Google Analytics now provides free cross channel Data Driven Attribution (DDA) Model to all users via the new ‘Attribution’ tool (still in beta):

Earlier the data driven attribution model was available only to Google Analytics 360 users. Now it is available to all Google Analytics users. 

You can access the new ‘Attribution’ tool from within your Google Analytics reporting view. 

In order to get started with this new attribution tool, you would first need to create an attribution project.

Introduction to Attribution Project

Attribution project is like a separate reporting view in Google Analytics which contains set of reports similar to the multi-channel funnel reports.

Following is the list of attribution projects I have access to:

Each attribution project acts like a reporting view. When I click on the name of an attribution project, I can access it. 

Following is an example of the set of reports which are available in each attribution project:

As you can see from the screenshot above, each attribution project provides the following set of reports:

• Conversion Paths

• Conversion Lag

• Conversion Path Length

• Model Comparison

Below these reports, you would see various settings for managing and configuring your attribution project:

• Project settings
• Connected properties
• Conversion types

An attribution project can report conversion totals (which are de-duplicated) across digital channels and can provide consolidated reporting of the performance of various marketing channels.  

In order for an attribution project to work as a reporting view, you would first need to connect your Google Analytics reporting view(s) and conversions to the attribution project.

In the context of Attribution project, a conversion can be a goal conversion and/or ecommerce transaction which has been defined in one or more GA properties.

What that means, if you have goal conversions and/or ecommerce transactions defined in multiple GA properties, you may choose to use them in the same Attribution project. 

Google allows connecting more than one GA property to a single attribution project.

Following is the screenshot of the attribution project which is connected to two GA properties: ‘Even Education’ and ‘Optimize Smart Live Property’:

When you add two or more GA properties to a single attribution project, the project contain attribution results for the two or more properties.

Unless you own multiple websites which are part of the same business, you should add only one GA property to a single attribution project.

You can set up different users permissions for each attribution project. 

If your attribution project is connected to only one GA property, the users who have access to the attribution project can see attribution results for only the connected GA property.

But when your attribution project is connected to multiple GA properties, the users who have access to the attribution project can see attribution results for all the connected GA properties.

This is something which you should keep in mind while connecting multiple GA properties to a single attribution project. 

Note: At present you can not add ‘App+Web’ property to your attribution project.

Google allows you to create up to 50 attribution projects per Google account. 

Google assigns each attribution project a unique ID called the ‘Project ID’:

Prerequisites for creating a new attribution project

Before you create a new attribution project in Google Analytics, you need to decide the reporting view you want to connect to your attribution project. 

Up to now I have been telling you that we connect a GA property to an attribution project. 

But what we really connect to an attribution project is the specific view of the GA property which must meet the following criteria:

#1 Your chosen view can not be a User ID view , rollup view (view used in a rollup property) or Apps+Web view. That is because the Google Attribution doesn’t support such type of reporting views.

#2 Your chosen view should not contain a filter that strip the URL parameters. Reporting views that use filters that strip the URL parameters can not be used for Attribution.

#3 Your chosen view must have Goal Conversion Tracking and/or ecommerce tracking (preferably enhanced ecommerce tracking) enabled with at least a few days of historical conversion data (however more the better). Without conversion data, attribution tool is of no use. 

If your chosen reporting view does not meet any or all of the above criteria then Google recommends to update or create a new reporting view which meet these criteria.

If you want to get the best results from your attribution reports and/or if later on, you want to use the free data driven attribution (DDA) model then select that reporting view for your attribution project which also meet the following criteria:

#1 Your reporting view belongs to the GA property which is connected to as many of your Google accounts as possible. These Google accounts can be: Google Ads, Google Search Console, Google Play, DoubleClick Campaign Manager, Google Big query etc. 

DDA model can analyze data from all Google accounts which are linked to your Google Analytics property. So when you link all of your Google accounts to your GA property, you would get exponentially better insight from your DDA model.

#2 Your reporting view contains at least 30 days of Goal conversion and/or ecommerce data (preferably enhanced ecommerce data) from as many marketing channels (you use) as possible. And this data should be of high quality (free from any tracking issues like ‘duplicate transactions’). 

#3 Your reporting view recorded at least 600 conversions (for one or more conversion type) in the last 30 days so that Google can create a DDA model for your conversion type(s). If your reporting view recorded less than 600 conversions for any conversion type (in the last 30 days) then you won’t be able to use DDA model for that conversion type in your attribution project(s).

#4 Your reporting view contains high volume of high quality data. Your DDA model is only as good as the data you feed to it. If you feed it garbage, it will produce garbage. As a result of flawed data, the data driven attribution insight you get is most likely to be flawed, misleading and could result in huge monetary loss.

Google recommends that you set up consistent URL parameters across all paid marketing channels. Use auto-tagging for Google ads and standard UTM parameters for non-Google ads.

Creating a new attribution project

Follow the steps below:

Step-1: Login to your Google Analytics account and then navigate to your chosen reporting view. Let’s call this reporting view ‘Optimize Smart Main View’ for easy reference.

Step-2: Scroll down and then click on the ‘Attribution’ link at the bottom of the left hand side navigation:

Step-3: Click on the ‘Get Started’ button:

Note: ‘Edit’ permission is required at the account level to create an Attribution project.

