To help B2B marketers gain a more reliable understanding of advertising responses, LinkedIn is updating its advertising attribution models with a new hybrid attribution methodology that the business-centric social-media company says offers “a more balanced view of the customer journey.”
While traditional attribution models help advertisers track the customer journey in response to their campaigns through assumptive models -- such as rule-based attribution (RBA) methods and data-driven attribution (DDA) methods -- LinkedIn contends that these approaches include blind spots, overlooking key stages of the customer journey.
In updating its own models, LinkedIn is zeroing in on other methodologies like multi-touch attribution (MTA) -- a bottom-up approach where member-level touchpoints are modeled and the touchpoint journey is reported -- and marketing-mix modeling (MMM), a top-down approach where channel touchpoints are modeled and account for seasonal and macroeconomic factors.
advertisement
advertisement
“We leverage the complementary value of both MMM and MTA approaches and have developed a unified system bridging the two methodologies in our attribution stack,” the company says.
LinkedIn says it has already deployed the system for its internal marketing, and plans to leverage the methodology for advertisers on the LinkedIn Marketing Solutions platform.
“Positional representations are combined with the sequential touchpoint data generated by members. These sequences are fed through a self-attention module,” LinkedIn explains, adding: “We concatenate member and company representations and feed these through a dense layer to create a representation of the acting member. The member’s representation and the output of the attention layers are combined and fed through a classification head for the learning task.”
In less technical language, LinkedIn’s new system takes into account more data points connected within a neural network that, according to the company, has the power to better track and measure audience response to advertisers’ in-app promotions, ultimately more accurately attributing campaign performance.
As for the results, LinkedIn says initial testing has resulted in notable improvements.
When a business, which LinkedIn did not specify, utilized the company’s new system for ads across video, digital display and social media compared LinkedIn’s traditional system with its current attribution system, the improved tracking process was better able to track user responses on broader tracking and attribution, providing more reliable insight in regards to how ads are eliciting user response.
“When comparing both models for non-search channels, Modeled Attribution was able to recognize and deliver credit whereas Last Click remained flat,” the company explains, adding that initial results showed a 150x increase in credit found in Modeled Attribution, “which paces well with Marketing’s spending increase during this time frame.”
LinkedIn says it has begun rolling out its new attribution methodology to all advertisers.