Marketing attribution models explained for B2B revenue teams

Attribution is not truth, it is a lens. Choosing the wrong lens leads to cutting the campaigns that actually create demand.

First touch

Credits the channel that created the contact. Best question: where does demand originate? Use it to fund awareness — content, SEO, generative search visibility, events. Weakness: ignores everything that converted the deal.

Last touch

Credits the final interaction before the deal. Best question: what closes? Use it for bottom-funnel budget decisions. Weakness: systematically overvalues branded search and direct traffic, because they are the last click of demand created elsewhere.

Linear

Splits credit evenly across all touches. Best question: which channels participate in winning cycles? Useful in long B2B cycles with many stakeholders. Weakness: treats a webinar and a newsletter click as equally decisive.

U-shaped (position based)

Weights first touch and lead conversion heavily, spreads the rest. Best for teams whose main job is generating and converting qualified leads.

W-shaped

Adds the opportunity-creation touch as a third weighted milestone. The most useful model for B2B with a defined sales process — it credits demand creation, conversion and sales acceptance.

What to do in practice

Do not pick one. Report first touch and W-shaped side by side. When they disagree, that is the interesting information: a channel with high first-touch and low W-shaped credit is creating awareness that your funnel is failing to convert.

Make attribution possible at all

None of this works without hygiene: consistent UTM conventions, a governed source property, offline sources recorded (referral, event, partner, WhatsApp), and closed-won revenue in the CRM. Attribution quality is a data-model outcome, not a report setting.

The number that ends arguments

Cost per closed-won revenue by source, over a period longer than your sales cycle. Everything upstream of that is a proxy.