Sales & Conversion
Personas
Ecommerce
Time to ROI
Short-term (< 3 months)
I used to spend hours crafting perfect audience segments for retargeting campaigns. Demographics, interests, behaviors—I'd create these intricate audience maps thinking I was being strategic. Then I worked on a B2C Shopify store that completely changed how I think about Facebook and Google ads.
The store was burning through budget with mediocre ROAS, and everyone was convinced the problem was audience targeting. "We need better data," they said. "Let's create more specific segments." But after analyzing their campaign performance, I discovered something that challenged everything I thought I knew about retargeting.
Here's what you'll learn from my experience with this failed-then-successful retargeting overhaul:
Why detailed audience targeting actually hurt our campaign performance
The simple framework that turned our creative testing into our main growth lever
How we shifted from complex funnels to one campaign structure that outperformed everything
The weekly testing rhythm that kept our ads fresh and converting
Why "creatives are the new targeting" isn't just a buzzword—it's reality
This approach works whether you're running retargeting for a small ecommerce store or a growing SaaS platform. The principles I'll share go against conventional wisdom, but they're based on what actually happened when we stopped overthinking audiences and started focusing on what people see. Check out our guide on ecommerce conversion optimization for more context on testing strategies.
Industry Reality
What every marketer thinks they need to do
Walk into any marketing conference or scroll through any ads-focused LinkedIn feed, and you'll hear the same advice repeated endlessly. The industry has convinced everyone that sophisticated audience targeting is the holy grail of advertising success.
Here's what every marketing guru will tell you about retargeting:
Segment everything - Create different audiences for cart abandoners, product page visitors, and email subscribers
Layer demographics - Age, gender, location, interests, and behaviors should all be precisely defined
Create custom funnels - Different ad creatives for different stages of the customer journey
Use lookalike audiences - Let Facebook find people similar to your best customers
Exclude converted users - Don't waste money showing ads to people who already bought
This advice exists because it feels logical and gives marketers a sense of control. When campaigns fail, it's easy to blame "bad targeting" and create even more granular segments. The tools make it possible to slice and dice audiences in infinite ways, so why wouldn't you?
The problem is that this conventional wisdom was developed when Facebook's algorithm was less sophisticated and privacy regulations didn't exist. iOS 14.5, GDPR, and the death of third-party cookies have fundamentally changed how ad platforms work.
But most marketers are still operating like it's 2018, creating complex audience strategies that the algorithms can't even use effectively anymore. Meanwhile, the real opportunity—creative testing—gets treated as an afterthought.
Consider me as your business complice.
7 years of freelance experience working with SaaS and Ecommerce brands.
When I started working with this Shopify store, they had the "perfect" retargeting setup. Multiple campaigns targeting different audience segments, each with its own budget and creative. Cart abandoners got one set of ads, product page visitors got another, and email subscribers got a third.
The results? Mediocre at best. ROAS was hovering around 2.5, which sounds decent until you factor in their margins. They were basically breaking even while spending hours managing complex audience segments.
My first instinct was to optimize what they had. I dug into the audience performance, adjusted targeting parameters, and tested different exclusions. But after two weeks of traditional optimization, nothing meaningful changed.
That's when I noticed something interesting in their account. The few ads that were actually performing well weren't tied to any specific audience strategy. They were just good creatives that resonated with people, regardless of which segment saw them.
One lifestyle-focused video ad was crushing it across all audience types. A problem-solving static image was converting cart abandoners and cold traffic equally well. Meanwhile, our "perfectly targeted" ads were flopping even when shown to their ideal audiences.
The data was telling me something the industry advice wasn't: the creative was way more important than the targeting. But I needed to test this hypothesis properly, which meant restructuring their entire approach.
The client was skeptical. "But won't we waste money showing cart abandonment ads to people who've never visited our site?" they asked. I convinced them to test it for 30 days, and what happened next changed everything.
Here's my playbook
What I ended up doing and the results.
Here's exactly what we did to transform their retargeting from complex audience management to simple creative testing:
Step 1: Campaign Consolidation
Instead of multiple campaigns with different audiences, we created one broad retargeting campaign. Anyone who had visited the website in the past 180 days was fair game. No demographics, no interest layering, no behavioral targeting—just website visitors.
