Facebook Campaign & Achievements

When running paid media advertising can feel like a guessing game. You spend hours crafting the perfect ad creative, only to see it underperform. I’ve been there—more times than I’d like to admit. But over the years, I’ve learned that the secret to success isn’t just about fancy creatives or writing good copy. It’s about finding the right ad angle—the core message that resonates with your audience and drives action.

Today, I want to share my step-by-step process for identifying winning ad angles through systematic A/B testing. Whether you’re running lead-gen campaigns or eCommerce ads, these strategies will help you cut through the noise and achieve real results.

Creatives - A/B testing

What exactly is an angle?
An ad angle is the essence of your messaging—the “why” behind your ad. It’s not just about the image or the copy; it’s about the problem you’re solving and the emotional trigger you’re tapping into. The right angle can make or break a campaign. But here’s the thing: identifying the right angle isn’t always straightforward. Sometimes, the ideas you expect to perform best can underperform, while seemingly “less promising” angles can deliver 200% better results.

How do I test angles?
Here’s my process for identifying the best-performing angles through systematic A/B testing:

1. Start with Creatives:

Before diving into messaging, I focus on finding the right visual. I like to begin by testing 10–20 different images to optimize for Click-Through Rate (CTR). At this stage, I don’t stress too much about messaging or angles—just focus on the images themselves. This helps me identify which visuals resonate most with my audience. (by the way, these are small budget tests)

2. Test Copy:

Once I’ve identified the winning image, I move on to testing the copy. I duplicate the ad set and test various angles by changing only the copy, while keeping the image constant. This ensures that I’m isolating the impact of the angle itself.

For instance, in one campaign, I tested three angles:

Angle 1: “Save 20% on Cross-Border Payments”
Angle 2: “Streamline Your Payment Process in Minutes”
Angle 3: “Join 10,000 Businesses Trusting Our Payment Solutions”

The third angle, which focused on social proof, outperformed the others by 15%. It was a reminder that sometimes, the emotional trigger of trust and credibility can be more powerful than discounts or features.

3. The Creative Process:

If my team is struggling with creativity, I would brainstorm with PMM or PM who specialize in the product or industry (In 2024, I would also use different AI). A diversity of perspectives often leads to unexpected winners. For example, in a campaign for a B2B SaaS client, a freelance writer suggested an angle focused on “reducing operational headaches.” It was a phrase we hadn’t considered, but it resonated deeply with our audience and became the winning angle.

4. Finding the Winning Creatives (Img + Copy)

Step 1: Find the Winning Image
- Create the ad set, select your targeting, and design your first ad.
- Duplicate the ad set 9–19 times, changing only the image in each ad. This allows me to pinpoint which visual resonates best with the audience.

Step 2: Refine the Best Angle
- Duplicate the winning ad set for each angle your team has developed.
- Edit each ad with a unique angle.
- Let the ads run for three days at $20 per ad set before making any cuts.

Optimize - When to Adjust or Scale

When you first launch a new ad set, platforms like Facebook need time to "learn" - testing how audience segments respond to your ad. This is where the platform shows your ad to smaller “pools” of potential customers to gauge engagement. It’s essential to let this process unfold for at least a few days before making changes.

During the testing phase, my primary metric is CTR. Once I know which ad has the highest CTR, I analyze conversion data to see how that correlates to cost and performance. Interestingly, a lower CTR ad can sometimes outperform a high-CTR one in terms of conversion, forcing a more nuanced optimization process.

For instance, if an expensive ad brings in the most conversions, I spend time understanding why that specific combination of image and copy works better. I often break down the ad into its key elements and test variations—adjusting the image or going more aggressive with copy to drive higher CTR without sacrificing conversions.

Scaling a campaign is where the real challenge lies, especially when maintaining a healthy ROI. The principles of scaling apply across different platforms, but on Facebook, it’s critical to understand that as you increase budgets, ROI tends to decrease. The key is balancing volume with profitability.

Incremental Scaling:

My preferred method for scaling is incremental increases. I raise budgets by 15–30% every two days, allowing Facebook’s algorithm to adjust gradually. This avoids the penalties associated with rapid budget increases.

For example:

Day 1: $50
Day 4: $60 (20% increase)
Day 7: $72 (20% increase)
This method is slow but safe, minimizing the risk of algorithmic disruptions while scaling to higher daily budgets.

Ad Set Duplication (My Favorite Method):

This approach allows for faster scaling but requires careful attention to audience overlap. Duplicating an ad set 20–30 times can cause self-competition, leading to lower delivery rates. My strategy is to duplicate winning ad sets 4–6 times with a budget increase of 100–200%. I monitor results over 2–3 days, pausing underperforming duplicates and scaling successful ones.


As a performance marketer, my ultimate goal is making positive ROI at scale. I don’t focus on metrics like CPC or CTR in isolation—what matters is overall spending and ROI. Whether my CPC is $0.10 or $10, if the campaign is profitable and achieves the desired ROAS, I’m happy with it.

Scaling paid media ads requires both patience and a solid strategy. Incremental scaling is safer but slower, while ad set duplication offers faster growth with more risk. Either way, the key is maintaining consistency, tracking performance rigorously, and giving Facebook’s algorithm the time it needs to optimize effectively.

Professional Portfolio © 2020 Daniel Liu