Sales & Conversion
Personas
SaaS & Startup
Time to ROI
Medium-term (3-6 months)
Here's what happened when I started working with an e-commerce client who was burning through their marketing budget because they believed Facebook's attribution was telling them the truth.
Their Facebook ads showed a beautiful 2.5 ROAS. The team was celebrating, planning to scale, ready to pour more money into what seemed like a winning campaign. But something felt off when I dug into the actual business numbers.
Within a month of implementing a comprehensive SEO strategy, their Facebook ROAS magically jumped to 8-9. Most marketers would pop champagne and claim they'd optimized their ads. But I knew better. The reality? Facebook was claiming credit for sales that were actually driven by organic search traffic.
This is the uncomfortable truth about attribution in 2025: the tracking you think you have doesn't exist anymore. iOS updates, GDPR compliance, and the death of third-party cookies have created what I call the "dark funnel" - where most customer touchpoints are invisible to your tracking tools.
In this playbook, you'll learn:
Why traditional attribution models are lying to you (and costing you money)
How to embrace "dark funnel" thinking for better decision-making
The distribution-first approach that actually works without perfect tracking
My framework for measuring what matters when you can't track everything
How to optimize for coverage instead of control
If you're tired of making marketing decisions based on lies your tracking tools tell you, this is for you. Let's dive into what I learned from building marketing systems that work in the post-cookie world.
Reality Check
What everyone's still trying to do
Walk into any marketing team meeting in 2025, and you'll hear the same conversation that's been happening for the past three years: "We need better attribution." "Can we implement server-side tracking?" "What about UTM parameters?"
The industry consensus is clear: fix the tracking, optimize the funnel, measure everything. Marketing teams are desperately trying to recreate the precision they think they once had. Here's what most are doing:
Implementing complex server-side tracking - Because if Facebook's pixel doesn't work, surely our own tracking will
Adding more UTM parameters - Tagging every possible touchpoint to "see" the customer journey
Investing in attribution platforms - Third-party tools promising to solve what Google and Facebook can't
Creating detailed customer journey maps - Assuming people follow predictable, linear paths to purchase
Obsessing over first-touch vs last-touch attribution - Debating which model is "most accurate"
This approach exists because it feels logical. If we can't see what's happening, let's build better eyes, right? The conventional wisdom says that proper measurement leads to proper optimization, and proper optimization leads to better results.
The problem? This entire approach is built on a fundamental misunderstanding of how customers actually behave in 2025. The linear funnel thinking that worked (or seemed to work) in the Facebook golden age is not just broken - it was always an oversimplification.
Real customer journeys look nothing like our attribution models. Someone might see your Facebook ad, Google your company name, read three blog posts, check reviews on a third-party site, see a retargeting ad, ask friends for opinions, and then buy a week later by typing your URL directly. Which touchpoint gets "credit"? The answer is both all of them and none of them.
Consider me as your business complice.
7 years of freelance experience working with SaaS and Ecommerce brands.
When I started working with this e-commerce client, they had what looked like a textbook Facebook Ads success story. Clean metrics, clear attribution, respectable ROAS. But the business numbers weren't adding up.
The client had built their entire growth strategy around Facebook Ads. They were spending about €2,000 monthly and seeing a 2.5 ROAS according to Facebook's reporting. On paper, this looked reasonable for their industry. The attribution model showed a clean story: user sees ad, user clicks ad, user buys product.
But there were red flags everywhere. Their organic traffic was basically zero. Their brand search volume was minimal. They had no email list growth outside of purchasers. Essentially, they had no distribution channels beyond paid ads - which should have been a warning sign.
Here's what I suspected: their entire business was dependent on a single channel, and that channel's attribution was probably overstating its impact. But I needed to test this theory.
My first approach was traditional. I tried to "fix" their tracking. We implemented better UTM parameters, set up Google Analytics 4 properly, installed server-side tracking for better data collection. We even tried a couple of attribution platforms that promised to solve the "iOS 14 problem."
None of it worked. The data was still fragmented, still inconsistent between platforms, still telling different stories. Facebook said one thing, Google Analytics said another, our server-side tracking showed yet another picture. We were spending more time trying to understand our data than actually improving our marketing.
That's when I realized we were solving the wrong problem. Instead of trying to get perfect attribution in an imperfect world, what if we focused on building distribution that works regardless of what we can measure?
Here's my playbook
What I ended up doing and the results.
I stopped trying to track everything and started thinking about coverage instead of control. The breakthrough came when I implemented what I call a "distribution-first, measurement-second" approach.
Step 1: I built multiple touchpoints without obsessing over attribution
Instead of optimizing Facebook ads based on their reported attribution, I focused on expanding the client's presence across all possible discovery channels. This meant:
Complete website restructuring for SEO optimization
Content creation focused on search intent, not just brand messaging
Email list building through lead magnets
Social media presence beyond just paid ads
Step 2: I embraced "dark funnel" thinking
The customer journey actually looks like this: Google search → Social media browsing → Retargeting ad exposure → Review site research → Email nurture sequence → Multiple touchpoints across channels. Most of these interactions are invisible to traditional tracking.
