Solved: How Many Creatives per Campaign Should We Use?

Hi there,

Thanks for reaching out! Happy to give you some of my initial thoughts and guidance on your question about creative testing. It's a question we get a lot, and you're right, there's a ton of conflicting advice out there.

The short answer is that the "one campaign, one asset" approach is almost always a recipe for wasting money. Real growth in paid advertising doesn't come from finding a single "magic" ad; it comes from building a systematic testing framework that constantly finds new winners and tells you what works and why. I'll walk you through how we approach this below.

You'll need a better system than "one creative per campaign"...

Let's get this out of the way first. The idea of running one campaign with a single ad creative is one of the most common and costly mistakes I see people make. It's like going fishing with a single hook and a single type of bait and hoping every fish in the lake loves it. It just doesn't work like that.

Why is it so bad? A few reasons:

-> No Learning: If the ad works, you don't know *why*. Was it the image? The headline? The audience? If it fails, you're in the same boat. You've learned nothing except that one specific combination didn't work. You're back to square one with zero actionable data.

-> No Scalability: Let's say you get lucky and the one ad does work. Great. What happens in three weeks when it stops working due to ad fatigue? You have no backup, no new angles to test, and no understanding of what else might resonate with your audience. Your growth stalls instantly.

-> Inefficient Spending: Ad platforms like Meta and Google are built on algorithms that thrive on data. By giving them only one option, you're tying their hands. They can't optimise delivery, they can't find cheaper pockets of your audience, and they can't test different placements effectively. You're essentially forcing the platform to be dumber than it is, which usually results in higher costs for you.

I remember one eCommerce client who came to us after spending thousands on this exact strategy. They had hundreds of tiny campaigns, each with one ad. It was an absolute mess. They had no idea what was actually driving sales and were just throwing money at whatever seemed to work that day. By implementing a proper testing structure, we not only increased their revenue but also made their account manageable and their results predictable. That's the goal here: predictable, scalable growth, not random luck.

I'd say you need to structure your campaigns properly...

Before you even think about how many creatives you need, you have to get the campaign structure right. This is the foundation for everything. Without it, your testing will be chaotic and your results will be messy. We structure almost every account we manage around the marketing funnel. It's a simple concept but incredibly powerful.

Here's what it looks like in practice:

1. Top of Funnel (ToFu) - Prospecting: This is your cold audience. These people have likely never heard of you. The goal here isn't necessarily an immediate sale (though it can happen). The goal is to introduce your brand, identify people who have the problem you solve, and get them to take a small first step, like visiting your website.

2. Middle of Funnel (MoFu) - Consideration: This is your warm audience. These are people who have engaged with you in some way—they've visited your site, watched a video, liked a post—but haven't purchased or become a lead yet. The goal here is to build trust, overcome objections, and move them closer to a decision.

3. Bottom of Funnel (BoFu) - Conversion: This is your hot audience. These people are on the verge of converting. They've added a product to their cart, initiated checkout, or visited your pricing page. The goal here is to give them that final nudge to get them over the line.

You should have separate, long-term campaigns for each of these stages. Why? Because the audience and the message for each stage are completely different. You wouldn't talk to a complete stranger (ToFu) the same way you'd talk to someone who's about to hand you their credit card (BoFu). By separating them, you can tailor your messaging, offers, and budget to what's most effective at each stage.

We'll need to look at prioritising your audiences...

Once your campaigns are structured, the next step is to test audiences within them. But you don't just test random audiences. There's a clear hierarchy of what's likely to perform best, especially on a platform like Meta. The closer an audience is to your desired conversion event (like a purchase), the better it will generally perform.

This is the order we usually test audiences in, starting with what's most likely to work:

You start testing within your ToFu campaign using an ad set for each of your most promising audiences. For a new account, you might start with 3-5 different interest-based ad sets. For an account with data, you'd start with your highest quality lookalikes. The key is to keep them separate so you can clearly see which audience performs best. Don't lump them all together. Once you have enough data (you generally need at least 100 people for a custom audience to work), you build out your MoFu and BoFu campaigns with retargeting audiences.

