The Pros & Cons of the Most Popular Multi-Touch Attribution Models
If you’re in marketing, you’ve no doubt heard of multi-touch attribution. Multi-touch attribution is the process of assigning credit to all the marketing campaigns that played a part in a conversion. In most marketing and analytics software, 100% of the credit for a conversion is given to either the campaign with the first touch or the last touch, but what about all the other campaigns that touched that prospect (or account). Didn’t those play a role in the conversion?
This is not a post to convince you to start using multi-touch attribution. We’ll assume for the sake of this blog post that you’ve suffered enough in the darkness of single touch attribution and now looking for the best multi-touch attribution model for your business.
Let’s look at 3 different models and the pros and cons of each.
In the Linear model, all prospect interactions are weighted the same. It’s the Communist Party of attribution - each prospect activity gets equal credit (even when some actually create more value).
Pro: It’s better than single touch attribution! While not the best form of Multi-Touch Attribution, it’s clearly a step in the right direction as it gives credit to all marketing campaigns up and down the funnel. It’s also the easiest model to implement.
Con: The Linear Model doesn’t take in consideration that some marketing campaigns are much more influential than others. As Avinash Kaushik explains, this is like giving everyone a participation trophy. It’s idealistic in theory, but simply not how the world works. Some campaigns will be winners and some losers. Your job is to separate the wheat from the chaff and you can’t do that if every campaign has a certificate that says “wheat” on it.
Takeaway: While the Linear model is still better than any single touch attribution model, you can do better.
Time Decay Attribution
Time decay attribution modeling gives credit to interactions that happen closer to the conversion event. Why? Because those touchpoints closest to the conversion were probably the most impactful. The further back from the conversion date, the less of a contribution that campaign made to the conversion. Earlier touchpoints are completely ignored because if they were truly effective why didn’t they influence the conversion to happen sooner? Because B2B sales cycles are typically much longer than B2C, early touchpoints from 6+ months ago aren’t as important as the ones in the weeks or days before conversion.
Pro: The time decay model places the most value on the activity that actually drove the conversion.
Con: Due to long sales cycles in B2B, time decay usually attribution ignores the activity that first introduced the prospect to your company. If you have shorter sales cycles, then time decay may not give you a full picture.
Takeaway: Time decay attribution is a more sophisticated model that allows you to see true campaign effectiveness.
Position Based Attribution
Position based (or U-shaped) attribution is a hybrid model that allows you to sit on the fence a bit. While it spreads credit out across all touchpoints, it gives 40% of conversion credit to the first touch (how the customer was introduced to your brand) and 40% to the last touch (the touchpoint that drove the desired action). The remaining 20% is spread out across the other in-between campaign touches.
Pro: The best of both worlds. Since most marketers value the first touchpoint that introduced customers to your brand and the last touchpoint that drove the conversion, Position Based is a favorite. It also accounts for the fact that other campaigns probably played a small role as well.
Con: In position based attribution, the marketer won’t have data on anything after the lead creation touchpoint. Anything that happens post-lead creation isn’t reported and thus you won’t know which touchpoint led to opportunity creation. This is all well and good if you are only worried about lead reporting, but if you want metrics from further down the funnel you’ll find Position Based comes up a bit short.
Takeaway: Position based gives you the most bang for your attribution buck in a model that rewards the keystone touchpoints (first and last interactions) while crediting those in-between campaigns that make up the rest of the customer journey.
While there are cons with any type of multi-touch attribution modeling, the positives far outweigh the negatives and all 3 models are vastly superior to single touch attribution when measuring the effectiveness of your campaigns. So pick a model that works for your business, be consistent, and start allocating marketing dollars to your most impactful campaigns today.