Stack Creative Diversity in Phases

Let’s talk about how to stack creative diversity.

So look, ever since Meta announced Andromeda, the phrase Creative Diversification has become a hot topic. To get the best results, Meta recommends leaning into diverse creative instead of levers like targeting inputs.

That means diversity in format, like static images, videos, and carousels. It means diversity in text, customer personas, pain points, and solutions. And Meta wants uniquely different visuals, even if the formats are different.

It’s honestly a whole lot, and it can be overwhelming. If you take Meta literally, it could mean a lot of ads at once. And the first thing I’ll say is that MOST advertisers shouldn’t be taking Meta literally.

If you have a big budget to work with, absolutely, lean into those creative resources you have and churn out a ton of options. But for the rest of us, it can be a complete waste of time, energy, and resources.

That doesn’t mean you ignore Meta’s recommendations on creative diversification. A phrase I’ve been using a lot lately is approaching ad creation in the spirit of creative diversity.

So I’m always conscious of generating more text and visual options, and leaning into things like Advantage+ enhancements to help. But if you have a modest budget, you’ll learn quickly that it rarely pays to churn out 10, 20, or more ads at a time in an attempt to make Meta and Andromeda happy.

Meta will almost always focus most of your budget on a small number of ads, and you’ll be asking yourself why you even bothered.

But there’s a way to lean into creative diversity while doing it in phases. And this is what I call STACKING creative diversity.

You’re not doing it all at once. You’re doing it in phases. And an important element is Meta’s creative testing tool.

So let me give you an example of this in action.

I started a new ad set to promote a new lead magnet. I worked out a general theme regarding how the ad would look, the customer persona, the pain points, and the messaging involved.

So I had the five primary text options and a general creative theme ready to go.

I was starting with images, and these would be a simple style of a flat background with text. But what I didn’t know is what specific color combination I wanted to use.

So I created five versions of this ad that all used the same primary text and headlines, five for each of course. But each ad leaned into a different color combination. There was a white background, black, green, blue, and orange.

Now, you probably know that Meta discourages this generally. Or at least, they say that five ads with only a color change isn’t diverse creative.

But that’s not the point here. My initial goal is to figure out which color combination works best.

So I start the ad set with a 7-day test using the creative testing tool. The entire budget goes to the test since these are the only ads at the beginning.

This will help me understand which version performs best, not just because the algorithm chosen a version but because budget was given to each version.

Then, after the test is over, budget will be distributed algorithmically. Meta will likely spend most of the budget on one or two ads.

And then I’ll let that ad set run normally for a week or two.

At this point, the ad set isn’t classically DIVERSE according to Meta’s definition. But if you’re getting good results, who cares?

So after a couple of weeks, I make that analysis. Is this ad set performing great in aggregate? If it is, I don’t touch it.

If I feel like it could still do better, I move on to Phase 2.

And this is where the stacking begins.

I learn from that first set of ads to create a uniquely different second set. Because the first set was static images, I may create entirely videos.

But they will be completely different from the first set in every way. Visuals, messaging, customer personas, everything.

This second stack begins to add to our desired creative diversity.

Of course, I don’t just publish five new ads. I start another test within that existing ad set for those five new ads. And I dedicate a portion of the overall budget to the test.

And then I repeat the process.

I let the test complete. I learn from it. Then, after a couple of weeks after the test, I ask how the ads are performing in aggregate.

Rinse and repeat. Add another set of ads to the stack if necessary.

So here’s the bottom of the glass.

I encourage creative diversity. But I discourage most advertisers from feeling the need to attack it all with the initial set of ads.

You’re doing it blindly. And most of the ads won’t even get shown.

Start with one general theme and one set of ads at a time, and start with a test. Learn from what works and what doesn’t to inform the next, uniquely different, set of ads.

By stacking, you’re generating diverse creative more slowly and strategically. And you’re not simply creating more ads for the sake of creating more ads.

You might not need more than a handful of ads at all if you get good results from the start.

And that’s what stacking does. It provides focus and limits wasted time and resources while being more strategic.