Skip to main content

Artificial intelligence (AI) advertising testing solutions are attracting more attention and budget than ever before. Our research tells us that 49% of global marketers plan to spend more on creative AI testing in the next year. And with good reason; promising quick responses and lower costs than survey-based approaches, the ability to test more content faster is clearly appealing. But where should AI solutions sit alongside survey-based testing? What are the strengths and advantages of each? And when should one or the other be used? These are questions marketers need to address to ensure the right product is used in every situation and ultimately to get the best return on advertising spend.

Creative quality is essential to the success of advertising campaigns – probably more so than you might think. Kantar data shows that creative quality is the second most important factor for profitable advertising, while marketers believe it is the fourth most important. We also know that great creativity drives both brand value and sales. Thus, the right creative approach, combined with the precision of execution details, adds up to memorable elements that form the basis of winning campaigns.

AI tools can augment the fundamentals of creative testing

From establishing your strategy and initiating ideation to executing your campaign and optimizing along the way, creative success is a journey, with testing and learning involved at every step. . This process can take months or days depending on whether it’s a new flagship campaign being designed or a change of assets in a digital campaign. But the fundamental process remains the same.

Add to that things are constantly changing which means standing still is not an option. New platforms are constantly emerging and new digital formats to explore. We’ve heard so much about the potential of the metaverse this year, with some brands leading the way in finding a role they can play in these virtual worlds. It’s something we expect to see more brands explore in 2022 and beyond.

Additionally, even traditionally offline channels are moving online with the unstoppable rise of VOD viewing, programmatic audio and digital out-of-home (DOOH) innovation. In DOOH we see brands getting really creative, using dynamic creative approaches that change based on weather conditions, time of day, traffic levels or location, as well as screen technologies. advanced, to deliver truly engaging experiences. All of this requires insight at every stage to understand where the creative is streaming or could be further improved.

AI enables more iterative testing throughout the campaign lifecycle

With all of these changes, marketers and agencies must constantly test and learn, and every business and campaign will need a framework to support it. Part of this framework is understanding what the right testing approach is for each stage of a campaign.

At Kantar, we view the role of creative AI-based testing as a predictive tool. It can give you a positive or negative result on a creative approach and the lower costs open up the possibility of testing competitors’ advertising – which has been largely prohibitive until now. We see this as a primary use case for AI solutions. But the data that feeds it and the assumptions that go into generating those quick indicative results are crucial to the output of any AI tool.

AI and survey-based ad testing work hand in hand

But it’s not all about AI. Advancements in AI simply reinforce why and when marketers should use survey-based testing. This will always be necessary to form the fundamental data that the AI ​​tool can rely on to obtain relevant information for the task at hand, whether it is to examine new creative routes or a launch campaign. complete. This fundamental data is crucial, because an AI tool is only as good as the data behind it. This is where scale matters and quality. The data fed into Kantar’s AI tools comes from a database of 230,000 actual ad tests.

When launching a new campaign, whether TV or digital, detailed information is important because details matter. At Kantar, we recommend testing a TV execution three times, whether for a new product, a new campaign, a new creative theme or even a cut-down: first, at an early stage, then again after a few key changes. and adjustments based on the results, and finally in the near final stage, to be refined. This means you maximize your chances of success and can only gain this granular, second-by-second understanding and optimization from survey-based pre-testing and insight into key themes and analytics. specific requirements such as celebrity use, music, I&D, etc.

To facilitate – and accelerate – more iterative testing, we recently launched Link AI for Digital on Kantar Marketplace, which offers creative effectiveness predictions for digital video ads in just 15 minutes, evaluating them against behavioral metrics and creatives that drive ad performance. It gives marketers the ability to predict digital ad performance before it hits the market, evaluate different versions of an ad, test competitor creative, and test large volumes of advertisements to identify trends and create creative references. Link AI for Digital is part of a suite of AI-powered features on Kantar Marketplace, which also includes Link AI for TV. Clients like Google and Unilever are already using these tools to predict how audiences will react to their ads.

The future we see is not one where AI replaces survey-based testing. We see them both working together, in packages, for specific use cases. This hybrid approach will allow pre-testing frameworks suitable for any client who wishes to maximize the return on their creation.