Adtech has been in the AI and automation business for far longer than the current AI revolution. The use of LLMs (large language models) to enable the optimization and maximization of ROAS for programmatic advertising predates generative AI by many years.
Still, as the technology becomes more advanced, it’s important to pause and really understand the ramifications of LLMs in adtech, particularly for creating a competitive, equitable and inclusive advertising ecosystem. While the technology is ubiquitous, it’s not benefiting every tier of the adtech industry the same; which means smaller brands need to get creative.
LLMs hold a world of possibility for the industry, but the overreliance upon them poses significant risk to the industry at large, and to middle-tier brands in particular. By understanding their potential and their broader impact, we can fully grasp the reality of how brands of all sizes can remain competitive in this new frontier.
How Adtech is Currently Leveraging LLMs
There are a number of ways that LLMs, AI and automation have been crucial to the adtech space over the past few years, especially with the development of generative AI. This technology is used to personalize content at scale and generate content for marketing campaigns and informational pages. It’s also key for recommendation engines and chatbots that can assist customers quickly and, again, at scale.
Advertisement
Beyond the creative side of generative AI, this technology also can be used for predictive analytics to forecast future trends and ad efficacy, which means more efficient and intelligent spending. It’s also been instrumental in audience targeting and segmentation, creating more focused and nuanced customer groups for more effective messaging and conversion.
The use of LLMs is broadly beneficial, but there are certain applications that create a vast imbalance within the adtech space that can be detrimental to the industry as a whole over time, if unchecked.
How Data Aggregation Puts Mid-Tier Brands at a Disadvantage
In short, LLMs, AI, ML and automation are engines powered by data. The more data you have, the more powerful your engine is. Large brands and advertisers have stores of actionable first-party data that can be used for everything from complex predictive analytics to personalization efforts and, perhaps most importantly, targeting precision and bidding strategies.
As a mid-tier brand, you’re saddled with a much weaker engine. The lack of data — from market and sales analysis to first-party consumer data — makes it much harder to compete on a number of fronts.
First and foremost, personalization is much more difficult to accomplish at scale because there’s a general lack of clarity around who the target audience is. The inability to identify audiences means putting forth a more broad-based approach to messaging, which can be less effective.
Next, as bidding becomes more automated and algorithm-driven, smaller advertisers may find themselves priced out of the market or unable to compete effectively for premium ad placements, limiting their visibility and reach. The use of LLMs for bidding in programmatic advertising is perhaps the key sticking point that could have broader industrywide ramifications over time.
The Challenges of Bid Inflation
The increased influence of LLM-driven targeting can lead to increased competition and higher bid prices across the board. This increased bidding competition for niche audiences drives up advertising costs, making it harder for advertisers with limited budgets to compete.
Generally speaking, this bid inflation has a much more pronounced impact on small and mid-sized advertisers. There are two paths in this current market: either bid high for niche audiences, which may result in inefficient allocation of advertising budgets and decreased ROI, or focus more on broad-based advertising, which can have less impact.
The Implications of Automated Targeting on Broad Reach Media
LLM-driven bidding impacts mid-tier advertisers, but it also has a broader effect on the media landscape. Because this brand of targeting favors precision over broad reach campaigns, traditional media channels like television and print are likely to see declining effectiveness as advertisers shift toward hyper-targeted digital ads.
On the advertising side, advertisers may end up paying more to reach smaller, fragmented audiences, diminishing the value of their advertising investments. And on the media side, this focused spend on niche, hyper-targeted ads can have broader impacts on the long-time viability of legacy media channels as well as give rise to highly fragmented audiences.
Why Automated AdTech is a Risky Long-Term Strategy
In a highly competitive environment like advertising, it can be easy to say, “Well, if mid-tier brands want to survive, they need to compete at the same level as the big players.”
If only big brands are able to leverage LLMs to the level of effectiveness that the technology promises, it may create an unlevel playing field for smaller brands; which has broader industry implications. The loss of mid-tier brands signals a lack of stability within the industry. It also takes away potential acquisition targets for larger brands.
Still, it’s naive to expect larger brands to change their course of action to hamstring their own efforts in service of the competition. So how should small and mid-tier advertisers proceed?
Understanding that they don’t have the resources — both financially and from a data perspective — means turning inward in a sense. By focusing on brand equity, customer experience and building great products, upstart brands can build the organic foundation to compete in areas where incumbent brands can leverage data and automation.
Additionally, smaller brands need to recognize that they can’t do it alone in the same way larger advertisers can. By forging relationships with agencies or third-party tech partners, upstart brands can leverage similar tools without needing to build the infrastructure to get it off the ground.
A healthy adtech ecosystem will cater to the needs of big brands, small brands and everything in between. Unfortunately, as LLMs and automation have become the tech du jour for the entire industry, too many are jumping in headfirst without truly assessing the ramifications of this new technology. And while larger brands stand to gain the most, it may come at the expense of smaller brands that will need to adapt or die.
Mark Zamuner currently serves as President of Juice Media, a data-driven omnichannel media activation platform he founded in 2020 that was acquired by Altice USA in 2022. Juice Media combines growth strategy, omnichannel media buying and analytics services with proprietary ad tech focused on audience identification, targeting, and attribution and optimization. Zamuner started his career leading marquee brand accounts at Ocean Media before joining the eHarmony marketing leadership team. In 2011, Zamuner founded TWO NIL, a leading independent growth consultancy designed to fill the gap between strategic consulting and effective execution of media investments. TWO NIL’s unique and award-winning approach had outsized impact for clients including unicorn companies Blue Apron, Wix, Zillow, Groupon, Dollar Shave Club and 23andMe.