The quality of your source audience for lookalikes is absolutely vital. A lookalike based on your best customers (highest LTV) will almost always outperform one based on general website visitors. The algorithm is only as good as the data you give it.

You probably should focus on testing angles, not just ads...

Okay, so now we have the structure. This is where we finally talk about creatives. Inside each of your ToFu ad sets (the audience tests), you should run multiple ads. But again, not just random ads. You need to be systematic.

I usually recommend starting with 3-5 creatives per ad set. But the number isn't as important as the *variation*. You want to test different hypotheses.

-> Test Different Angles: An "angle" is the core message or hook. For example, if you sell high-quality kitchen knives, your angles could be 'Durability' (lasts a lifetime), 'Precision' (perfect cuts every time), or 'Status' (the choice of professional chefs). You need to find which pain point or desire resonates most with each audience.

-> Test Different Formats: Within each angle, you can test different formats. A static image, a carousel, a short video, a user-generated content (UGC) style video. Some audiences respond better to video, others to a simple, clear image. For a number of our SaaS clients, UGC videos have been incredibly effective, much more so than slick, corporate-style productions. You won't know until you test.

So, instead of "one ad per campaign," you now have, for example, one ToFu campaign, 3 ad sets (each testing a different audience), and 4 ads inside each ad set (e.g., Angle 1 as an image, Angle 1 as a video, Angle 2 as an image, Angle 2 as a video). Now you're generating real data. You'll quickly see which audiences and which creative approaches are your winners.

You'll need to know when to kill an ad...

Testing is pointless if you don't know how to interpret the results and act on them. A common question is, "How long should I wait before turning an ad off?"

The answer isn't about time; it's about money. My rule of thumb is pretty simple: if an ad set or a specific ad has spent 2-3 times your average target Cost Per Acquisition (CPA) without a single conversion, it's time to turn it off. It's probably not going to become profitable.

For example, if you know a new customer is worth £100 to you and you're aiming for a £25 CPA, once an ad set hits £50-£75 in spend with no sales, it gets paused. Don't get emotionally attached to your ads. Trust the data.

This is where understanding your numbers becomes absolutely crticial. You can't make good decisions if you don't know what a lead or a customer is actually worth to you. This is where calculating your Lifetime Value (LTV) is so important.

Let's run through a quick example. Let's say you're a SaaS business.

• Average Revenue Per Account (ARPA): £100/month
• Gross Margin: 80%
• Monthly Churn Rate: 5%

Your LTV would be (£100 * 0.80) / 0.05 = £1,600. So, each customer is worth £1,600 in gross margin. A healthy business might aim to spend 1/3 of that on acquisition, meaning your target CPA is around £533. All of a sudden, paying £50 for a lead that might convert doesn't seem so bad, does it? This maths frees you from chasing cheap, low-quality leads and allows you to invest in acquiring proper customers.

Once you identify winning audiences and creative angles, you don't just leave them. You scale them. This usually means gradually increasing the budget on the winning ad sets. At the same time, you're taking what you've learned from the winning creative (e.g., "The 'Precision' angle works best with video") and creating new variations to test against your current winner. This creates a cycle of continuous improvement, which is the complete opposite of the static "one ad" aproach.

This is the main advice I have for you:

To wrap this up, let's put it into a clear, actionable plan. Forget about the "right number" of creatives and instead adopt a systematic process. The mishmash of advice exists because people are looking for a simple answer to a complex problem. The real answer is having a solid framework.

As you can probably tell, this is a lot more involved than just setting up an ad and hoping for the best. It's a continuous process of hypothesising, testing, learning, and optimising. It requires a deep understanding of the platform, your audience, and your own business metrics.

Getting this framework right is often the difference between an ad account that struggles to break even and one that becomes a reliable engine for growth. If you feel like this is a bit overwhelming or you'd just rather have an expert team handle it for you, that's what we do. We help businesses build and manage these kinds of scalable advertising systems every day.

We offer a free, no-obligation initial consultation where we can take a look at your specific situation and ad account together. We can give you some more tailored advice and show you exactly what we'd do to improve things. It's a great way to get some expert eyes on your strategy, with no strings attached.

Regards,

Team @ Lukas Holschuh