By Will Burghes, Head of Professional Services, DV Rockerbox
Multi-touch attribution (MTA) models are marketing analytics methods used to identify and measure the impact of individual marketing channels or campaigns on customer acquisition and sales. MTA models provide a comprehensive view of how potential customers interact with a business's marketing efforts across different channels, devices and touchpoints before completing a purchase or taking a desired action. By analyzing these touchpoints, businesses can better understand which channels most effectively drive conversions and optimize the customer journey going forward.
The four most common MTA models are linear, time decay, position-based and data-driven. These models vary in complexity and accuracy, each with its own strengths and weaknesses.
Linear or even-weight models assume all channels or touchpoints in the customer journey have equal impact on conversion. This simple, straightforward model assigns equal credit to all touchpoints a customer interacted with before converting. It is the most basic model, but is also the least nuanced of the MTA models.
If the linear model were a soccer game, all players would get the same credit for scoring a goal. Even-weight is more sophisticated than a single-touch model; it doesn't ignore all the touchpoints in the middle of the conversion path. But not all touchpoints are created equal. This model essentially gives every soccer player credit for participating, which is unfair to the rockstars.
Time decay models, also known as recency models, assign more credit to touchpoints closest to the conversion. This model helps teams prioritize recent interactions over older ones or businesses that know their path to conversion is relatively short. It captures the impact of touchpoints better, but it can still miss the impact of touchpoints further back in the journey.
With a time decay model, the soccer players who touched the ball right before the goal was scored receive the most credit. Players who touched the ball earlier in the play still get some credit, just less. While time decay acknowledges that not all touchpoints are created equal, it's still flawed. Let's say someone reads a powerful article on your website that makes them want your product. The next day, you retarget this person with an ad. Is the retargeting ad really doing more heavy lifting in earning the conversion?
Position-based models assign different levels of credit to marketing channels based on their position in the path to conversion. This model assigns the most credit to the first and last touchpoints, with the remaining touchpoints receiving less credit. This model helps businesses prioritize the touchpoints most likely to influence the customer's purchase decision.
In a soccer game, the players who initiate play and score the goal get the most credit, while the remaining players are still acknowledged equally, but to a lesser degree. If the touchpoint that grabs customers' initial interest and the touchpoint that ultimately converts them are deemed most important for an advertiser, this model works well. The drawback is that if your business really needs a second or third touchpoint to convert a customer, this model won't properly acknowledge it.
Data-driven or machine learning-driven models are the most advanced and accurate of the MTA models. They use machine learning and artificial intelligence to analyze customer journeys and identify the channels and touchpoints that most impact conversion. This model suits businesses with the resources to invest in data-driven marketing analytics and a need for the most accurate view of how their marketing efforts influence customer behavior.
In the soccer game, each player's value would be determined by their performance across multiple plays. Implementing a machine learning-driven MTA model used to be challenging. For instance, it requires access to historical data on your marketing activities, as well as a tool to analyze performance patterns. Despite these challenges, this model brings you much closer to determining the true value of each marketing touchpoint.
MTA models help businesses understand how customers interact with their marketing efforts across channels and touchpoints. The most common MTA models are linear, time decay, position-based, and data-driven, each with its own strengths and weaknesses. When choosing an MTA model, businesses should consider their marketing channel mix, needs, resources and the nuances of their customers’ typical paths to conversion to get the most actionable view of marketing performance.
Contact a DV representative.
Learn more about DV Rockerbox.