Solved: Best Way to Test Creative and Copy on Facebook Ads?

Hi there,

Thanks for reaching out!

I saw your post about structuring your Facebook ads and thought I'd give you some more detailed thoughts. It's a really common point of confusion, especially when you're starting out but have a good amount of creative assets to work with. It's a good problem to have, mind. A lot of people struggle to get enough decent creative, so you're already a step ahead.

You've basically hit on the central question of ad testing: how to test multiple variables at once without making a mess of your ad account or spending a fortune to get answers. You're right to think about structure, it's the foundation of any successful campaign. Without it, you're just throwing stuff at a wall and hoping something sticks, which is a quick way to burn through your budget.

I'm happy to give you some initial guidance based on what we do for our clients. Hopefully it'll give you a clear path forward.

We'll need to look at how you structure your tests...

Your question is mainly about testing creatives and copy at the same time. You mentioned Dynamic Creative, and you're spot on, that's one of the main ways to handle this. Let's break down the two main approaches you can take: a structured manual test and a Dynamic Creative test.

1. Dynamic Creative Testing (DCT)

This is what you were alluding to. It's a feature within Meta's ad platform that's built specifically for this situation. You give it a bunch of 'assets' – images, videos (like your UGC), headlines, primary text (your copy variations), and calls-to-action – and its algorithm will automatically mix and match them in real-time. It then works to find the highest-performing combinations and serves them more often to your audience.

For someone in your position, this is probably the best place to start. It's efficient and lets you test a lot of combinations without having to build dozens of individual ads. You said you have UGC, infographics, and good product photos. And you have testimonial, long-form, short-form, and story-based copy. This is a perfect scenario for a DCT.

The process would be to create a single ad within your ad set, toggle the 'Dynamic Creative' option on, and then upload all your assets. You can load in up to 10 images/videos and 5 versions of everything else (primary text, headline, description). Facebook does the heavy lifting for you.

The main advantage is speed and efficiency. The potential downside is that the reporting can be a bit less granular. You can see which individual assets are performing best (e.g., 'UGC_video_1' got the best results, or 'testimonial_copy_A' performed best), but it's sometimes harder to see the exact winning *combinations*. Still, it's a powerful tool to quicky identify your strongest creative angles and messages without a massive amount of manual work.

2. Manual Split Testing

The other way is the old-school manual approach. This gives you more control and cleaner data but takes more effort to set up. Here, you'd create multiple, seperate ads within a single ad set. For example:

• -> Ad 1: UGC Video + Testimonial Copy
• -> Ad 2: UGC Video + Short Copy
• -> Ad 3: Product Photo + Testimonial Copy
• -> Ad 4: Product Photo + Short Copy

...and so on. You'd keep the audience the same in the ad set and let the ads compete against each other. The ad with the best results (e.g., lowest cost per purchase) is your winner. This method gives you crystal-clear data on which specific combination works best. However, if you have lots of assets like you do, the number of ads can quickly get out of hand, and it can split your budget too thinly across too many ads, making it take longer to find a clear winner. For the amount of assets you have, this would be a real headache to manage.

Given you're pretty new to this, I'd strongly recommend starting with Dynamic Creative. It's designed for this exact purpose and will give you directional data on your best assets much faster. You can always take the winning elements from a DCT and run a more focused manual test later if you want to refine things even further.

I'd say you need a solid audience strategy first...

This is probably the most important bit. You can have the best ads in the world, but if you show them to the wrong people, you'll just be breaking even at best, or losing money at worst. A lot of people I see getting started with ads spend 90% of their time on the creative and 10% on the audience. It really should be the other way around, especialy at the start.

When we take on a new client, before we even think about ad creative, we map out their entire audience strategy. For an eCommerce business, this usually breaks down into a funnel structure: Top of Funnel (ToFu), Middle of Funnel (MoFu), and Bottom of Funnel (BoFu).

• -> ToFu (Top of Funnel): These are cold audiences. People who have never heard of you before. This is where you find new customers.
• -> MoFu (Middle of Funnel): These are warm audiences. People who have engaged with your brand in some way (e.g., watched a video, liked a post, visited your website) but haven't made a purchase or added to cart.
• -> BoFu (Bottom of Funnel): These are hot audiences. People who have shown strong buying intent (e.g., added a product to their cart, initiated checkout) but haven't completed the purchase.

The audiences in BoFu are almost always going to perform better (give you a higher Return On Ad Spend) than MoFu, and MoFu will perform better than ToFu. It just makes sense – someone who has already put your product in their basket is much closer to buying than a complete stranger. So, we prioritise them. Since you've only run one ad, you might not have much data yet, but this is the framework you should be building towards.

Here's a list of audiences we typically prioritise for eCommerce clients. You should be thinking about how you can build these over time.

For you, starting out, you'll be focused on the ToFu audiences. The most important thing here is to get your Detailed Targeting right. This is where people go wrong all the time. They pick interests that are way too broad. For example, if you sell high-end coffee beans, targeting people with an interest in "Coffee" is a poor strategy. It includes millions of people who just drink instant coffee from the supermarket. You'd be better off targeting interests like "Specialty coffee", "Aeropress", "Third Wave Coffee", or followers of specific roaster brands or coffee influencers. You have to think about what interests your ideal customer has that the general population doesn't. That's the secret to good interest targeting.

Once you've run your ads for a while and your Meta Pixel has gathered some data (you need at least 100 people in an audience to use it for retargeting or lookalikes), you can start building out your MoFu and BoFu campaigns. But for now, focus on nailing your cold audience targeting.

