As marketers, we’ve all faced that uneasy moment when a client asks, “So… how do we know this is working?”
You’re not alone. Recent research shows that 39% of marketing decision-makers struggle to measure the impact of each marketing channel. In truth, that figure may even be optimistic. Measurement has always been a challenge — but in recent years, it’s become exponentially more complicated.
Welcome to what Gen Z might call “a new era” — one defined by stricter privacy, weaker targeting, and the slow, drawn-out demise of the cookie.
The Privacy-First Shift
Not long ago, digital advertising was like the Wild West — data flowed freely, user tracking was simple, and regulation was minimal. Then came GDPR, cookie consent banners, and data privacy laws sweeping across the globe, all of which have fundamentally changed how marketers collect and use personal data.
Apple’s App Tracking Transparency (iOS 14) and the rise of ad blockers have further empowered users to opt out of tracking entirely. The result? A fragmented, increasingly opaque measurement landscape — with giants like Google, Meta, and TikTok now operating within their own data silos.
And while GA4 promised to simplify things, it’s created its own set of challenges:
- Session-level insights are harder to access.
- Attribution models have shifted from last-click to data-driven — but with less transparency.
- Cross-platform visibility remains extremely limited.
Put simply: the marketing ecosystem has lost clarity.
Why Paid Social Measurement Is So Difficult
Paid social has never been a last-touch channel. It influences rather than closes, which makes it inherently tricky to measure — especially when many clients rely on Google Analytics dashboards that still favor last-click attribution.
Imagine a customer who sees ads for a brand on Spotify, Instagram, and connected TV before finally searching the product on Google and making a purchase. Google Analytics will give full credit to Search — even though 84% of paid search conversions are influenced by other channels, according to Meta’s (admittedly self-interested) research.
That’s what I call the streetlight effect: looking for your keys only where the light shines, even though they’re probably in the dark.
The Problem with Platform Attribution
1. Meta’s View
Meta provides robust in-platform analytics — from granular audience data and stronger cross-device tracking to fast cost-efficiency feedback. It’s ideal for optimizing performance within Meta’s walls.
But there’s a catch: Meta often over-claims credit. It can’t see beyond its own ecosystem, assumes ownership of any visible conversion, and can’t be easily validated.
In one client example, Meta reported a dramatic spike in conversions. The truth? The client had launched a major sale and email campaign at the same time. Meta simply claimed credit for purchases that were already happening elsewhere.
2. Google’s View
Google Analytics, meanwhile, under-credits social impact — overlooking impressions, view-throughs, and engagement-driven conversions.
That leaves marketers caught between two extremes: Meta’s over-attribution and Google’s under-attribution. Neither provides the full picture, and both miss offline or cross-channel influence.
Attribution vs. Measurement
It’s essential to distinguish between the two:
- Measurement is statistical — broad, long-term, and privacy-resilient.
- Attribution is user-level — immediate, click-based, and cookie-dependent.
Attribution helps optimize short-term spend, while measurement reveals true long-term effectiveness.
Both matter, but to make smarter strategic decisions, marketers must evolve beyond attribution and toward incrementality.
Understanding the “Messy Middle”
The traditional marketing funnel — Awareness → Interest → Desire → Action — no longer reflects reality.
Today’s journey is nonlinear. Google calls it the Messy Middle — a looping, dynamic space where consumers continuously shift between research, evaluation, and purchase across multiple touchpoints.
The upside? Consumers are more open than ever to discovering new brands. The downside? Proving which touchpoints drove them to act has never been harder.
From Attribution to Contribution
To get closer to the truth, marketers are increasingly combining three complementary methods:
- Multi-Touch Attribution (MTA): Still valuable for optimizing digital performance.
- Incrementality Testing: Isolates a channel’s true impact through controlled experiments.
- Marketing Mix Modeling (MMM): Combines everything into a unified, predictive framework.
Of these, incrementality testing is where first-party data and smart experimentation really shine.
Case Study: Norse Atlantic Airways
When we began working with Norse Atlantic Airways, they didn’t even have a website. Four years later, they’re flying globally with over a million passengers.
As a challenger brand competing against industry heavyweights like British Airways and Virgin Atlantic, Norse couldn’t rely on name recognition or organic demand. Paid social — particularly Meta — became a critical driver of awareness and intent.
However, Google Analytics was showing Meta contributing less than 5% of total revenue, sparking the inevitable question: “Should we just turn off social and see what happens?”
We knew that wasn’t the right test. Instead, we proposed a geo holdout experiment — turning off Meta ads across half of Norway for four weeks while keeping all other activity constant.
What Happened?
Within the first week, revenue in the no-ad regions fell 40% below forecast. The following week, it was still 20% down. We ended the test early — confident that Meta was driving real incremental sales.
When we compared the results, the contrast was striking:
- GA4 said: Meta = <5% of revenue
- Incrementality test showed: Meta = 32% of revenue
It was a clear example of how digital attribution alone can dramatically understate the true impact of paid social.
What’s Next for Norse?
Following the success of that experiment, Norse fully adopted a test-and-learn mindset, which includes:
- Running new incrementality tests for TikTok and PPC.
- Conducting “positive” tests — scaling spend instead of cutting it.
- Building their first Marketing Mix Model to unify channel data and forecast investment impact.
Steps to Improve Your Measurement Maturity
No matter your client’s size or budget, you can start strengthening your measurement approach:
- Get your data foundations right.
Ensure clean analytics, deduplication, server-side tagging, and first-party tracking (like CAPI). - Start with small experiments.
Launch geo holdouts or audience splits to capture real-world impact data. - Build and refine your model.
Feed experiment results into a Marketing Mix Model that evolves with your brand and market.
In Summary
Don’t rely blindly on misleading metrics. Avoid the streetlight effect — the tendency to analyze only where the data is easiest to find.
Instead, start small, start testing, and start measuring what truly matters.
