predictive analytics

Emotional Drivers of Long-Term Effectiveness of YouTube Ads

Emotional Drivers of Long-Term Effectiveness of YouTube Ads

Manuel Garcia-Garcia, Ph.D. – Global Lead of Neuroscience, Ipsos

Ariane Pol – Global Head of Research for Creative Works, Google

Geraldine Rodriguez – Client Manager Applied Research, Ipsos

Can YouTube help drive long-term brand building? How do you measure long-term brand building? When brands want to air strategic long-term campaigns, they typically revert to traditional media. Most people are not in need of a brand’s immediate offering, but they represent the biggest sales opportunity. Ten years ago, the IPA demonstrated that campaigns whose primary focus was emotional were the most effective. Emotions are the fuel that allow high conversion over time. Brands should tap into emotions of consumers that may not be interested in a product now but may be relevant in the future. Ipsos partnered with Google Creative Works to study the observed and declared behaviors. Methodology: A triangulation of methods were used. They were Creative/Spark (market validated KPIs of creative impact); Ipsos Bayesian Nets (models the impact of emotion); Ipsos Emotion Framework (captures emotional responses). Ipsos Emotion Framework defines emotions as physiological changes we experience in response to the environment. These are complex emotions that are heavily driven by culture and context, and they are therefore, not universal. This complicates measuring emotions. While emotions are not universal, we can explain emotions based on valence, arousal and control. This maintains the cultural authenticity but can be compared across cultures. The experimental approach to measuring long-term brand growth included a brand relationship index (BRI), comprised of brand performance = how would you rate [brand] in terms of what you are looking for in a [category] + brand closeness = how close do you feel to [brand]? Findings:
  1. Valence alone explains 28% of variance of long-term brand sales growth for YouTube videos. Highly pleasant residual emotions on YouTube ads have predictive power over long-term brand growth. This works for both YouTube ad formats (skippable and forced).
  2. Highly pleasant YouTube ads make people willing to pay more, reducing price sensitivity.
  3. The highly pleasant emotions that correlated with valence were warmth, happiness, calmness, love, nostalgia and excitement.
  4. Empathy and surprise become important predictors of the brand relationship change index in the long term.
  5. To analyze how respondents group emotions when reporting how ads make them feel, a sophisticated analytic technique based on Bayesian network was applied. This method shows that ads can awaken different emotions, not just one emotional note. Empathy and surprise are more neutral by nature, and this can lead to either positive or negative emotions. They can be bridge emotions between negative and positive emotions.
Key takeaways:
  • Digital media like YouTube can be a prime brand building vehicle, not only for short-term tactical business objectives.
  • Highly pleasant emotions account for 28% of long-term brand growth. Brands should leverage this knowledge to create powerful, emotional storytelling to get closer to current and prospective clients.
  • Positive emotional storytelling supercharges performance. It makes people more willing to pay more for a brand.
  • Emotional storytelling doesn’t mean focusing on one single tone—brands can experiment with several emotions to create powerful and emotionally stirring narratives.

Download Presentation

Member Only Access
Determining the Value of Emotional Engagement to TV

Determining the Value of Emotional Engagement to TV

Pedro Almeida – CEO, MediaProbe

Context matters—not all reach is equal, and so, we need a way to qualify each impression and valuate each of these impressions. Metric of valuation needs to be valid, reliable and have predictive power for business outcomes. The research focus: 1) What can we say about the value of emotional engagement (EE)? 2) Can we model the value of EE via its impact on memory? 3) Can we use EE to optimize and valuate content and ad positions? How? Methodology: MediaProbe used Galvanic Skin Response with participants who were exposed to content through a MediaProbe panel (U.S., 2,700 households). Data gets delivered second by second and data extracted goes toward creating an impact measure of how much people are reacting to what they are watching. The platform calculates an impact value that enables comparisons across media platforms. There was an added layer to see whether participants are leaning into the content and are engaged. U.S. TV dataset includes over 45,000 participants, reaching over 85,000 hours. More than 1,000 TV hours are monitored and over 42,500 ads. Using a subset of 16,351 ads and 329 “premium pod” formats, participants watch content and are then asked which ads they remember. Findings:
  1. Enhancing the emotional impact of an ad in 150 EIS points equates to adding a second 30’ ad unit. This will increase probability of brand recall by 15%. For each 100 points, this increases probability of brand recall by 10%.
  2. Single best predictor of whether someone will respond to an ad is how much a person was engaged with the content prior to the ad. EE carries over to the ad break. It’s more engaging pre-break, in earlier breaks and earlier position in break, which leads to higher ad impact.
  3. However, this is different across genres. Genre moderates pre-break emotional patterns. This is further differentiated within genres. For instance, people will react differently to ad breaks when watching soccer vs. some other sport. MediaProbe shows that there is 66% similarity between various award shows in terms of EE to ad breaks. They use this data to realize the value of different ads placed in different breaks (1st, 2nd, etc. break) and pods. Emotional engagement helps better predict ads performance.
  4. Additional findings show that first-in-break still rules and that premium pods deliver higher recall.
Key takeaways:
  • Ad EIS is systematically associated with ad recall.
  • It is possible to optimize ads for estimated impact by advertising in the most engaging content and being present after the most engaging moments.
  • Different genres tend to have typical pre-break engagement morphologies. This allows to estimate the delivered value of each pod position (and order in break when relevant).

