This November, we saw boundary-pushing AI-generated ads, new insights on how to improve campaign effectiveness, the continued investment in ID-less solutions, the potential risks of running out of data, a retrospective on banner ads, and shifting marketing budgets to reflect a flattened funnel
One of the buzziest stories from the month was Coca-Cola’s launch of three new AI-generated ads inspired by their 1995 “Holidays Are Coming” campaign. The ads were produced using a variety of AI models to accomplish the innovative feat. They are also part of a wider GenAI experience where consumers scan a QR code on a can which takes them to the “Create Real Magic” site, featuring a virtual Santa Claus and allows them to generate holiday post cards and share holiday images on social media.
Coca-Cola has a history of pushing boundaries with ad tech, in 1993 they used computer-generated imagery (CGI) to create their now iconic holiday polar bears, and, more recently, the brand has notably been on the forefront of integrating AI into marketing strategy. Ad Age spoke to Coca-Cola’s VP and global head of generative AI, Pratik Thakar, on the strategy behind the commercial, where he noted, “We keep our roots in our heritage and what Coke is all about as a brand, but then connect the dots with the future and technology, […] and that was kind of a starting point.”
The ads represent a significant milestone in the GenAI evolution. However, there are some skeptics. Some have stated that the ads, no matter how much GenAI technology has improved, still feel cheap and have the undesirable “uncanny valley” effect. But the fact of the matter is, production was cheap, and fast. As Secret Level Founder Jason Zada described to Ad Age on if the commercials had been shot live, “It would be, you know, several million dollars and a lot of time in the cold […] and we were able to do all of that, you know, from the comfort of everyone’s home, we have global artists all over the world that we worked with.” Furthermore, Thakar noted in the article that these ads were produced five times faster than average––which is a massive benefit. “And then you can do more, more variety, and more customized and more personalized,” Thakar said. “And that’s the way to go, with resources, rather than doing less and spending less.”
Cost savings and budget efficiency are, of course, appealing to every company. However, brands should keep quality content at the center of their strategies even if it means spending a bit more. As Forrester VP, Principal Analyst, Jay Pattisall notes “By all accounts, some agencies leveraging brand AI systems are removing 25% or more of the costs for building campaigns. But fast, cheap content creation is worthless if not manifested as effective, engaging content outputs.”
Though a machine technically created these ads, it’s important to note that a human touch was still integral to the process. As Zada described, “I think a lot of people feel like you just press a button, and you get something like that [Coca-Cola commercial] out, […] And I think that it’s so much of the human side of it that makes that warmth that you see in that spot.” A partnership between human intuition and AI automation, which Forrester has dubbed “Intelligent Creativity”, is necessary for success in this space as audiences are generally wary of AI generated content–-when they are aware that AI was involved. When tested among consumers who did not have AI context, the ads were rated 5.9 out of 6. However, for those who were aware of AI involvement, the ads sparked controversy. The solution, though, as Pattisall writes, is not to omit disclosure of AI involvement, as once brands start down this path “it becomes more susceptible to abuse and misuse, contributing to the already eroding trust of the 21st-century digital media environment.”
While the dialogue on the topic of employing AI in a creative capacity is ongoing, it remains important to keep human connection at the center. Data Axle’s own Tom Zawacki recently spoke to Forbes on how he believes brands benefit from emotionality and overreliance on AI may lose sight of that.
As marketers, we’re always looking for ways to glean insights on ad performance and how to improve our content. MediaPost covered Meta’s recent research on how intimacy and immediacy in ad content improves performance. The research, conducted on 13 billion consumer impressions over five countries and across three continents on Facebook and Instagram, showed that designing ads in a way that feels intimate and immediately recallable leads to major boosts in performance.
They defined intimacy in two ways: first, having a human connection in the ad, and second, having seamlessly integrated brand and product placement (the key word being seamless, having the products be a natural fit for the story was crucial). When human connection was featured prominently in the ad, they saw an 81% increase in effectiveness, but only 31% of campaigns analyzed used this tactic. For seamless brand and product integration for higher engagement, they observed a 46% increase in effectiveness.
