7-Eleven, she held multiple positions at Giant Eagle, notably as VP of own brands.

Food Navigator: Insite AI Harmonizes Retail Data Into Actionable Solutions for Brands

Insite AI leverages artificial intelligence and predictive analytics to identify actionable recommendations for brands from data insights, like those provided by Walmart’s recently upgraded Luminate data platform, which tracks the behaviors of its 180 million weekly shoppers and provides suppliers with insights into their business, including customer behavior, product performance and category trends, Kristine Joji, EVP strategy consulting, Insite AI told Food Navigator-USA.

About the Author

Kristine is a highly accomplished retail executive and former VP of Merchandising at Walmart.  Kristine led strategic initiatives that resulted in substantial revenue growth for the company across Grocery and prior to that Personal Care.  Widely recognized as a visionary leader, she played a pivotal role in optimizing Walmart’s merchandising with large CPGs.

More than Halloween: Maximizing consumer trends in candy

A recent Advantage Solutions survey of Halloween shopping habits found that more than a third of shoppers reported price as their main purchase driver—the top ranking influence within the survey.

In this article, Brooke Hodierne (former SVP of Merchandising at 7-Eleven), shares recommendations for candy brands can make more informed pricing decisions, maximize their presence in stores, and collaborate with retailers to foster continued growth in the candy category beyond Halloween.

View full article featured on Candy Industry.

About the Author

Brooke Hodierne serves as EVP of strategy consulting for Insite AI. She previously worked at 7-Eleven as SVP of merchandising for the leading c-store. Before joining 7-Eleven, she held multiple positions at Giant Eagle, notably as VP of own brands.