A Blueprint to Multi-Touch Attribution
Multi-Touch Attribution 101
In the ever complex marketing world, understanding the effectiveness of various touch points in a customer's journey is crucial for optimizing marketing strategies and maximizing return on investment. Multi-Touch Attribution (MTA) refers to a model of attributing credit for a conversion or sale to multiple touch points along a customer's journey.
The result is a more accurate representation of what media is driving impact, and what optimizations can be made to increase business impact.
This is the standard user journey - multiple exposures to a single brand’s ad across various channels “touch points”
The goal of MTA modeling is to help define which media channel should get the credit for a conversion.
Mapping Attribution
There are a number of ways to attribute value to the media customers are exposed to. Before the emergence of ad tech measurement tools, brands traditionally used a single touch point attribution methodology like “last touch”. Now, with machine learning and out of the box MTA software performance marketers can see more complex multi touch views: position, linear, U-shaped.
But, these views still don’t provide the sophistication and accuracy brands need in today’s complex media sphere.
Having the right approach to MTA modeling is one of the most critical components in understanding the impact a brand’s media spend is having, and more crucially how to drive even greater scale.
AKA, how do you get more juice out of the squeeze.
3 Unique Elements Make Juice Media’s MTA a Game Changer for Brands
Custom Built ML and AI Models
Juice Media’s MTA product suite brings together two of the godfather’s of math:
Lloyd Shapley and Andrey Markov to provide brands with not just one, but two highly sophisticated custom models.
Shapley Values
Shapley values provide a method of calculating payouts to players in a cooperative game, where players are part of a coalition to achieve a shared aim. In the case of media attribution, the payout would be a conversion. Each player may contribute unequally to the outcome of the game; some players may deserve the lion’s share of the credit, and others may barely increase the coalition’s probability of winning.
Markov Chain
The Markov Chain is a mathematical model that represents a sequence of events and the probability of transitioning from one “state” to another. An MTA model assumes that a customer's journey through various media touch points can be modeled as a series of states, each representing a touchpoint and the transitions between states represents the customer moving from one touch point to another.
Providing advanced mathematical models alongside Last Touch and U Attribution in a self-serve dashboard gives brands a dynamic and insight rich view of the effectiveness of their media spend and how to more accurately assign credit, allocate their weekly or monthly media budgets, and better understand overall return on ad spend (ROAS) and critically customer acquisition cost (CAC) at a granular level.
Within Channel Model for OTT
As streaming ad viewers get increasingly exposed to ads on multiple platforms, it is
critical to understand which platform, or what sequence of exposures are driving conversion. The Juice Media MTA model creates an exposure string and assigns credit across digital
publishers and streamers to ensure pinpoint accuracy on what channels are driving conversions, verses assigning credit to all or just the last publisher.
This type of actionable insight leads to highly performant digital video campaigns for brands looking to scale their reach and find new audiences via OTT, CTV and online video.
Predictive Optimizations
Juice Media’s Optimizer triangulates the Markov and Shapely MTA models, giving both equal weighting, and layers on top historical benchmark data captured over the past decade of scaling brands, and response curve analysis to account for diminishing returns.
A set of recommendations on how to best distribute media dollars for optimal conversion and scale potential is delivered based on this intelligence. Brands see the impact of these recommendations rapidly with real time bidding via the Juice Media DSP alongside optimizations driven by an internal team of media buying pros.
MTA in Action
Juice Media’s MTA model identified the huge synergistic effects of layering the OTT channel on top of search campaigns, resulting in a 96% increase in conversion rate for a major ecommerce customer.
For a leading SAAS platform Juice Media’s MTA model identified that a single exposure was driving 80% of conversion within one sales funnel, enabling a significantly more efficient allocation of media spend.
AI optimizations generated from the MTA model for a leading D2C brand drove a 56% cost efficiency in the brand‘s OTT holiday mix enabling continued spend during a critical seasonal period.
Ready to drive more impact with your media?