As for immediacy, the definition was split into visual dynamism and distinctive atmosphere. Visual dynamism refers to the ad involving animation, quick shot changes, animated supers and pacing, and overall creating an energy with the arrangement of visual elements. Only 37% of campaigns utilized this element, however, the research revealed a 74% increase in effectiveness when deployed. For distinctive atmosphere, Meta defined this as having a distinct color palette, music, or use of semiotic storytelling (non-verbal symbols and cues to convey meaning). This strategy was more frequently used, and saw increases in effectiveness both in the short term (67%) and long term (19%).
Overall, the research demonstrated that in order to create a memorable and effective ad, marketers should prioritize creating a distinct and compelling atmosphere, employing human connection where it will pack a punch, and telling a story that’s a natural fit for the brand and product.
It’s no surprise that GenAI and ML models have seen immense growth this year. As companies and consumers alike invest in this technology, the ongoing question of how to source the necessary data to train these models is growing with importance. As noted by Forbes, experts are saying there is a 20% chance that the scaling of ML models will slow down significantly by 2040 due to lack of training data. This prediction may feel surprising, with over 5 billion images on the internet to use, how could we possibly run out of data?
Well, the issue is that not all data is created equal. High quality images, which are most useful for training ML models, are more likely to be behind paywalls and free images sourced from places like social media are more likely to have bias, rendering it less useful for AI model training. Generally speaking, most high-quality information is sitting in a walled garden, behind a paywall, which can be frustrating for both humans and machines alike. Some have suggested that a way to circumvent this issue is with synthetic data, which is data that is extrapolated from other data. However, this data is really only as good as the core data it was generated from. Another potential solution could be to digitize pre-internet texts, and use them to feed the voracious data appetites of these ML models. Overall, the problem at hand of running out of data and the questions of how we will harvest more and who will be doing the labor remain. It’s critical to consider these factors as the relationship between humans and AI evolves.
Regardless of if the cookie is going to be phased out, marketers are struggling to find reliable ways to identify and segment audiences. Many have turned to emphasizing and ramping up collection of first party data, but it’s still only part of having a fully complete data-backed picture of your audience. As alternative, companies such as Experian have begun to invest in contextual targeting, a tried-and-true traditional method. However, in 2024, contextual advertising is being supercharged by AI. Because AI is excellent at combing through data to understand what content people are engaging with, this method has become certainly more efficient. And the best part––it does not require cookies.
Ad Age recently reported on how a variety of companies are responding to the shift away from cookies and IDs and are incorporating more contextual targeting strategies. The following is a brief summary of their findings:
Hershey has been on the forefront of moving away from reliance on IDs. First party data is less valuable to them because their market does not rely as much on direct-to-consumer selling. Instead, they’re utilizing sales data and AI algorithms to direct ads where inventory needs to move faster, all without using Personally Identifiable Information (PII).
Other companies such as Clorox, Procter & Gamble, and Reckitt are investing in first-party data and contextual targeting. By putting out ads on CTV, podcasts, and other digital mediums they hope to glean insights from the type of content people are engaging with. They also are striving to work with a more diverse creator pool to reach multicultural audiences.
Boost Mobile recently partnered with GumGum, a contextual intelligence company, to assist in their marketing efforts. By using GumGum’s Mindset Graph tool, they’ve outperformed competitors by 90%, with their cost per conversion coming in at $2 or $3 versus competitors at $20. Additionally, Boost has concentrated efforts on matching its first party data with the contextual data from GumGum. As Caine Junginger, marketing lead at Boost Mobile said to Ad Age, “’I view contextual holistically as where things are going,’ Junginger said. He believes GumGum’s Mindset platform and increased precision around why people are viewing particular content as particularly promising. ‘If we can narrow it down to people on local news websites who are extremely willing to buy, then we’ll go out and target those.’”
