Multi-touch attribution (MTA) provides insights into the impact of each interaction on a customer’s conversion journey, but there is more than one multi-touch attribution model out there, and not all of them may be the right fit for your business.
Before you implement an MTA strategy, it’s important to familiarize yourself with the ins and outs of each model. Below is a breakdown of some of the most common MTA models you can use to gain valuable insights into the customer journey.
What Is Multi-Touch Attribution?
At its core, multi-touch attribution is a means of understanding which marketing efforts — be they ads, emails, or social media posts — contribute the most to sales or customer sign-ups. Its goal is to identify every interaction a consumer has along the customer journey and then determine which touchpoints had (and continue to have) the biggest influence on your audience.
In determining the impact of each interaction, MTA models assign credit to various touchpoints. For instance, say that a customer interacted with your brand five times before making their purchase. Theoretically, under an MTA model, each touchpoint could receive 20% of the credit for the conversion.
Why Do MTA Models Matter?
Choosing the right attribution model is like picking the right ingredients for a recipe: Use the wrong one, and you might devote your time, attention, and resources to a touchpoint that ultimately has minimal impact on your audience. The appropriate model, on the other hand, will provide relevant insights you can use to understand and improve the customer journey.
Common MTA Models
There are five core MTA models that you’ll encounter. These are as follows:
First-touch attribution (FTA) gives most (if not all) of the credit to the first interaction someone has with your brand. An FTA model functions on the premise that all subsequent interactions would not have occurred if not for that initial touchpoint. Though it is a straightforward approach, it can oversimplify things, especially if your customers navigate complex journeys before making their purchases.
In contrast to the latter, last-touch attribution (LTA) models credit the last interaction the prospect had before converting and making their purchase. For instance, a customer may click on an email link and proceed to buy something from your business. If you are operating under an LTA model, you will assign the email with most or all of the credit for the conversion.
Like an FTA model, however, an LTA approach oversimplifies the customer journey, largely ignoring everything that led up to the final touchpoint.
Under linear attribution, you distribute credit across every documented touchpoint and value all of them as equally important parts of the customer journey.
The downside to the approach, however, is that it doesn’t differentiate between interactions that had a huge impact on a buyer’s journey and those that minimally influenced their decision.
A time-decay attribution (TDA) model is like an LTA approach but with a twist: The closer the interaction is to the sale, the more credit it gets. For instance, in a scenario in which a new customer interacted with your brand five days in a row, the distribution of credit under a TDA model might appear as follows:
- 1st Interaction: 5%
- 2nd Interaction: 10%
- 3rd Interaction: 20%
- 4th Interaction: 25%
- 5th Interaction/Conversion: 40%
All interactions receive a share of the credit, but later touchpoints are deemed more valuable. As such, TDA models account for the lead nurturing process and demonstrate the importance of following up with promising leads.
Position-Based Attribution (U-Shaped)
A U-shaped attribution model gives most of the credit to both the first and last interactions, with some recognition spread out in between. For instance, the first and last touchpoints might receive 40% of the credit each, and the remaining 20% of the credit will be divided among every interaction that occurs between the two.
Which MTA Model Is Right for Your Business?
When exploring MTA models, you need to consider such factors as:
- The Customer Journey: If most customers come from a single ad or email, FTA or LTA models might work; if the journey is more complex, consider a linear or time-decay approach
- The Length of Your Sales Cycle: The longer the sales cycle, the more complex your chosen MTA model needs to be, and vice versa
- Your Marketing Channels: Linear or position-based models are often better fits when several marketing channels are in play
If you are new to MTA as a whole, consider starting with a simpler model, like FTA or LTA, and if the model you choose doesn’t provide enough insight into the customer journey, pivot.
You may have to try multiple MTA models before you find the one that is the best fit for your organization, but once you nail it, you’ll have access to timely, relevant insights and harness the ability to create frictionless customer journeys.