Sales & Conversion

How I Stopped Wasting Ad Spend by Testing Creative Blocks Instead of Audience Obsession


Personas

Ecommerce

Time to ROI

Short-term (< 3 months)

When I started managing Facebook Ads for a B2C Shopify store, I made the same mistake every marketer makes: I became obsessed with finding the "perfect audience." I spent weeks meticulously crafting different audience segments, targeting specific demographics, interests, and behaviors, convinced that the secret to profitable ads was hidden in some magical audience combination.

But after burning through budget testing audience after audience with mediocre results, I discovered something that completely changed my approach to Shopify advertising: creatives are the new targeting. Privacy regulations have fundamentally broken detailed targeting, but most marketers are still fighting yesterday's war.

In this playbook, I'll show you the exact split testing framework I developed that shifted focus from audience hunting to creative testing – and how it transformed our ROAS. You'll learn:

  • Why traditional audience testing is dead (and what replaced it)

  • The 3-creative weekly testing rhythm that scales

  • How to structure ad sets for maximum creative insights

  • The creative angles that actually move the needle

  • Real examples from a campaign that went from break-even to profitable

If you're tired of chasing audiences that don't convert and ready to embrace the new reality of Facebook advertising, this playbook will save you months of wasted spend.

Industry Reality

What every marketer thinks they need to master

Walk into any marketing conference or scroll through any advertising forum, and you'll hear the same broken record playing: "It's all about audience targeting." The industry has built an entire mythology around finding your perfect customer through Facebook's targeting options.

Here's what conventional wisdom tells you to focus on:

  • Detailed Demographics - Age ranges, income levels, job titles, relationship status

  • Interest Layering - Stacking multiple interests to narrow down your "ideal" customer

  • Lookalike Refinement - Creating lookalikes of your best customers, then layering interests on top

  • Behavioral Targeting - Targeting based on purchase behavior, device usage, travel patterns

  • Geographic Precision - Drilling down to specific cities, zip codes, or radius targeting

This approach made sense in 2018. Facebook had granular data on user behavior, iOS 14 hadn't destroyed attribution, and you could actually trust the platform to find your customers based on detailed parameters.

But here's the uncomfortable truth: Privacy regulations killed detailed targeting. When iOS 14.5 launched and GDPR tightened data collection, Facebook lost access to the behavioral signals that made precise targeting possible. Yet most marketers are still optimizing for an advertising landscape that no longer exists.

The result? Marketers waste months testing audience combinations that perform identically because the platform can't actually differentiate between them anymore. They're solving yesterday's problem with yesterday's tools while their competitors figure out what actually works in 2025.

Who am I

Consider me as your business complice.

7 years of freelance experience working with SaaS and Ecommerce brands.

I fell into this exact trap when working with a B2C Shopify store. The client sold fashion accessories – the kind of products that should be perfect for Facebook advertising. Beautiful visuals, clear target demographic, proven market demand. On paper, everything looked right.

My initial approach followed the textbook playbook. I created detailed audience segments: fashion enthusiasts aged 25-35, layered with interests in specific brands, income levels, and shopping behaviors. I spent two weeks building what I thought was a sophisticated targeting strategy with 12 different audience combinations.

The results? Absolutely mediocre. ROAS hovering around 2.5 with a €50 average order value. With their margins, we were barely breaking even. Every audience performed almost identically, which should have been my first red flag.

Here's what I was doing wrong: I was treating Facebook Ads like it was still 2019. I was trying to outsmart an algorithm that no longer had access to the data I was optimizing for. While I obsessed over audience nuances, I was using the same three creative variations across all my "precisely targeted" ad sets.

The turning point came when I discovered a simple truth that changed everything: if all my audiences were performing the same, maybe the problem wasn't the targeting. Maybe the platform was already finding the right people, and I needed to focus on giving those people better reasons to buy.

That's when I learned about the fundamental shift in Facebook advertising: your creative IS your targeting now. Instead of telling Facebook who to target, you let your creative signal who should be interested. A lifestyle-focused creative attracts lifestyle buyers. A problem-solution creative attracts people with that specific problem. Your message becomes your audience filter.

My experiments

Here's my playbook

What I ended up doing and the results.

Once I realized that creatives had become the new targeting mechanism, I completely restructured my approach. Instead of multiple audiences with the same creatives, I flipped it: one broad audience with multiple creative angles.

