Growth & Strategy
Personas
SaaS & Startup
Time to ROI
Short-term (< 3 months)
Last year, I was working with a B2B SaaS client whose Facebook ROAS supposedly jumped from 2.5 to 8-9 overnight. The marketing team was celebrating their "improved ad performance," but I knew something was off.
While they were congratulating themselves on better targeting and ad copy, I had just implemented a comprehensive SEO strategy that was driving significant organic traffic. The reality hit me: Facebook's attribution model was claiming credit for my SEO wins.
This wasn't an isolated incident. After working with dozens of clients across SaaS and e-commerce, I've seen the same pattern repeated: tracking pixels lie, attribution models mislead, and businesses make costly decisions based on incomplete data.
Here's what you'll learn from my experience debugging tracking implementation hell:
Why tracking pixels systematically over-credit paid channels
The hidden costs of attribution mistakes I've witnessed
My framework for tracking what actually matters
When to fix tracking vs when to ignore it completely
Alternative measurement approaches that tell the real story
This isn't about perfect tracking—it's about making better business decisions with imperfect data.
Technical Issues
Every tracking "solution" creates new problems
Walk into any marketing meeting and you'll hear the same complaints about tracking pixels. Everyone knows the data is messy, but the industry response has been predictable: add more tracking, implement better pixels, use advanced attribution models.
The conventional wisdom looks like this:
iOS 14.5+ problems: Install server-side tracking and Conversions API
Attribution window issues: Implement multi-touch attribution models
Cross-device tracking: Use first-party data and customer matching
Dark funnel attribution: Add UTM parameters and pixel events everywhere
Pixel firing errors: Implement redundant tracking and validation scripts
This advice isn't wrong—it's just expensive and often unnecessary. The tracking industry has convinced us that measurement problems require measurement solutions. But after years of implementing these "fixes," I've realized something fundamental: perfect tracking is impossible, and chasing it often costs more than the insights are worth.
Most businesses spend more time debugging their tracking than actually improving their marketing. They hire specialists, buy attribution software, and still end up with data they don't trust. The problem isn't the implementation—it's the expectation that digital marketing can be measured like a science experiment.
Every tracking solution introduces new failure points. Server-side tracking needs constant maintenance. Attribution models make assumptions about customer behavior. UTM parameters get stripped by email clients and social platforms.
Meanwhile, your actual customers are taking messy, non-linear journeys that no tracking system can fully capture.
Consider me as your business complice.
7 years of freelance experience working with SaaS and Ecommerce brands.
The moment that changed my perspective on tracking happened while working with a B2B SaaS startup that was heavily dependent on Facebook Ads. Their ROAS had been sitting at a respectable 2.5, and they were spending about €5K monthly on ads with small but steady growth.
The client approached me for a complete SEO overhaul—they had zero organic presence and wanted to build a sustainable traffic channel. This was a classic case of distribution-led growth where they needed to move beyond paid dependency.
Here's what made this project unique: their product had over 1,000 features and integrations, making it perfect for programmatic SEO. Think of it like a complex B2B tool where every feature combination could target different long-tail keywords. The kind of product that's impossible to sell through simple Facebook ads but perfect for capturing search intent.
I spent three months building a comprehensive SEO strategy:
Complete website restructuring for search optimization
Content creation targeting their extensive feature set
Technical SEO improvements for better crawling
Integration pages for every major software they connected with
Within a month of launching the SEO strategy, something "magical" happened: Facebook's reported ROAS jumped from 2.5 to 8-9. The marketing team was ecstatic. They started talking about how their improved ad copy and targeting was finally paying off.
But I knew the truth. The organic traffic from SEO was driving significant conversions, but Facebook's attribution model was claiming credit for users who had first discovered the company through search, then later clicked on a retargeting ad.
This is the dark funnel in action: someone googles "project management integration," finds the client's website, spends time evaluating the product, and then weeks later clicks on a Facebook retargeting ad before converting. Facebook claims the conversion, but SEO did the heavy lifting.
The client was about to double their Facebook ad spend based on the "improved" ROAS. I had to break some uncomfortable news about what was actually driving their growth.
Here's my playbook
What I ended up doing and the results.
After the Facebook attribution wake-up call, I developed a completely different approach to tracking and measurement. Instead of trying to fix tracking pixels, I learned to work around their limitations.
Step 1: Separate Brand vs Non-Brand Traffic
The first thing I do with any client is segment their traffic into brand and non-brand categories. This immediately reveals attribution issues. When "direct" traffic spikes after launching SEO or content marketing, it's usually branded search traffic that tracking systems can't properly attribute.
I use tools like Google Search Console to identify branded keyword growth, which often correlates with increases in "direct" traffic. This simple analysis reveals how much credit other channels deserve for building brand awareness.