Step 2: The Creative Testing Framework
We established a weekly rhythm of launching 3 new creative variations. Each week, we'd test different angles:
Week 1: Lifestyle vs. Product-focused imagery
Week 2: Problem-solving vs. Benefit-driven copy
Week 3: Video vs. Static formats
Week 4: User-generated content vs. Professional photos
Step 3: Budget Allocation Strategy
Rather than splitting budget across audience segments, we let the algorithm decide. Facebook got the full budget to optimize for conversions, and we monitored which creatives were driving results.
Step 4: Performance Measurement
We stopped obsessing over audience-level metrics and focused on creative performance. Which ads had the highest CTR? Which drove the most conversions? Which had the best ROAS? The data guided our next creative tests.
Step 5: Creative Retirement Process
When an ad's performance declined (usually after 2-3 weeks), we'd retire it and replace it with a new variation. This kept our ads fresh and prevented audience fatigue.
The key insight was that Facebook's algorithm had become sophisticated enough to find the right people if we gave it good creative signals. A compelling lifestyle image would naturally attract lifestyle-conscious buyers. A problem-solving video would resonate with people actively looking for solutions.
By focusing our energy on creative quality instead of audience complexity, we turned our ads into the targeting mechanism.
Creative Quality
Focus on making ads that naturally attract your ideal customers rather than trying to find them through targeting.
Weekly Testing Rhythm
Launch 3 new creative variations every week to keep your campaigns fresh and prevent audience fatigue.
Algorithm Trust
Let Facebook's machine learning find your customers—it's better at this than manual audience creation.
Performance Simplicity
Track creative-level metrics instead of audience-level data to make better optimization decisions.
The transformation was immediate and sustained. Within the first week of the new approach, ROAS jumped from 2.5 to 4.2. But more importantly, it stayed there.
Over the 90-day test period, we saw consistent improvements:
Average ROAS increased to 4.8 (68% improvement)
Cost per acquisition dropped by 35%
Campaign management time reduced by 75%
Creative performance insights led to better organic content
But the most surprising result was how this simplified approach actually gave us better customer insights. When you're not constraining the algorithm with complex targeting, you see who naturally responds to different messages. This data informed everything from email marketing to product development.
The client went from spending hours weekly on audience optimization to focusing entirely on creative strategy. They started seeing their ads as content marketing that also drove sales, which improved their overall brand presence.
What I've learned and the mistakes I've made.
Sharing so you don't make them.
Here are the key lessons from completely overhauling our retargeting approach:
Creative is the new targeting - In the post-iOS 14.5 world, good creative does more for audience selection than manual targeting ever will
Complexity doesn't equal sophistication - Simple campaigns often outperform complex ones because they give algorithms more data to optimize with
Test cadence matters more than test volume - Consistent weekly testing beats sporadic campaign overhauls
Algorithm trust requires patience - It takes 7-14 days for Facebook to optimize properly, so don't panic and change things too quickly
Creative insights inform everything - What works in ads should influence your email marketing, product descriptions, and content strategy
Retirement is as important as creation - Knowing when to kill underperforming creatives prevents wasted spend and audience fatigue
Budget consolidation beats segmentation - One well-funded campaign usually performs better than multiple underfunded ones
If I were doing this again, I'd start with creative testing from day one instead of trying to optimize targeting first. The biggest mistake was assuming that conventional wisdom about audience segmentation still applied in today's privacy-focused advertising landscape.
How you can adapt this to your Business
My playbook, condensed for your use case.
For your SaaS / Startup
For SaaS companies implementing this creative-first retargeting approach:
Test different problem-solution angles weekly—feature benefits vs. outcome-focused messaging
Use trial signup data to inform creative themes rather than complex audience targeting
Focus on video demos and social proof rather than traditional display ads
For your Ecommerce store
For ecommerce stores adopting creative-focused retargeting:
Rotate between lifestyle imagery and product-focused creatives based on performance data
Test user-generated content against professional photography weekly
Use seasonal and trending themes to keep creative fresh and relevant