Multi-Touch Attribution
Splitting a conversion across several touchpoints
In short
Multi-touch attribution splits the value of a conversion across several touchpoints in the customer journey instead of crediting only the first or last contact.
Also known as: multi touch attribution, multi-touch attribution model, MTA
Multi-touch attribution is an approach to attribution that splits the value of a conversion across several touchpoints in the customer journey. Unlike first-click attribution or last-click attribution, no single contact gets all the credit.
The main models
- Linear: Every touchpoint gets the same share. With four contacts, that’s 25% each.
- Time decay: Contacts close to the conversion get more weight than early ones, usually controlled by a half-life.
- Position-based (U-shaped): The first and last contact get 40% each, the contacts in between share the remaining 20%.
- Data-driven: An algorithm compares journeys with and without a conversion and derives how much each channel actually contributes. This requires large amounts of data.
Example
Before closing a $4,000 deal, a lead has four touchpoints: Meta ad, Google search, newsletter, branded search.
| Model | Meta | Google search | Newsletter | Branded search |
|---|---|---|---|---|
| Linear | $1,000 | $1,000 | $1,000 | $1,000 |
| Position-based | $1,600 | $400 | $400 | $1,600 |
| First click | $4,000 | $0 | $0 | $0 |
| Last click | $0 | $0 | $0 | $4,000 |
The same journey leads to completely different channel ratings depending on the model.
Multi-touch attribution for lead generation
In theory, multi-touch attribution reflects long decision processes best. In practice, it often fails because of the data: touchpoints get lost through declined consent, device switches and short cookie lifetimes. Many contacts, like a conversation at a trade show or a referral, never show up in tracking at all. Especially for smaller accounts with a few hundred leads a month, there often isn’t enough data for data-driven models.
A pragmatic approach is to look at first click and last click side by side: the first shows which channels create demand, the second which ones close it. Either way, the conversion shouldn’t stop at the lead – it should include deals and revenue from the CRM.
Common mistakes
- Picking the model that confirms your opinion: Every model makes certain channels look better.
- Ignoring incomplete journeys: A model can only split what was captured.
- Too little data for data-driven models: With few conversions, the calculated weights are hardly reliable.
Ready for better tracking?
Frequently asked questions
Articles on this topic
Related terms
All termsAttribution · Assigning conversions to marketing channels
Attribution assigns leads, purchases and revenue to the channels, campaigns and ads that drove them – the basis for every budget decision you make.
First-Click Attribution · The attribution model that credits the first contact
First-click attribution gives 100% of the credit for a conversion to the first touchpoint – the channel through which a user found you for the first time.
Last-Click Attribution · The attribution model that credits the last contact
Last-click attribution gives 100% of the credit for a conversion to the last touchpoint before it – the click that led directly to the lead or purchase.
Customer Journey · The path from first contact to closed deal
The customer journey describes every step and touchpoint a prospect goes through, from the first contact with your brand to the purchase or signed deal.
Touchpoint · A point of contact between customer and brand
A touchpoint is any single point of contact between a prospect and your brand – for example an ad click, a website visit, an email or a phone call.
Attribution Window · The time frame for crediting a conversion
An attribution window defines how many days after a click or view a conversion is still credited to an ad, for example 7 days after the click.