Download Presentation

Member Only Access
Neuro: TV Brand Attraction Advantage Over Digital

Neuro: TV Brand Attraction Advantage Over Digital

Bill Harvey – Executive Chairman, Bill Harvey Consulting

Elizabeth Johnson, Ph.D. – Executive Director & Senior Fellow, Wharton Neuroscience Initiative, UPenn

Michael Platt, Ph.D. – Director, Wharton Neuroscience Initiative, UPenn

Audrey Steele – EVP Sales Research Insights & Strategy, FOX Corp.

The presenters discussed their study focused on the link between attention and sales. Attention is required for engagement. Eyes on screen do not predict sales well. However, three main brain measurement dimensions account for sales and branding effects: brand attraction/joy (=motivational signals in fMRI and EEG), memory (=Theta power in EEG) and synchrony (=collective resonance across audience brains, in fMRI and EEG)—all require more than 1-2 seconds to unfold and measure. Using neuroanalysis can help unmask hidden thoughts and feelings (via fMRI). Additionally, scaled up, other predictive bio and neuro metrics can be just as predictive. The research shows that patterns of brain activity predict sales best: the sum of all perceptual, attentional, emotional, social and memory processes. We can also use EEG to tell us about frustration, attention, memory, sleep/introspection. Research using EEG shows that EEG measuring brand attraction/joy can predict 80% variance in sales. Notably, brand attraction/joy takes 15 seconds to peak. Brain memory also predicts sales. Notably, memory encoding picks up after 10 seconds. Finally, synchrony—collective audience response—predicts more than 90% of sales but also has temporal dynamics, peaks at 5 seconds and picks up again after 15-20 seconds. Wharton Neuroscience investigated predicting how different content and platform impact sales lift. The study design: eight ads in eight verticals tested in each of the 10 experimental cells (7 TV, 2 smart phones, 1 control condition). Four ads at a time are shown between TV show. Each viewer will only see one kind of content. Findings from the 3% of total sample: attraction and memory are sustained for ads shown in premium channels compared to YouTube. Value of context is enormous! YouTube has a drop at 4 seconds whereas TV continues. Key takeaways:
  • Attention is an incomplete measure by which to select media contexts and platforms for specific campaigns.
  • Premium longform content and contexts have more sales and branding impact than digital, especially in new customer growth due to emotional immersion in TV context vs. brevity of ad attention/engagement in digital.

