Attribution is Broken. Here's What to Do Instead.

The Attribution Fantasy

Here's what typically happens: a marketing team invests in a multi-touch attribution platform. They spend 3-6 months implementing it. The platform produces beautiful dashboards showing fractional credit across channels. And then nobody changes their behavior.

Why? Because the models are built on flawed assumptions. They assume linear customer journeys. They can't account for offline touchpoints. They're blind to brand awareness effects. And they break completely when cookies are blocked — which is now the majority of traffic.

What We Do Instead

We use a three-layer approach that's less elegant but far more actionable:

• Layer 1: Ask. Add a "How did you hear about us?" field to your forms. It's not scientific, but it captures awareness channels that no tracking pixel will ever see.
• Layer 2: Isolate. Turn channels on and off one at a time. Pause Google Ads for two weeks. Did leads drop? By how much? That's your actual attribution for that channel.
• Layer 3: Track margins, not clicks. Stop optimizing for cost-per-click. Start optimizing for profit-per-channel. A $50 lead from SEO that converts at 15% is worth infinitely more than a $5 lead from social that converts at 0.3%.

Perfect attribution is a fantasy. Good-enough attribution that drives action is worth more than a perfect model that nobody uses.

The Incrementality Question

The real question isn't "which channel gets credit?" It's "what happens if I stop spending on this channel?" That's incrementality testing, and it's the only attribution method that produces genuinely reliable answers.

For most businesses under $10M in revenue, the isolation method (Layer 2) is more than sufficient. You don't need a $50K/year attribution platform. You need the discipline to run controlled experiments on your media spend.