Here's the exact framework I developed:

The Simple Setup
I eliminated all the complex audience targeting and went with the simplest possible structure:

  • One campaign

  • One broad audience (basic demographics only: gender, country, age range)

  • Multiple ad sets, each testing a different creative angle

The 3-Creative Weekly Testing Rhythm
This became the backbone of the entire strategy. Every single week, without fail, I produced and launched 3 new creative variations. Not 1, not 5 – exactly 3. This wasn't about quantity for the sake of it. It was about:

  • Giving the algorithm fresh data points to work with

  • Preventing creative fatigue before it impacts performance

  • Building a library of proven creative concepts

  • Discovering unexpected angles that resonated with different segments

Creative Angle Categories I Tested
Rather than random creative variations, I systematically tested different psychological triggers:

  1. Lifestyle Integration - Showing the product in aspirational lifestyle contexts

  2. Problem-Solution - Leading with a specific problem the product solves

  3. Social Proof - User-generated content and testimonials

  4. Benefit-Focused - Highlighting the main functional benefit

  5. Scarcity/Urgency - Limited availability or time-sensitive offers

The Testing Structure
Each week's 3 new creatives would test different angles from these categories. For example:

  • Week 1: Lifestyle + Problem-Solution + Social Proof

  • Week 2: Benefit-Focused + Scarcity + Lifestyle (different execution)

  • Week 3: Social Proof (different format) + Problem-Solution + Urgency

The Platform's Learning Process
Here's what I discovered was actually happening: Facebook's algorithm was using each creative as a signal to find different micro-segments within my broad audience. The lifestyle creative attracted aspirational buyers. The problem-solution creative found people actively experiencing that pain point. The social proof creative resonated with validation-seekers.

By testing creative angles instead of audiences, I was essentially letting the platform do the sophisticated targeting work while I focused on crafting messages that would resonate with each micro-segment it discovered.

Creative Variety

Testing different psychological triggers weekly instead of audience segments

Algorithm Training

Letting Facebook learn from creative signals rather than manual targeting

Performance Tracking

Measuring which creative angles consistently drive better ROAS

Scaling Winners

Expanding successful creative concepts across multiple campaigns

The results spoke for themselves. Within six weeks of implementing this creative-first approach, the campaign performance transformed completely.

Instead of the flat 2.5 ROAS we'd been stuck at, we started seeing consistent ROAS improvements. The algorithm began identifying high-value customers more effectively because each creative was attracting different types of buyers.

What surprised me most was the diversity of winning creatives. Some of our best-performing ads came from angles I never would have prioritized under the old audience-targeting approach. A simple product demonstration video outperformed our "sophisticated" lifestyle content. A customer testimonial beat our professionally shot brand campaign.

The creative testing rhythm also solved another problem I hadn't anticipated: creative fatigue. By consistently introducing fresh creative angles, we maintained performance levels that would typically decline as audiences became oversaturated with the same messages.

More importantly, this approach built a sustainable advertising system. Instead of constantly hunting for new audiences to test, we had a proven framework for generating new creative angles that would attract different customer segments within our broad targeting parameters.

Learnings

What I've learned and the mistakes I've made.

Sharing so you don't make them.

After running this creative-first approach across multiple e-commerce campaigns, here are the most important lessons I learned:

  1. Consistency beats perfection - The weekly 3-creative rhythm was more important than having "perfect" creatives. Regular testing kept the algorithm fed with fresh signals.

  2. Broad targeting works better than you think - When you let creatives do the targeting work, you don't need complex audience setups. Basic demographics are enough.

  3. Creative fatigue is real and predictable - Even winning creatives need to be refreshed every 2-3 weeks to maintain performance.

  4. Different creative angles attract different customer values - Lifestyle creatives often brought higher AOV customers, while problem-solution ads drove more volume.

  5. The algorithm is better at micro-targeting than manual targeting - Facebook can find micro-segments within broad audiences more effectively than we can define them manually.

  6. Creative testing is more scalable than audience testing - You can generate new creative angles indefinitely, but audience combinations are limited.

  7. Documentation is crucial - Tracking which creative angles work helps inform future creative development and prevents repeating failed approaches.

The biggest mindset shift was realizing that modern Facebook advertising is about creative strategy, not targeting strategy. Once I stopped trying to outsmart the algorithm with targeting tricks and started giving it better creative tools to work with, everything improved.

How you can adapt this to your Business

My playbook, condensed for your use case.

For your SaaS / Startup

  • Use broad targeting with specific creative angles for different customer segments

  • Test 3 new ad creatives weekly to maintain algorithm performance

  • Focus creative testing on different psychological triggers rather than demographics

For your Ecommerce store

  • Implement systematic creative testing with product demonstration and lifestyle integration

  • Test user-generated content and customer testimonials as social proof angles

  • Track which creative angles drive higher AOV vs higher conversion volume

Get more playbooks like this one in my weekly newsletter