Samba TV launched an AI product that can identify contextual signals to a granular level in a show or sports broadcast. The product can identify when certain themes or topics are being discussed, or even if there’s a relevant logo present. HP recently used this technology to quantify viewer exposure to their logo from its Real Madrid uniform sponsorship. The results of this study were favorable, which showed the potential in contextual measurements.
Raptive announced an AI-powered contextual targeting tool called Mindset Targeting that predicts emotional responses to brand ads or integrations across creator media sites. The tool aims to place ads where consumers are most likely to respond. Toyota used the tool to analyze 25 million pieces of creative content to identify key personas. The results showed 7% higher than benchmark for post-exposure ad recall. Ultimately, the goal for Raptive and other contextual technology companies is to enhance personalization without the need for identity signals. As Raptive co-founder Andrew Shue described to Ad Age, “Everyone can do contextual to some extent. […] But it’s the ability to go deep, to have nuance, that brings a different level of view into that consumer.”
In the words of Jonathan Nelson, CEO of Omnicom Digital, “Banner ads walked so generative AI could run.” Thirty years ago, the first banner ads were released, a historic moment that Data Axle’s own Tom Zawacki helped to execute. It was a revolution in digital advertising that changed the web into the commerce hub it is today with users actively engaging with ads as a way to find new information, products and services. Banner ads marked the beginning of tracking user interactions, which paved the way to performance-based marketing. Targeted advertising also found its start in this era, eventually leading us to the hyper-personalization and customization we see today. As exciting as these innovations were for marketers, consumers, however, became increasingly concerned with data collection, privacy and manipulation as a result of the new tracking and targeting methods that banner ads ushered in. Sounds familiar, doesn’t it?
With every innovation comes ethical considerations. The advent of generative AI finds many similarities with the early stages of banner ads. Both mark a seismic shift in marketing strategy; a new technology that completely disrupts the old guard. Furthermore, both technologies need immense amounts of user data in order to thrive, and use it to improve over time. The amount of data needed to train AI models has raised new concerns of privacy, consent, and potential for bias. In response, companies are creating frameworks to ensure ethical use––just as they did with banner ads in 1994.
The difference between these two technologies however, as Jonathan Nelson notes, is that banner ads were primarily used by ad tech experts, whereas GenAI is more democratized––nearly everyone has could benefit from utilizing GenAI and has the technology acumen to do so. Nelson suggests that companies encourage organization wide use of GenAI as it will create a more cohesive, AI-integrated ecosystem, as ultimately, this will better inform marketing efforts. And, as Nelson writes for AdExchanger, “when the right message is delivered to the right person at the right time, then the final ingredient will fall into place: achieving the right outcome.”
In a new report sponsored by PubMatic, 82% of brands surveyed are no longer thinking of the marketing funnel as linear. As marketers face new challenges of addressability and rapid shifts in consumer behavior, it’s becoming increasingly difficult to streamline performance marketing strategy. One shift PubMatic’s report observed is that performance campaigns are being increasingly deployed across the open web rather than staying in walled gardens. With the open web typically being brand marketing’s domain, performance’s entrance into the open web brings in the idea that branding and performance budgets are mixing more than previously thought.
There are many reasons for this heightened focus on the open web, as described in an article from Digiday, “The open web is also a source of significant innovation right now as it relates to privacy-first, ID-agnostic targeting. At the same time, the open web enables brands to lessen their dependence on walled-garden platforms and their associated risks. On the open web, brands can stand out with high-quality, relevant content, while simultaneously supporting premium publishers and outlets.”
As brands are making the shift to the open web, so too are their budgets toward investment in technology. The PubMatic report showed that in 2024, slightly less than half (49%) of brand respondents said that technology investment accounted for over 40% of their performance budgets. In 2025, that percentage of brands will grow to 55%.