Download Presentation

Member Only Access
Human Experience: Why Attention AI Needs Human Input

Human Experience: Why Attention AI Needs Human Input

Dr. Matthias Rothensee – CSO & Partner, eye square

Stefan Schoenherr – VP Brand and Media & Partner, eye square

Speakers Matthias Rothensee and Stefan Schoenherr of eye square discussed the need for a human element and oversight of AI. Beginning their discussion on the state of attention and AI, Matthias acknowledge that race for attention is one of the defining challenges of our time for modern marketers. He quoted author Rex Briggs, who noted the "conundrum at the heart of AI: its greatest strength can also be its greatest weakness." Matthias indicated that AI is incredibly powerful in recognizing pattern from big data sets but at the same time there are some risks attached to it (e.g., finding spurious patterns, hallucinations, etc.). Stefan examined a case study using an advertisement for the candy M&Ms, which considered real humans using eye tracking technology and compared it to results using AI. The goal was to better understand where AI is good at predicting attention and where does it still have to optimize or get better. Results from a case study indicated areas for AI improvements in terms of gaze cueing, movement, contrast, complexity and nonhuman entities (e.g., a dog). The static nature of AI (e.g., AI prediction models are often built based on static attention databases) can become a challenge when comparing dynamic attention trends. Key takeaways:
  • Predictive AI is good at replicating human attention for basic face and eye images, high-contrast scenes (e.g., probability of looking at things that stand out) and slow-paced scene cuts where AI can detect details.
  • AI seems unaware of a common phenomenon called the "cueing effect" (e.g., humans not only pay attention to people's faces but also to where they're looking), which leads to an incorrect prediction.
  • AI has difficulties deciphering scenes with fast movements (e.g., AI shows inertia) in contrast to slow-paced scenes where AI excels in replicating human feedback. In this case human feedback is more accurate.
  • AI is more consumed with attention towards contrast (e.g., in an ad featuring a runner, AI gave attention to trees surrounding the runner), whereas humans can decipher the main aspect of an image.
  • AI decomposes human faces (e.g., AI is obsessed with human ears), whereas humans can detect the focal point of a human face. In addition, AI hallucinates, underestimating facial effects.
  • AI has difficulties interpreting more complex visual layouts (e.g., complex product pack shots are misinterpreted).
  • AI is human centric and does not focus well on nonhuman entities such as a dog (e.g., in scenes where a dog was present, AI disregarded the dog altogether).
  • AI tends to be more static in nature (e.g., AI prediction models are often built based on static attention databases), which can be a problem when comparing this to dynamic attention trends.

Download Presentation

Member Only Access
Forecasting & Optimizing Reach in a PII Compliant Measurement Ecosystem

Forecasting & Optimizing Reach in a PII Compliant Measurement Ecosystem

Spencer Lambert – VP, Product & Partnership Success, datafuelX

Matthew Weinman – Sr. Director, Advanced Advertising Product Management, TelevisaUnivision

Reach and frequency planning requires access to unique viewership data, which has become increasingly restricted due to identity restrictions. However, challenges exist with panel-only measurement, including the undercounting of Hispanic and Spanish language coverage, stated Matthew Weinman (TelevisaUnivision). Panel data undercounts Hispanics audiences by upwards of 20%, even for broad demographics. The benefits of big data exist across audience planning, viewership measurement and outcomes. Excessive frequency can be limited while maintaining or expanding reach, as well as improving ROAS. However, there are barriers to working with big data, including PII compliance. Additionally, the size and scale of big data leads to lengthy ID forecast times and computing costs. Spencer Lambert (datafuelX) presented details of their approach to ID-level forecasting which included their reach and frequency clustering methodology. Key takeaways:
  • Advantages of clustering methodology over identity methodology for reach and frequency:
  • Efficiency and accuracy: Delivers comparable accuracy metrics
  • Lower error rates: Seven percent for cluster reach forecasts vs. 20% error rate on identity-scaled reach forecasts
  • Cross-platform reach and frequency: By scaling cluster assignments to digital IDs, this methodology can empower cross-platform management and optimization
  • Lower compute time and costs
  • PII compliant: Preserves the use of identity-level planning

