Imagine you’re running a Meta advertising campaign for an automobile. Your market research reveals that potential customers generally fall into four groups:
- Those who value the car’s technical features and performance.
- Those who care most about its appearance and design.
- Those who prioritise safety features and ratings.
- Those who are primarily concerned with affordability.
From a creative perspective, rather than designing a single ad that attempts to highlight all four selling points, you could develop four separate ads, each built around one of these themes. This allows you to tailor the messaging to the motivations and priorities of different audience segments.
This is a simplified example of creative diversification: developing different creative approaches to appeal to different people, rather than relying on audience targeting alone to reach different segments.
Adromeda’s Role
There’s a lot of discussion around creative diversification at the moment, and much of it is framed as though it’s a new tactic brought about by Andromeda, Meta’s new ad delivery model. We’re seeing people describe Andromeda as something that has “flipped advertising strategy on its head.“
When the conversation is framed this way, I think too much emphasis is placed on the ad tech, and people lose sight of the fact that this is fundamentally a strategy about communicating effectively with diverse groups of people with different motivations and values.
At its core, creative diversification is rooted in the age-old principles of advertising: communicating value through the right message, to the right people, at the right time. While it may not have always been called “creative diversification,” the concept existed long before Andromeda.
In Meta’s own words, Andromeda supports creative diversification strategies by “working behind the scenes to power more complex AI models that can better personalise which ads are shown to each person”. This isn’t Andromeda flipping advertising on its head. It’s Meta improving its creative-audience matching technology, so that advertisers can focus less on targeting tactics and more on creative strategies that deliver the right message to the right people, and do so at scale.
Human Response At The Heart
Human response is at the heart of the entire process. Meta uses people’s interactions with your ads as signals to optimise delivery towards more leads or sales. No matter how much you diversify your creative, without people taking action, Meta cannot gather the meaningful data it needs to learn and improve delivery. It’s why campaigns need a sufficient volume of results to exit the learning phase.
And so, creative diversification is a means to an end, not the goal. By reaching more people with messages that connect, you increase your opportunities to generate quality results and give Meta more meaningful data to learn from. But diversification alone doesn’t guarantee better results or more useful data. A single ad that consistently generates qualified leads will provide Meta with more meaningful learning signals than a diverse set of ads that generate little or nothing.
Ultimately, this post isn’t about questioning the merits of creative diversification as a strategy. Its effectiveness is undeniable. It’s about exploring why it works and grounding the conversation in the fundamental principles of advertising and marketing. Our job as advertisers is to communicate value through content and messaging that connects with people, not to “hack” algorithms. Algorithms are simply tools that help us find the right people more effectively and at scale. I feel like that’s something that sometimes gets lost in the way we talk about things.
