Why creative quality needs a rethink in the age of GenAI
Why It Matters
Generative AI has made it easier than ever for brands to create content at scale. But as the volume of advertising explodes, the real challenge is no longer producing content; it is ensuring that it actually works.
Key Takeaways
- Traditional pre-launch testing is struggling to keep pace with the volume of content produced in the GenAI era.
- Brands should shift from testing a few advertisements before launch to evaluating all content after it goes live.
- Linking creative quality to business outcomes can help organisations improve marketing effectiveness and media returns.
The AI Content Boom Is Creating a New Problem
Generative AI has transformed the economics of content creation. Marketing teams can now produce thousands of advertisements, social media posts and videos faster and more cheaply than ever before. As digital platforms demand a constant stream of fresh content, brands are increasingly turning to AI tools to keep pace.
But this surge in productivity has created a new challenge: quality control.
For decades, marketing research has consistently shown that creative quality is one of the most important drivers of advertising success. Strong creative work influences whether consumers notice, rememberand respond to a campaign. Yet as AI makes it possible to generate content at unprecedented scale, brands risk flooding digital channels with material that is merely adequate rather than effective.
The result is what many industry observers now describe as “AI slop”: vast quantities of content that compete for attention but fail to make an impact.
Why Traditional Testing Is No Longer Enough
Historically, brands relied on pre-testing. Before launching a campaign, marketers would show advertisements to consumers, gather feedback and refine the content before investing media budgets.
This approach worked when campaigns consisted of a handful of advertisements. However, today's digital environment requires brands to create hundreds or even thousands of variations tailored to different platforms, formats and audiences.
AI-powered testing tools have emerged to address this challenge. These systems analyse creative assets and predict how effectively they may perform. While useful, they still largely serve as filtering mechanisms, helping marketers choose which advertisements to launch.
The limitation is that these tools evaluate only a small part of the overall content ecosystem. They may improve individual campaigns, but they do little to help organisations understand the quality of all the content they produce or identify broader patterns that drive success.
A Shift Towards Continuous Measurement
The article proposes a different approach: measuring creative quality after content is deployed rather than relying primarily on testing beforehand.
Using AI-powered assessment tools connected directly to advertising platforms, brands can evaluate every piece of content that runs in the market. This creates a comprehensive picture of the quality of their entire media investment rather than a snapshot of selected campaigns.
Such an approach offers several advantages. First, it allows organisations to monitor whether teams consistently follow proven creative best practices across regions, brands and business units. This is especially valuable in large organisations where content creation is increasingly decentralised.
Second, it enables marketers to link creative quality with real-world outcomes such as brand awareness, ad recall and purchase intent. By identifying which creative characteristics contribute most to performance, organisations can establish continuous feedback loops that improve future campaigns.
Third, creative quality data can be incorporated into marketing mix models and other econometric analyses. This allows firms to quantify the financial impact of creative quality on media return on investment – something that has historically been difficult to measure at scale.
Business Implications
As AI lowers the cost of content production, competitive advantage will depend less on how much content brands can create and more on how effectively they can identify what works.
The future of creative evaluation is likely to be continuous rather than episodic. Instead of relying solely on pre-launch testing, organisations may increasingly assess creative quality throughout the lifecycle of a campaign, using real-time data to refine and improve performance.
For marketing leaders, this means investing not only in content generation tools but also in measurement systems that connect creative quality to business outcomes. In an era where every brand can produce content at scale, the winners may be those that learn fastest from the content they create.
Author and sources
Author: Ramanathan Vythilingam (Nanyang Technological University)
Original article: WARC, January 2026
---
For more research, click here to return to NBS Knowledge Lab.