Download Presentation

Member Only Access
The Value of Attention is Nuanced by the Size of the Brand

The Value of Attention is Nuanced by the Size of the Brand

Karen Nelson-Field, Ph.D. – CEO, Amplified Intelligence

This presentation discussed the importance of nuance and interaction effects and how understanding interaction effects are critical in building products. There were four use cases—campaign strategy, planning, verification, buying. Two sets of data—inward and outward facing—looked at tag-based data through tags, outward facing—device based, panel data, gaze tracking, pose estimation, etc. One is observed while the other is human. Both are valuable. Each set has limitations. Looking at actual humans has a scale issue, whereas impression data has limited ability to predict behavior. Human behavior is complex. It is also varied by platforms. Metrics without ground truth misses out on this. Three types of human-attention were measured: active attention (looking directly at an ad), passive attention (eyes not directly on ad), non-attention (eyes not on screen, not on ad). Attention outcomes and attention are not always related. Underneath how attention data works there is a hierarchy of attention—the way ad units and scroll speeds and other interaction effects all mediate with each other. It is not as simple as saying look at this ad unit and we will get this amount of attention. If products don’t include these factors they fail. Amplified Intelligence built a large-scale validation model for interaction effects and “choice” using Pepsi. They employed logistic regression using Maximum Likelihood Estimation (MLE), analyzing observations and tested critical factors—brand size and attention type, to demonstrate strong predictive accuracy with CV accuracy. They found significant interaction effects, particularly brand size and attention type as key influencers of consumer brand choice. Key findings:
  1. Passive and active attention work differently. Passive attention works harder for bigger brands, while active attention works harder for smaller brands. Put differently, small brands need active attention to get more brand choice outcomes.
  2. Attention switching (focus) mediates outcomes. The nature of viewing behavior mediates outcomes. Not just attention yes or no, and what level, but about behavior across time. This is why time-in-view fundamentally fails even though it is considered one of the critical measures of attention. Humans are constantly switching between attention and non-attention. There’s attention decay—how quickly attention diminishes (sustained attention x time). There’s attention volume—the number of people attentive (attentive reach x time).
  3. Eyes on brand attention is vital for outcomes. If the brand is not at the point when people are looking (or hearing), this impacts outcomes. When the brand is missing, we fill in the blanks, but the next generation of buyers are being “untrained.”
Implications:
  1. Human attention is nuanced, complicated, making it difficult to rely merely on aggregated non-human metrics for accuracy. We must constantly train these models, just like GenAI, to ensure that all these nuances are fit into the model. A human first approach is critical.
  2. Outcomes cannot predict attention. Attention can predict outcomes but not the other way around.
  3. Attention strategies should be tailored to campaign requirements (not binary quality or more/less time). Overtime attention performance segments will start to think about other AI.
Key takeaways:
  • Human attention is nuanced. This makes it difficult to rely only on aggregated non-human metrics for accuracy.
  • A human-first approach is critical.
  • Outcomes cannot predict attention.
  • Attention strategies should be tailored to campaign requirements.

Download Presentation

Member Only Access
The Power of AI for Effective Advertising in an ID-free World

The Power of AI for Effective Advertising in an ID-free World

Rachel Gantz – Managing Director, Proximic by Comscore

Amidst heightened regulations in the advertising ecosystem, Rachel Gantz of Proximic by Comscore delved into a discussion of diverse AI applications and implementation tactics, in an increasingly ID-free environment, to effectively reach audiences. Rachel highlighted signal loss as a "massive industry challenge," to provide a framework for the research she examined. She remarked that the digital advertising environment was built on ID-based audience targeting, but with the loss of this data and the increase in privacy regulations, advertisers have placed their focus on first-party and contextual targeting (which includes predictive modeling). In her discussion, she focused on the many impacts predictive AI is having on contextual targeting, in a world increasingly void of third-party data, providing results from a supporting experiment. The research aimed to understand how the performance of AI-powered ID-free audience targeting tactics compared to their ID-based counterparts. The experiment considered audience reach, cost efficiency (eCPM), in-target accuracy and inventory placement quality. Key takeaways:
  • Fifty to sixty percent of programmatic inventory has no IDs associated with it and that includes alternative IDs.
  • Specific to mobile advertising, many advertisers saw 80% of their IOS scale disappear overnight.
  • In an experiment, two groups were exposed to two simultaneous campaigns, focused on holiday shoppers. The first group (campaign A) was an ID-based audience, while the second group was an ID-free predictive audience.
    • Analyzing reach: ID-free targeting nearly doubled the advertisers’ reach, vs. the same audience, with ID-based tactics.
    • Results from cost efficiency (eCPM): ID-free AI-powered contextual audiences saw 32% lower eCPMs than ID-based counterparts.
    • In-target rate results: Significant accuracy was confirmed (84%) when validating if users reached with the ID-free audience matched the targeting criteria.
    • Inventory placement quality: ID-free audience ads appeared on higher quality inventory, compared to the same ID-based audience (ID-free 27% vs. ID-based 21%).

Download Presentation

Member Only Access
The Seven Answers Every CMO Should Demand

The Seven Answers Every CMO Should Demand

On January 23, Ken Roberts, Founder and President of Forethought™, Mary Wilson Avant, Senior Consultant of Forethought™ and Elizabeth Windram, EVP, Marketing & Communications of CLEAR discussed the seven answers every CMO should demand, ranging from predicting change in market share, creating internal alignment between marketing and the rest of the organization, how generative AI can be leveraged, and more. Currently this content is only available to event attendees.

Member Only Access
 Next »