Pack testing is a vital tool for companies in the packaged goods sector, serving as a crucial step in understanding how consumers will react to new packaging designs before they hit the shelves. As we move through 2024, several emerging challenges are reshaping the landscape of pack testing, impacting how brands forecast consumer response and market success.
Using Predictive AI in Advertising Research
Predictive AI is a cutting-edge approach that merges academic knowledge from neuroscience with machine learning to forecast consumer viewing behaviour. Trained on a vast database of thousands of ads, this technology is capable of predicting where viewers are likely to focus their attention in both static and video ads. The outputs from Predictive AI include heatmaps that visually represent areas of interest and quantitative KPIs that offer deeper insights into ad effectiveness.
One of the primary KPIs generated by Predictive AI is Cognitive Demand. This metric measures the amount of human processing power required to understand an ad, gauging the level of visual information viewers need to process. The second KPI, Focus, assesses the degree of concentrated attention a viewer maintains while engaging with an ad. Ads cluttered with multiple attention-drawing elements tend to diffuse viewer focus, whereas a more streamlined design can enhance viewer engagement.
Validating Predictive AI: A Rigorous Approach
At DVJ, we don’t just blindly adopt new techniques but instead, always validate their added value. This principle guided our approach to Predictive AI as well. To learn whether Predictive AI adds to our capabilities, we applied the model to hundreds of campaigns we have tested in the past three years. These campaigns were not limited to ad evaluations but also included assessments of campaign effects, enabling us to assess the Predictive AI output’s ability to forecast in-market performance.
Across the different KPIs measured through Predictive AI, one stood out in our validation process: Cognitive Demand emerged as a significant predictor of an ad’s in-market breakthrough ability. Of course, factors such as media spend, competitor activity, and seasonality influence campaign effect, but Cognitive Demand adds a unique explanatory dimension.
The Inverted U-Shape Relationship
Interestingly, our analysis revealed a nuanced relationship between Cognitive Demand and ad performance. Rather than a straightforward linear connection, we discovered an inverted U-shape relationship. This means that ads with a moderate level of cognitive demand—neither too simplistic nor overly complex—tend to perform the best in capturing viewer attention and achieving in-market breakthroughs.
Ads that are too easy to process may fail to engage viewers, leading to lower impact and recall. Conversely, ads that are overly demanding on cognitive resources can overwhelm viewers, causing them to lose interest. This suggests that great ads ensure their storyline and video cues fall within the optimal bandwidth of cognitive demand to engage viewers. This optimal range strikes a balance, ensuring that the ad is engaging enough to hold attention without being excessively taxing.
Implications for Advertisers
The insights derived from our exploration of Predictive AI have significant implications for advertisers. By understanding the optimal cognitive demand for their ads, marketers can design campaigns that maximise viewer engagement and market impact. Predictive AI provides a powerful tool for fine-tuning ad content, enabling advertisers to create visually appealing and cognitively balanced ads that resonate with their target audience.
Moreover, the combination of Predictive AI with survey-based methods offers a holistic approach to ad testing. While our existing solution is already able to assess the breakthrough and brand impact of the ad overall, Predictive AI helps to quantify how difficult different scenes within the ad are to process for viewers. This integrated approach enhances the accuracy and depth of our ad evaluations, ensuring that our clients receive comprehensive and actionable insights.
Case Study: Optimising Ad Effectiveness with Predictive AI
Recently, we tested an ad that effectively conveyed the brand but struggled to deliver its core message. Using Predictive AI we were able to provide deeper insights into the underlying cause.
Using Predictive AI, we discovered that the Cognitive Demand of the ad spiked significantly just as the key message was introduced. At this point, there was an overload of both visual and audio cues. This high cognitive load at a crucial moment overwhelmed viewers, making it difficult for them to process the message amidst the other visual and audio cues.
Based on these findings, our recommendation was to simplify the scene where the message was presented. By reducing the number of visual and audio elements, we aimed to lower the Cognitive Demand, allowing the message to stand out more clearly and be better processed by viewers.
This case study demonstrates how Predictive AI can pinpoint specific areas within an ad that need adjustment, providing actionable insights to enhance ad effectiveness. By integrating these advanced techniques, we can ensure that ads not only capture attention but also communicate their intended messages more effectively.
