The Debrief
L7L14L30L90All
PaidSearchIndustryTechDataBrandConversion
Data · 3 min read23 May 2026

The Web Is Now as Fragmented as Mobile Was a Decade Ago. The Measurement Fix Is the Same One Mobile Already Built.

Digiday's new piece argues that the web's signal degradation, cookie loss and platform divergence look identical to what mobile faced a decade ago. The fix is the same mobile playbook: independent attribution, server-side postbacks and unified web-and-app measurement above platform self-reporting.

The signals are degrading. The platforms are diverging. The customer journey does not stay in one place. That is mobile in 2015. That is the web in 2026.

3 min read

Digiday just put a name on the thing every web marketer is now living through. The fragmentation problem on the open web (degrading cookies, divergent platform reporting, broken referral paths, contradictory conversion numbers) is the same problem mobile faced a decade ago. The mobile industry already built the fix. The web is finally borrowing it.

The mobile measurement discipline grew up around three structural realities. The OS owners (Apple and Google) controlled the rules. The platforms (Meta, TikTok, Snap) all wanted credit for the same install. And every platform's self-reported numbers added up to more than the actual install count. Sound familiar.

The mobile playbook has three components. Independent attribution that sits above platform self-reporting, so performance claims from different networks reconcile against a consistent standard. Server-side postbacks that send validated conversion signals back into ad platforms for optimisation. And unified measurement across web and app, so the customer journey is not fragmented by the device boundary.

The web has not had an MMP-equivalent layer at scale. Last-click via GA4 was the default. Platform self-reported conversions on top. Anyone trying to triangulate the truth has had to do it manually in a spreadsheet, and most teams either gave up or stopped looking.

Why it matters

For Australian marketers running multi-platform campaigns across Google Ads, Meta, TikTok, LinkedIn and Microsoft, the platform-attributed conversions almost always add up to 200-400% of the actual conversions. Every platform claims credit. Most platforms inflate it.

The mobile playbook is the only practical answer because it does not require the platforms to cooperate. It sits above them. You feed validated server-side signals back to each platform's optimisation algorithm and you keep a single source of truth for the business.

200-400%

Approximate range of platform self-attributed conversions as a percentage of actual conversions in most mid-market multi-channel accounts

The mobile measurement layer also opens up the harder questions. Where in the journey did the channel actually contribute. Did the assist matter or did it cannibalise. Are paid social CPAs improving because the channel is working harder or because they are stealing credit from organic search. None of those questions have honest answers in a last-click GA4 dashboard.

What to do about it

Pick one independent measurement standard for your business this quarter. It can be a server-side GA4 model, a CDP-based attribution, a media-mix model from Mutinex or Meridian. The choice matters less than the discipline.
Audit your conversion API setup on Meta, TikTok, LinkedIn and Google. Most accounts have it half-wired. The half that does work fixes 60% of the signal loss with no platform cooperation needed.
Set up server-side tagging if you have not already. Browser-side measurement is now structurally unreliable and will only get worse.
Build a monthly reconciliation between platform-attributed conversions and your independent source of truth. Document the gap. Adjust the spend mix based on the truth, not the platform claims.
Stop letting the platforms be the scoreboard. The platforms have a structural conflict of interest in their own reporting. That is not a moral judgement. It is the basic economic incentive of any ad network.

The quiet implication is that the marketers who treat measurement as infrastructure (not as a dashboard) over the next 24 months will outperform the ones who do not. The win is not in finding the right tool. It is in choosing one consistent truth and running the business off it.

Share this brief
Send it to a colleague who'll find it useful.
Filip Ivanković
The Debrief / From Filip Ivanković
One every morning. Six months in, you'll see the patterns most don't.
Strategy, benchmarks, and what's actually moving in Australian marketing. Four-minute read. The reps compound.
Filip Ivanković·Founder, New RebellionLinkedIn