The multi-touch attribution delusion

I have a file of vendor pitches I've kept over the years. Attribution SaaS companies promising "complete customer journey mapping" and "100% touchpoint visibility."

Every one of them had the same fatal flaw: they required tracking to work. And in 2026, tracking is broken.

iOS privacy changes. Corporate ad blockers and firewalls. Browser ITP stripping cookies after 24 hours. VPN usage among executive-level buyers. Third-party cookie deprecation.

The "perfect 360-degree view of your customer" is mathematically impossible. The data doesn't exist.

What you're actually buying

When an MTA vendor says they can show you every touchpoint in the customer journey, they mean: every trackable touchpoint, in the channels where their tracking pixels have coverage, for the percentage of users who haven't blocked tracking.

In industrial B2B, where buyers often research on corporate networks with strict firewall policies and use company laptops where IT has disabled third-party cookies, that coverage might be 30-50% of your actual buyer population.

You're paying $50-200K/year for a highly confident map of half your customers. The other half is invisible.

The real cost

It's not just the subscription. It's the organizational debt.

When executives trust MTA data, they make decisions based on it. Budget shifts from channels that look bad in the model to channels that look good. Often the channels that look bad are doing work that isn't trackable — brand-building, dark social, conference networking.

The model tells a confident, incomplete story. The organization follows that story. Then pipelines dry up from channels that "weren't performing" but were actually doing invisible work.

What actually works

Stop trying to track every touchpoint. Start measuring macro outcomes at the channel level.

Media mix modeling (MMM) uses statistical regression to understand the relationship between aggregate spend levels and aggregate pipeline outcomes. No individual-level tracking required. Works even when individual clicks are invisible.

You run it quarterly. You look at: when we increased spend in channel X by 20%, did pipeline grow? By how much? With what lag?

The answers are directional, not precise. But they're more honest than a pixel-based attribution tool telling you LinkedIn drove exactly 23.7% of pipeline when it can't actually track half the LinkedIn users.

Triangulate with lead source surveys. On every contact form, add one question: "How did you first hear about us?" Not as the primary attribution signal, but as a sanity check against your tracking data.

Look at organic brand search volume. It rises when awareness campaigns are working, even if those awareness campaigns aren't getting click-through credit. An uptick in branded searches six weeks after a new demand gen push tells you something.

Compare pipeline velocity pre/post major spend changes. When you paused paid search for four weeks during a budget crisis, what happened to inbound pipeline volume six months later? That's your incrementality estimate.

None of these are precise. All of them are more honest than a multi-touch attribution tool built on incomplete tracking data.

The conversation to have

When your CFO demands perfect attribution, the honest answer is: "We can't provide perfect attribution because the tracking infrastructure doesn't exist. What we can provide is directional evidence that our investment in X is correlated with pipeline growth."

That's a harder conversation than showing a dashboard. But it's the right conversation. And it leads to better decisions than trusting a confident lie.

Stop chasing perfect attribution. Triangulate toward good-enough attribution. Invest the savings from canceled MTA subscriptions in actual campaign spend.