Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
LinkedIn has made some improvements advertising attribution modelsto ensure that its data is more representative of the actual response to the ad, as opposed to the assumptions of traditional ad methodologies.
Because as marketers know, most attribution models are based on assumptions, like last click attribution, that don’t really take into account the specific nuances of how modern web users react to and engage with linked content.
This is especially true for B2B campaigns, LinkedIn’s bread and butter, and to address this, LinkedIn is now introducing a new attribution methodology that takes a completely different approach.
As he explained LinkedIn:
“Methodologies such as Multi-Touch Attribution (MTA) and Marketing Mix Modeling (MMM) take into account a wider range of factors and offer a more balanced view of the customer journey, from initial awareness to final conversion. At LinkedIn, we leverage the complementary value of the MMM and MTA approaches and have developed a unique system that bridges the two methodologies in our attribution suite. We have successfully implemented a system for our internal marketing (i.e. marketing for LinkedIn products) and will leverage this methodology for advertisers on the LinkedIn Marketing Solutions platform.”
The new approach, which incorporates elements of both of these alternative tracking systems, uses multiple data points within LinkedIn’s assessment process to better gauge user response across all of these factors, as opposed to making assumptions based on limited sampling.
Although the full explanation is quite technical.
For example:
“Positional representations are combined with sequential touchpoint data generated by members. These sequences are entered through the self-monitoring module. We merge the member and company views and pull them through the dense layer to create the current member view. The member representation and the output of the attention layers are combined and fed through the classification head for the learning task.”
Yes, it’s not built to be layman-friendly, but if that sequence of words gives you a headache, LinkedIn says it’s built a system that takes into account multiple factors, connected via a neural network, that can better track and measure audience reaction to promotions, how the performance of the campaign would be more accurately attributed.
And in its initial testing, LinkedIn says it led to significant improvements in results.
“For example, work [in testing] looked at the performance of both models for early and mid-funnel campaigns across video ads, digital display and social media. When comparing both models for non-search channels, Modeled Attribution was able to recognize and deliver credit, while the last click remained the same. This is due to the model’s ability to connect impressions from those campaigns in the user journey, which RBA models cannot do. Initial results show a 150x increase in credit found in Modeled Attribution, which is consistent with an increase in Marketing spend over this timeframe.”
In other words, LinkedIn’s improved tracking process was able to better track user responses based on broader tracking and attribution, providing more accurate insight into how ads drive actual user response.
Which is key to making sure you’re allocating your ad spend to the right elements.
“LinkedIn marketing has historically relied on rules-based attribution (RBA) based on the last click, where full credit for the conversion point was given to the last click event. This re-indexed the credits towards low-flow channels that convert demand, such as search or email. However, last-click attribution underestimates the value of the top and middle channels, leaving marketers unable to see their performance or how to optimize it.“
So, instead of doubling down on elements that may not be key drivers of your results, you’ll now have better insight on which to make spending decisions.
Which should lead to improved results.
LinkedIn says it is rolling out this new methodology to all advertisers.