You probably should build a simple, scalable campaign structure...

Right, let's tie this all together into a structure you can actually use. You mentioned being at a loss with structuring the campaign and whether everything should be in the same ad set. The answer is almost always no. You want to seperate your cold audiences from your warm/hot audiences.

A simple, effective structure to start with is a single campaign using Campaign Budget Optimisation (CBO).

Campaign: CBO - Conversions (e.g., Purchases)

With CBO, you set the budget at the campaign level, and Meta will automatically distribute it to the ad sets that are performing the best. This is much more efficient than setting individual budgets for each ad set.

Inside this campaign, you'd start with two ad sets:

-> Ad Set 1: ToFu - Cold Prospecting

• Audience: Detailed Targeting. Pick a theme of highly relevant, specific interests. For instance, if you sell sustainable yoga mats, you might layer interests like "Yoga", "Lululemon", AND "Sustainable living". Don't just put "Yoga".
• Ads: Create one Dynamic Creative ad in here. Load it up with all your UGC, infographics, product photos, and all your different copy angles. Let Meta figure out what resonates with a cold audience.

-> Ad Set 2: MoFu/BoFu - Retargeting

• Audience: Since your budget and audience sizes are probably small to begin with, you can group your warm and hot audiences together. Create a custom audience of "All Website Visitors - Last 30 Days" and another for "All Instagram Engagers - Last 30 Days". Put them both in this ad set. Make sure to exclude "Purchasers - Last 180 Days" so you're not annoying existing customers.
• Ads: You could run the same Dynamic Creative ad in here. Or, you could create a seperate, more direct ad. For retargeting, sometimes a simple ad with a strong offer ("Still thinking about it? Here's 10% off to help you decide") or a carousel showing the products they viewed can work wonders. A testimonial-based ad would also work very well here to build trust and push them over the edge.

This two-ad-set structure is simple, allows you to test everything you want to, and seperates your cold traffic from your warm traffic, which is a best practice. The CBO will ensure your money flows to whichever ad set is getting you cheaper purchases. Over time, as you get more data and a bigger budget, you can split these out further (e.g., seperate ad sets for BoFu and MoFu, multiple different ToFu ad sets testing different interests or lookalikes).

I remember a client in the women's apparel niche we worked with. They had a similar situation to you – lots of great UGC but messy campaigns. We restructured their campaigns into a simple ToFu/MoFu/BoFu CBO campaign and used DCT to test their creative, which led to their return on ad spend increasing to 691%. Structure and methodical testing really does work.

You'll need to analyse the results properly...

Running the ads is only half the battle. You need to know how to read the data to make smart decisions. Since you're using DCT, after a few days (you need to let it run long enough to get meaningful data, don't make decisions after just 24 hours), you can use the "Breakdown" feature in Ads Manager.

Go to your ad level, select the dynamic ad, and click 'Breakdown'. You can then choose to break down the results by 'Image, Video and Slideshow' or by 'Text'. This will show you which of your individual assets are performing the best. You'll see metrics like Amount Spent, CTR (Click-Through Rate), and Cost Per Purchase for each creative and each piece of copy.

This is your goldmine of information. If you see that your UGC videos are getting all the spend and have a much lower cost per purchase than your infographics, that's a clear signal to double down on UGC. If your short, punchy copy is outperforming your long story-based copy, you've learned something valuable about your audience's preferences. It's not about finding one single perfect ad; it's about learning these broader lessons that you can apply to all your future marketing.

You also need to look at the whole funnel, not just the ad metrics:

• -> High Impressions, Low CTR? Your ad isn't grabbing attention. The image/video hook isn't working, or the first line of your copy isn't compelling enough. Your creative is likely the problem.
• -> High CTR, Low Add to Carts? People are interested enough to click, but something on your product page is putting them off. It could be your price, your product photos, your product description, or even a slow-loading page.
• -> High Add to Carts, Low Purchases? They want to buy, but something in your checkout process is stopping them. Unexpectedly high shipping costs are a classic culprit here. Lack of trust signals (reviews, security badges) can also cause people to drop off.

Understanding where the drop-off is happening is key to knowing what to fix. It's not always the ad's fault. The best ad in the world can't fix a broken website or an uncompetitive offer. The process is one of continuous improvment. You test, analyse the data, find the winners, turn off the losers, and then create new ads to test against your previous winner. That's how you go from breaking even to achieving the kind of results we see with our clients. For example, we helped a cleaning products company achieve a 633% return using Meta Ads. It's a marathon, not a sprint.

To sum it all up, this can seem like a lot to take in at once. So, I've put my main recommendations for you into a simple table to give you a clear, actionable plan.

I know this is a tonne of information, but hopefully, it gives you a much clearer picture than you had before. Following this structure will put you lightyears ahead of most people who are new to Facebook ads.

While this advice will definitely get you on the right track, the real challenge comes in the day-to-day management, ongoing analysis, and constant iteration required to not just get results, but to scale them profitably. It's a full-time job for a reason, and it's what seperates campaigns that 'break even' from those that become a major growth engine for a business.

If you decide you'd rather focus on running your business and leave the complexities of paid advertising to an expert, we're always happy to help. We offer a free, no-obligation initial consultation where we can take a proper look at your ad account and website, and talk through a bespoke strategy for your specific goals.

Either way, I hope this detailed breakdown has been genuinely helpful for you. Best of luck with the campaign!

Regards,

Team @ Lukas Holschuh