Using Attribution Modeling in Google Analytics

Using Attribution Modeling in Google Analytics

May 30, 2019 3:01 pm

Marketing attribution is the method of figuring out sources or channels that led to a desired end result, similar to a sale or a subscription. Attribution can turn into difficult when a number of channels contribute.

For instance, a shopper might first click on a hyperlink on Fb to view a product on an ecommerce website. She might revisit that product web page from an e mail promotion. Lastly, she might click on an natural search itemizing, assess the product, and consummate the acquisition. Deciding which of these sources — Fb, e-mail, natural search — is liable for the sale is attribution.

Google Analytics presents attribution modeling at Conversions > Attribution > Mannequin Comparability Device. There are a handful of fashions to pick from. You possibly can examine fashions to determine the effectiveness of every supply.

On this publish, I’ll describe attribution modeling in Google Analytics.

Google Analytics' attribution modeling tools are at Conversions > Attribution > Model Comparison Tool.

Google Analytics’ attribution modeling instruments are at Conversions > Attribution > Mannequin Comparability Device.

Default Attribution Fashions

“Final Interplay” is the most typical mannequin. It attributes the complete conversion to the final contact. Within the instance above, a shopper first visited a website from Fb, then from e mail, and, lastly, from an natural search itemizing, when she accomplished the acquisition. Final Interplay would attribute the complete sale to the natural search itemizing. This mannequin tells you what supply closed the deal. It helps set up the messaging or promotion that satisfied the buyer to buy.

“First Interplay” assigns one hundred pc of the acquisition to the primary supply that introduced the consumer to the website. Within the instance above, the primary supply is Fb. First Interplay is useful in model consciousness.

“Linear” attribution assigns equal weighting to all contact factors. In our instance, Linear attribution would allocate 33 % to every of Fb, e mail, and natural search. This mannequin is useful when calculating return on funding because it provides worth to every contributing supply.

“Time Decay” attribution is just like Linear, the place all sources obtain credit score. Nevertheless, with Time Decay the sources which might be closest to the time of sale would obtain a better proportion. That is useful if monitoring the time of the conversion is essential. For instance, a service provider with a weekend sale might use Time Decay as it might credit score the sources on, say, Friday, versus these from, say, Monday.

“Place Based mostly” takes all sources under consideration. Nevertheless, it assigns forty % to the primary and final interplay and splitting 20 % among the many remaining sources. This mannequin is just like Linear attribution, however it presumes the primary and final contact factors are most necessary.

“Final Non-Direct Click on” excludes direct visitors from getting credit score.

“Final Google Advertisements Click on” attribution assigns worth solely to Google Advertisements.

Customized Attribution

If these aren't sufficient, you'll be able to create as much as 10 customized attribution fashions. Entrepreneurs with Google’s Analytics 360 can use knowledge-pushed attribution, which makes it simpler to create customized fashions based mostly on knowledge from Google Analytics. In any other case, you'll be able to export the info from Google Analytics and run your personal fashions in statistical platforms comparable to SAS or SPSS Statistics.

There are numerous causes to create customized attribution fashions, together with:

  • Distinctive weighting for particular sources. You ran a mannequin outdoors of Google Analytics and realized a given supply persistently produced, say, 60 % of the worth.
  • Timing. You've gotten a selected timetable for the attribution window. For instance, chances are you'll want to assign worth solely from the final 5 days.
  • Particular instances. You could need to assign larger or decrease values to particular key phrases, channels, or sources. Or chances are you'll need to exclude sure sources.
  • Altering the default fashions. For instance, you could need to change the values or order in “Place Based mostly” attribution. For the “Time Decay” mannequin, you could need to assign the very best worth to the primary supply.

Whatever the cause, comply with these steps in Google Analytics to create a customized attribution mannequin.

  1. Click on Conversion > Attribution > Mannequin Comparability Device. From there, choose “Create new customized mannequin” beneath the “Final Interplay” dropdown menu.
  1. Identify the mannequin and choose the “Baseline Mannequin” earlier than altering its weight.
  1. As desired, set the “Lookback Window,” “Regulate credit score based mostly on consumer engagement,” and “Apply customized credit score guidelines.”
Name the model and select the "Baseline Model" before changing its weight. As desired, set the "Lookback Window," "Adjust credit based on user engagement," and "Apply custom credit rules."

Identify the mannequin and choose the “Baseline Mannequin” earlier than altering its weight. As desired, set the “Lookback Window,” “Regulate credit score based mostly on consumer engagement,” and “Apply customized credit score guidelines.”


Determine the aim of your attribution mannequin and maintain it constant over time. For instance, in case you use attribution to determine ROI of every supply, don't use “Final Interplay” in April and “Place Based mostly” in Might. Achieved appropriately, attribution can optimize campaigns, key phrases, and sources to enhance efficiency.

You may also like...