Predicting the potential of pack designs to grow in-market sales

By Sheryl Brie

The classic fairy tale of "Snow White" features a magical mirror that can answer any question it is asked. The mirror is similar to artificial intelligence (AI), which can also be used to provide answers to a wide range of questions and topics. And just like the magic mirror, AI tools can be used for good or for evil, depending on how it is being utilized. 

One of the most famous lines from the fairy tale is when the evil queen asks the mirror, "Mirror, mirror on the wall, who is the fairest of them all?" The mirror responds time and time again, "You, my queen, are the fairest of them all." The queen is still not satisfied and becomes jealous of Snow White, leading to the story's main conflict. 

The power of AI insights 

Like the magic mirror, when used incorrectly, AI insights can be used to provide answers that people want to hear. The issue with AI insight is that it can only provide the answers that it is programmed to give or the ones that it learns over time. AI does not have its own thoughts or opinions, and it cannot make decisions or take actions on its own. 

The ultimate responsibility for the actions taken as a result of using AI insights lies with the person using them. Just as the evil queen in "Snow White" chose to use the magic mirror for her own vain purposes, it is up to the human user to decide how to use the insights, what data to utilize to program the technology and what actions to take based on the information provided.

Growing in-market sales: AI and better pack designs

Last year, we launched our proprietary PackPower Score™, a simple one-number metric powered by AI that predicts the potential of pack designs to grow in-market sales for brands and products. The Score uses a neural network model to train a computer to see images as a human would, analyzing the pixel-level details of shapes, sizes, colors and more in order to find patterns between the visual and the data.

In addition to its AI capabilities, the Score also uses the industry’s leading behavioral pack database with 32 million data points to collect and analyze data across 42 different measures for all PackFlash® online pack design studies. The product also incorporates sales data from IRI including 156 weeks of historical sales data that has been standardized and normalized across categories and time frames.

Like the magic mirror, the PackPower Score is a powerful tool that will provide an honest, unbiased answer based on the data and machine learning capabilities. How one utilizes the results depends on the professional so they may confidently move forward with the right pack design into market.

One of the benefits of the PackPower Score is that it can be infused into our full product stack including our recently announced PackPath™ suite of qualitative services to help marketers develop more effective pack designs to own the most valuable moment in marketing: the sales transaction. PackPath combines proven, tried-and-true qualitative methods with the latest innovations to bring insights to the next level. By using PackPower Score and our qualitative tools at each step of the PackPath process, we can guide our clients towards success with confidence and ease.

The PackPath process consists of three steps: On your Mark! Get Set!! and Go!!!

  1. On your Mark! This is the start of the journey, where we aim to understand the current category and the competition. We use focus groups and shelf context to understand the perceptions, values and usage of the products in the category. We also infuse PackPower Scores to establish a baseline, which helps inform the design brief that is created for the design agency.
  2. Get Set!! During this phase, we test and learn about the design territories that are being explored. We see which territories make the most sense for the client's strategy and how shoppers feel about them. We also run preliminary PackCepts through PackPower Score to help guide decision-making.
  3. Go!!! In this final phase, we bring in three-to-four design options into the category and understand how they perform in a shopping scenario. We receive PackPower Scores on each option, which helps inform the final decision on which design to pursue.

Overall, the PackPath process is designed to make the packaging design development process as easy and straightforward as possible so clients can succeed. Research can be conducted among consumers either in-person, in one of our ShopperLab® facilities, online or even the metaverse. With a strong foundation and a focus on innovation, PackPath allows our clients to soar with confidence when making impactful business decisions. Like the magic mirror, PackPath infused with the PackPower Score provide answers to help guide marketers – not for the purpose of evil but to improve and design better packaging.

To learn more about PackPath and how to infuse AI technology into your qualitative methodologies, please visit or contact us at

About the author 

Sheryl Brie is vice president, head of behavioral qualitative, at Behaviorally. With over two decades in market research and spending her recent 14 years working in CPG shopping and packaging insights, Brie has conducted over 500 research projects and has spoken to more than 10,000 individuals about their shopping decision process. She is a behavioral science geek and is fascinated with how human perceptions filter their world and guide their choices and behaviors. Brie is also a recent empty nester with two adult children. Her daughter is following in her footsteps and is also in the market research industry. Her son is a U.S. Marine. 

Connect with Sheryl Brie on LinkedIn.