Step 2: Time-Based Attribution Analysis
Instead of relying on platform-reported ROAS, I track performance changes over time relative to marketing activities. When I implement SEO strategies, I document:
Baseline performance across all channels before starting
Week-by-week changes in "direct" and "organic" traffic
Corresponding changes in paid channel "performance"
Overall business metrics (total leads, revenue, customer acquisition)
This approach revealed the pattern with my B2B SaaS client: as organic traffic grew, Facebook's reported ROAS improved because more people were in retargeting audiences.
Step 3: Channel Pause Testing
The most reliable way to understand true channel contribution is systematic testing. I work with clients to temporarily pause specific channels and measure the real impact:
Pause Facebook ads for 2 weeks and track total conversion changes
Stop content production and monitor organic traffic decline
Redirect blog traffic to measure SEO's conversion contribution
These tests often reveal that pausing high-ROAS channels has minimal impact on total business performance, while pausing "lower-performing" channels like content marketing causes significant drops.
Step 4: Cohort-Based Revenue Analysis
Instead of tracking clicks and conversions, I focus on customer cohorts and lifetime value. This approach looks at:
Customer acquisition dates relative to major marketing initiatives
Revenue trends by first-touch channel (when reliable)
Long-term retention rates by acquisition source
For SaaS businesses especially, this reveals which channels bring customers who actually stick around and grow their accounts.
Step 5: Embrace the Dark Funnel
Rather than fighting attribution problems, I plan for them. Modern customer journeys are inherently messy:
Google search → Social media browsing → Retargeting ad → Review sites → Email nurture → Purchase
No tracking system can capture this complexity accurately. Instead of trying to track every touchpoint, I focus on expanding visibility across all possible discovery points. More distribution channels mean more opportunities for customers to find and trust the brand, regardless of which touchpoint gets "credit."
Testing Framework
Document baseline performance across all channels before making any tracking changes to understand true impact
Attribution Analysis
Use time-based correlation analysis rather than pixel data to understand which activities actually drive business growth
Channel Validation
Systematically pause specific marketing channels to measure real conversion impact vs. reported attribution
Business Metrics
Focus on total customer acquisition, revenue trends, and cohort analysis rather than pixel-perfect click tracking
The results of this approach have been consistently revealing across multiple client projects. With the B2B SaaS client, we discovered that SEO was responsible for approximately 60% of their growth, even though Facebook was claiming credit for much of it.
Here's what the data actually showed when we implemented proper measurement:
Organic traffic grew from 300 to 5,000+ monthly visitors over 3 months
"Direct" traffic increased by 150% (actually branded search spillover)
Facebook's retargeting audiences grew, improving their reported ROAS
Total customer acquisition increased 40%, but only 15% was attributable to "better" Facebook performance
When we paused Facebook ads for two weeks, total conversions dropped only 12%—far less than the 60%+ that their ROAS numbers suggested. Meanwhile, a temporary redirect of blog traffic caused a 35% drop in total lead generation.
The most important outcome wasn't perfect attribution—it was strategic clarity. The client shifted budget from increasing Facebook spend to expanding their content and SEO efforts. Six months later, their cost per acquisition had dropped 40% and they were less dependent on paid channels.
This pattern has held across dozens of projects: tracking pixels systematically over-credit late-funnel touchpoints while under-crediting awareness and education activities that do the heavy lifting.
What I've learned and the mistakes I've made.
Sharing so you don't make them.
After years of debugging tracking implementations and analyzing attribution discrepancies, here are the key lessons that changed how I approach measurement:
Tracking pixels favor last-touch attribution, systematically over-crediting retargeting and undervaluing top-funnel activities
Business growth often correlates inversely with tracking complexity—companies obsessing over attribution often neglect actual marketing effectiveness
Time-based analysis beats pixel data for understanding true channel contribution
The dark funnel is real and growing—accept that modern customer journeys are unmeasurable
Channel pause testing reveals truth faster than any attribution model
Focus on total business metrics, not channel-specific vanity metrics
Embrace distribution over attribution—be everywhere customers are, regardless of tracking challenges
The biggest mistake I see businesses make is spending more on tracking solutions than the insights are worth. Perfect attribution is impossible and unnecessary. Focus on understanding directional impact and making strategic decisions with imperfect but sufficient data.
Sometimes the best tracking fix is to stop tracking altogether and focus on growing the business.
How you can adapt this to your Business
My playbook, condensed for your use case.
For your SaaS / Startup
For SaaS businesses dealing with tracking pixel errors:
Focus on cohort revenue analysis over conversion tracking
Track trial-to-paid conversion rates by acquisition source
Monitor customer lifetime value trends rather than click attribution
Use time-based correlation analysis for channel performance
For your Ecommerce store
For e-commerce stores struggling with attribution:
Implement systematic channel pause testing
Separate brand vs non-brand traffic analysis
Focus on total revenue trends over ROAS optimization
Track customer repeat purchase behavior by source