Step-4: Select the Google Analytics account from the ‘Account’ menu which contains the reporting view you want to use in your attribution project (in our case ‘Optimize Smart Main View’):

Step-5: Select the Google Analytics Property from the ‘Property’ menu which contains the reporting view you want to use in your attribution project (in our case ‘Optimize Smart Main View’):

Step-6: Select the reporting view from the ‘View’ menu which you want to use in your attribution project (in our case ‘Optimize Smart Main View’):

Step-7: Click on the ‘Next’ button. Your screen should now look like the one below:

Step-8: Click on ‘Select Conversion Types’ button:

Step-9: Select all the conversion types that you want to use in your attribution project (I like to select all the conversion types) and then click on the ‘Save’ button:

Step-10: Click on the ‘Complete Setup’ button:

Once you clicked on the ‘Complete Setup’ button, you will be able to access your new attribution project:

At this point your attribution project can take: 

• Up to 72 hrs to collect enough data (from your reporting view) to create your first attribution model.
• Up to 30 days to collect enough data to create robust attribution models (including DDA model, if applicable in your case).

Best practice is to wait for at least 30 days (from the date you first created your attribution project) before you start analyzing the various reports in your attribution project.

Step-11: Add users to your attribution project. To do that follow the steps below:

#11.1 Click on the ‘Project Settings’ link under the ‘Configure’ section:

#11.2 Click on the ‘User Management’ button:

#11.3 Click on the blue plus button on the top right hand side:

#11.4 Click on ‘Add Users’ item from the drop down menu:

#11.5 Enter the email address of the person you want to give access to:

#11.6 Click on the checkbox next to the permission you want to give to the new user:

You can provide following types of permissions to a user:

1. Edit – A user with ‘Edit’ permission can change configuration data and/or connect the attribution project to additional GA properties. But he can’t manage users.

2. Read & Analyze – A user with ‘Read & Analyze’ permission can see report and configuration data. He can also manipulate data within reports (example: filter a table, add a secondary dimension). But he can’t manage users.

3. Manage Users – A user with ‘Manage Users’ permission can add or delete other users. He can also change users permissions.

#11.7 Click on the ‘Add’ button at the top right hand side:

Step-12: Click on the pencil icon next to ‘Project Details’ if you want to rename your attribution project and/or change the timezone for your attribution project:

Step-13: Click on the pencil icon next to ‘Lookback Window’ if you want to change the lookback window setting to 60 days or 90 days (the default setting is 30 days):

The period of 1 to 90 days prior to conversion is called the lookback window.

The lookback window determines how far back in time a touchpoint should be considered eligible for getting credit for conversion(s). 

For example, 30 days lookback window means, a touchpoint should be considered eligible for getting credit for conversion(s) up to 30 days prior to the date of conversion. 

Similarly, a 90 days lookback window means, a touchpoint should be considered eligible for getting credit for conversion(s) up to 90 days prior to the date of conversion. 

The lookback window applies to all attribution models and all conversion types. Changes to the lookback window apply going forward.

That’s why it is important that you determine and change your lookback window as soon as you create a new attribution project.

I like to keep the lookback window setting to 30 days. 

Step-14: Click on the ‘Connected Properties’ link (under ‘Configure’) if you want to add more GA properties to your attribution project:

Note: When your attribution project is connected to multiple GA properties, the users who have access to the attribution project can see attribution results for all the connected GA properties.

This is something which you should keep in mind while connecting multiple GA properties to a single attribution project. 

Other articles on Attribution Modelling in Google Analytics

1. Touch Point Analysis in Google Analytics Attribution Modelling
2. 8 Google Analytics Conversions Segments You Must Use
3. Default and Custom Attribution Models in Google Analytics
4. Attribution Model Comparison Tool in Google Analytics
5. Which Attribution Model to use in Google Analytics?
6. How to create Custom Attribution Model in Google Analytics

7. How to do ROI Analysis in Google Analytics
8. Google Analytics Attribution Modelling – Complete Guide
9. Guide to Data Driven Attribution Model in Google Analytics
10. Conversion Credit distribution for Attribution Models in Google Analytics
11. You are doing Google Analytics all wrong. Here is why

12. Marketing Mix Modelling or Attribution Modelling. Which one is for you?
13. Introduction to Nonline Analytics – True Multi Channel Analytics
14. How to set up Data driven attribution model in Google Analytics
15. How to valuate Display Advertising through Attribution Modelling
16. Understanding Shopping Carts for Analytics and Conversion Optimization

17. View-through conversion tracking in Google Analytics
18. Understanding Missing Touch Points in Attribution Modelling
19. Guide to Offline Conversion Tracking in Google Analytics
20. How to explain attribution modelling to your clients
21. 6 Keys to Digital Success in Attribution Modelling

22. How to use ZMOT to increase Conversions and Sales exponentially
23. How to Measure and Improve the Quality of SEO Traffic through Google Analytics
24. How to analyse and report the true value of your SEO Campaign
25. How to allocate Budgets in Multi Channel Marketing
26. What You Should Know about Historical Data in Web Analytics

27. Google Analytics Not Provided Keywords and how to unlock and analyze them
28. Selecting the Best Attribution Model for Inbound Marketing
29. Introduction to TV attribution in Google Analytics Attribution 360
30. Cross Device Reports in Google Analytics via Google Signals
31. Data-Driven Attribution Model Explorer in Google Analytics
32. What is Attribution Modelling and why it is the ‘key’ to online business success?
33. How Does Attribution Work?
34. How is Attribution Modelling helpful for e-commerce and non-e-commerce websites?
35. Introduction to Attribution Tool & Project in Google Analytics