Instead of fighting this reality, I designed our strategy around it. I assumed that every channel was contributing to every sale, and optimized for maximum visibility across all touchpoints rather than trying to prove which one "worked."
Step 3: I implemented proxy metrics that actually matter
Rather than obsessing over attribution, I focused on leading indicators that predict business growth:
Brand search volume growth
Direct traffic increases
Email list growth and engagement
Organic keyword rankings
Overall business revenue (the only metric that truly matters)
Step 4: I used cohort-based measurement instead of session-based
Instead of trying to track individual customer journeys, I started measuring the performance of user cohorts over time. This meant looking at the lifetime value of customers acquired during specific periods, regardless of which "touchpoint" got credited in our analytics tools.
The results were immediate and eye-opening. Within a month of launching the SEO strategy, Facebook's reported ROAS jumped from 2.5 to 8-9. This wasn't because our ads got better - it was because SEO was driving significant traffic and conversions, but Facebook's attribution model was claiming credit for organic wins.
This confirmed what I suspected: most attribution is fiction. People were finding the brand through organic search, but if they had ever clicked a Facebook ad (even weeks earlier), Facebook got the credit. The "last-click" and "view-through" attribution windows were creating false causation stories.
Coverage Strategy
Instead of trying to track every touchpoint, focus on being visible everywhere your customers might look. Cast a wide net across all possible discovery channels.
Proxy Metrics
Measure leading indicators like brand search volume, direct traffic, and email growth instead of obsessing over attribution data that's mostly fiction.
Dark Funnel
Accept that most customer interactions are invisible. Design your strategy assuming every channel contributes to every sale, even if you can't prove it.
Business Focus
The only metric that truly matters is total business revenue. Everything else is just directional data to help you make better bets.
The transformation was dramatic, but not in the way most case studies would tell this story. The real results weren't about optimizing a single channel - they were about building a business that could grow regardless of what our tracking tools could see.
Revenue Growth: Within six months, the client's monthly revenue had doubled. But here's the kicker - Facebook's attribution still claimed credit for about 60% of sales, even though we knew organic traffic was driving significant conversions.
Channel Diversification: Their traffic sources went from 80% paid ads to a healthy mix: 35% organic search, 25% direct traffic, 20% paid ads, 15% email, and 5% social. This diversification made the business far more resilient.
Brand Search Growth: Monthly brand searches increased by 400%. This was our clearest signal that the distribution strategy was working - people were actively looking for the company by name, indicating strong brand awareness.
Customer Quality: Perhaps most importantly, customers acquired through the diversified approach had 35% higher lifetime value than those attributed solely to Facebook ads. They stayed longer, bought more, and referred others more frequently.
The most telling result? When iOS 15 launched and decimated Facebook's tracking capabilities, this client barely noticed. Their business continued growing because it wasn't dependent on a single channel or attribution model.
What I've learned and the mistakes I've made.
Sharing so you don't make them.
This experience taught me five critical lessons that completely changed how I approach marketing measurement and attribution:
Attribution models create false confidence - The precision of attribution reports is largely an illusion. Clean numbers don't mean accurate numbers.
Customer journeys are messier than our tools can capture - Real buying behavior involves multiple touchpoints across days or weeks, most of which are invisible to tracking.
Distribution beats optimization - Being visible everywhere is more valuable than optimizing a single channel based on questionable attribution data.
Business metrics trump marketing metrics - Focus on total revenue, customer lifetime value, and brand awareness rather than click-through rates and attribution percentages.
Resilience requires diversification - Businesses built on single channels or attribution models are fragile. Multiple touchpoints create antifragile growth systems.
What I'd do differently: I would embrace the dark funnel thinking from day one instead of wasting months trying to fix tracking. The measurement-first approach delayed real progress.
When this approach works best: For businesses with complex buying cycles, B2B SaaS with longer sales cycles, and any company that wants sustainable growth rather than quick optimization wins.
When it doesn't work: If you're running a simple e-commerce store with impulse purchases and need immediate optimization feedback, some level of attribution tracking is still valuable - just don't trust it completely.
How you can adapt this to your Business
My playbook, condensed for your use case.
For your SaaS / Startup
For SaaS companies, implement this approach by:
Focus on building SEO content around use cases and integrations
Track trial-to-paid conversion rates by cohort, not by attribution source
Measure brand search volume and direct signups as leading indicators
Build email nurture sequences that work regardless of initial touchpoint
For your Ecommerce store
For e-commerce stores, apply this framework through:
Invest in SEO for product and category pages, not just paid ads
Track customer lifetime value by acquisition month, not attribution source
Build retargeting that works across email and social, not just ads
Focus on repeat purchase rates as a proxy for attribution accuracy