Generative AI in Creative Management: An Opportunity for Advertisers and AdTech Vendors

By Maria Novikova On May 19, 2023

Advertisers are so focused on finding the right inventory, reaching target segments, and choosing optimal channels, all the while keeping costs at bay, that the quality of advertising creatives often becomes an afterthought.

Yet, this approach can hold brands back from hitting benchmarks. In fact, research by KANTAR shows that a high-quality creative can improve the performance of an ad campaign by 12 times and generate a 30% increase in ROI.

Leveraging the power of machine learning can help agencies and advertisers streamline their creative workflows without shifting the focus from other operations. In this post, I will explore the applications of generative AI in creative management, its benefits, and implementation challenges for AdTech vendors.

Why brands struggle to create top-notch creatives

Here’s a quick snapshot of the challenges that prevent creative teams from maximizing the impact of their brand assets.

Lack of in-house resources for manual DCO. Having little time, talent, or budget to fuel a consistent creative production engine is a top-of-the-mind concern for 66% of creative teams. Since producing multiple versions of the same ad is too time-consuming, teams struggled to go all-in on DCO – many still prioritize A/B testing.
Time and budget constraints. According to data from Nielsen Ad Intel, the advertising market in the US has shrunk by 7%. As they cut back on advertising, brands tend to downsize creative teams and budgets. As a result, marketers and agencies are pressured to prioritize other aspects of their campaigns.
Responses to creatives vary by audience type. There’s no one-size-fits-all approach to creating a high-performing ad. Creative teams need to have a granular understanding of their target segments in order to create an appealing message – according to IBM, that’s the case only for 34% of advertisers. Without a solid grasp of their audience’s preferences and behaviors, teams struggle to produce conversion-generating creatives.

How generative AI can help teams overcome challenges in ad creative design

Bootstrapped and overloaded, creative teams don’t have the capacity to spend a lot of time on fine-tuning their creatives. With advertisers and agencies falling short on resources, technology is becoming crucial in facilitating tasks like testing visuals and copy or optimizing ad sizes.Generative AI, in particular, is proving to be extremely useful in multiple areas of creative management.

Ad testing. Machine learning and data science can enable advertisers and agencies to optimize their campaigns more efficiently. For example, Ad-Lib.io uses machine learning to automate the production, testing, and optimization of ad creatives, allowing marketers to quickly and easily generate a variety of ads across multiple channels.
Producing copy and visuals. AI-based tools can help advertisers and agencies automate the process of finding the right copy or choosing a compelling visual for their next ad. Phrasee, for example, helps create powerful copy for creatives by analyzing historical campaign data and coming up with ideas that resonate with audiences.
DCO. AI tools allow marketers to quickly and easily generate ad variations tailored to different audience segments. io, one of the leaders in the field, uses AI to build ad creatives based on a wide range of variables, including audience demographics, location, and behavior.

Why creative management platform vendors should add AI to their offerings

Here’s why I believe that introducing AI to AdTech/MarTech platforms is an intelligent and future-oriented decision.

High demand from marketers and advertisers. According to Statista, AI in the advertising market will reach over $107 billion by 2028, which also proves consistent demand for AI tools. Thus, the early adopters of AI technology can gain a significant competitive advantage.
Ability to support clients in optimizing production costs. Amidst a recession, advertising teams are likely to choose tools based on their cost reduction potential. AI solutions can help drive efficiency by allowing teams to generate multiple ad variations tailored to their audiences at low costs
Media and investor interest. Adding AI capabilities to their solutions can be a push for AdTech vendors toward media and investor interest. For example, Zeta Global, a MarTech company, known for ambitious bets on AI, like the acquisition of Temnos, an AI-based language recognition platform, is successfully differentiating itself as a frontrunner in creative management.

Challenges of introducing machine learning to AdTech projects

Although the applications of machine learning in creative generation and DCO are straightforward, teams often struggle to find the right place for AI in their offerings and make sure that models bring tangible value.

Here are a few challenges I’ve seen AdTech projects face in AI implementation:

Risks of misinformation Large generative AI models like GPT and DALL-E can create copy and visuals for an ad in seconds – but the outputs of these algorithms are error-prone. At the time of writing, many cases of AI generating ads with wrong prices have been reported. Thus, AdTech vendors need to develop strategies for fact-checking AI-generated creatives before their clients include those in a campaign.
Latency of AI models gets in the way of performance. Introducing AI to DCO platforms is challenging due to the need for low latency in AdTech projects. Complex ML models are computationally intensive, so adding them to DCO and creative management platforms increases response times and reduces performance.
High cost of deployment can become a hurdle for AI implementation if the tech team has little experience in diagnosing the root causes of high costs behind training AI models and no plan for cost reduction (autoscaling CPU, automating AWS instances, etc).

Having access to machine learning engineers on the in-house team is a solid starting point for introducing generative AI to DCO platforms or CMPs. However, due to the technical complexity of ML models and the high specificity of AdTech as a domain, it’s a lot more helpful to form a team of experienced engineers focused solely on AI implementation.

Expanding machine learning talent is a strategic priority

Hiring a skilled tech partner can be essential for AdTech vendors looking to add generative AI to their solutions. Here’s how working with a team of ML engineers can protect AdTech vendors from judgment errors in mapping out AI strategies.

Access to specialized expertise: joining forces with a tech partner allows vendors to build specialized models (for example, algorithms that help identify attention-catchers in creatives, predict the success of individual assets, and optimize creatives for multiple channels).
Accelerated time-to-market: tech teams with a track record in developing generative AI models usually have a tried-and-true approach to collecting use-case-specific data and training models. Partnering with such a team means in-house engineers no longer need to build a strategy from scratch. As a result, AdTech vendors can cut significantly cut time-to-market.
Customization and scalability: A tech partner who works closely with the in-house team will understand the needs of the project and develop custom solutions that meet business goals.

For AdTech vendors venturing into creative management solutions, adding AI to their projects is a way to differentiate and strengthen their offerings. Generative AI, for one, has a lot of impact on creative management platforms and DCO tools: it can instantly generate creatives and optimize them in flight, as well as personalize assets so that they resonate with target segments.

How should AdTech vendors go about introducing generative AI to their platforms? Defining a clear use case, collecting reliable data for training models, and joining forces with a tech team specializing in AI development can help project teams be more confident in the success of their model.