Overall, these trends indicate that brands are more interested in bigger picture concepts like ROI, efficiency and sales rather than rote numerical data of clicks and conversions. As Digiday concludes, “This fact is highlighted in their merging of brand and performance budgets to mirror consumer behaviors, as well as their deepening emphasis on the flexibility and power of open web strategies. For brands that continue to embrace these shifts, an exciting future lies ahead: one where a flattened funnel and smarter investments in tech and the open web lead to not just performance gains, but true market leadership.”
Building a Trust-First Brand: Transparency and Consent in Marketing (CMS Wire) With news of data breaches seemingly increasing by the day and consumer privacy concerns at an all-time high, brands are rethinking how they source, use and store customer data. Consumer trust is a delicate and valuable entity for brands. Earning it takes time, patience and consistency. Losing it can happen in an instant. Consumers are holding brands to a higher standard, expecting transparency, respect and consent when it comes to their personal information. Today’s consumers are more data-aware, have higher standards and can easily switch brands if their expectations aren’t met. The challenge (and opportunity) for brands lies in building a long-term strategy that places trust — underpinned by data transparency and consent — at the core of marketing.
– David Hegarty, VP of Solutions Consulting
16 Big Trends In Marketing That May Soon Be On Their Way Out (Forbes) If only artificial intelligence had a heart. While generative AI boosts efficiency, it lacks the ability to create sentient connections, a crucial element in driving consumer affinity and loyalty. Research shows emotionally connected customers are 52% more valuable, driving higher long-term growth and brand advocacy. Brands must balance AI’s speed with genuine emotional engagement to create lasting impact and empathetic connections.
– Thomas Zawacki, President, Axle Agency / Chief Marketing Officer
Creating Real Connections: AI’s Role in Modern Omnichannel Strategy (CMS Wire) The customer experience (CX) landscape is undergoing a profound transformation. With consumers interacting with brands across multiple channels — social media, websites, mobile apps, emails and even physical stores — businesses are challenged to provide a cohesive and consistent experience. The expectation is that no matter where or how a customer engages with a brand, their customer journey should be seamless. This is where omnichannel customer strategies and AI come into play. AI-driven solutions are increasingly crucial in managing and enhancing the omnichannel strategy. From real-time customer engagement to predictive analytics, AI is empowering brands to craft more personalized, efficient and responsive customer experiences across every touchpoint.
Navigating the Future of B2B Tech Marketing: Enhancing Data Quality and Processes (MarTech Series) The marketing landscape is undergoing a seismic shift and B2B tech marketers must stay nimble to keep pace with change. The phase-out of third-party cookies, the introduction of IP address masking, and increasingly complex privacy laws are fundamentally reshaping how data is collected and used. In this new environment, data quality has become the cornerstone of effective marketing strategies. To gain sharper insights, improve targeting, and maximize ROI, marketers need to focus on building accurate, reliable, and unified data processes. This not only strengthens current marketing efforts but also future-proofs brands against the evolving challenges of online and offline marketing.
– Marc Sabatini, SVP of Enterprise Solutions
Data Axle Nonprofit Promotes Adam Tatro to SVP of Client Services to Drive Continued Growth for Apogee & DonorBase Today, Data Axle Nonprofit is excited to announce the appointment of Adam Tatro as Senior Vice President (SVP) of Client Services for Apogee and DonorBase, its industry-leading cooperative donor databases. Adam’s leadership will be essential in expanding the growth and innovation of the company’s successful omnichannel and email cooperatives as the company looks toward 2025 and beyond. Prior to joining Data Axle, Adam had extensive experience with Wiland both in the sales and solutions capacity. He comes with over a decade of experience developing tactical campaigns for nonprofits that help move the needle. His experience executing in multiple roles within client services and business development has given him a deep understanding of cooperative databases and their role in helping nonprofits accomplish their missions.
Courtney is a seasoned communications and public relations professional with 17+ years of experience working in both the public and private sectors in diverse leadership roles. As Data Axle’s Senior Public Relations Manager, she is intently focused on elevating the company’s media relations presence and increasing brand loyalty and awareness through landing coverage in top-tier media outlets.