Why use data-driven attribution? And which Google product to use?

Home Insights Why use data-driven attribution? And which Google product to use?

Jimmy McCann

Head of Digital Strategy

April 10th 2019

Analytics and Data Science

Being able to understand what marketing channels have contributed to converting a customer, and appropriately assign credit to each one involved, is a crucial part of any digital marketing strategy.

Google’s products have a range of attribution models and each vary greatly with the level of customisation and how much of the whole picture they give you. Different models let you allocate how much credit each action gets for your conversions; depending on the model chosen, the credit is given out in different ways.

Why use data-driven attribution?

Data-driven attribution (DDA) uses predictive algorithms to analyse your data to determine which channels, campaigns, creatives and keywords have the biggest impact on conversions. This means that, unlike other attribution models, data-driven attribution can identify which steps throughout the whole journey, increase the likelihood of a customer converting and gives these interactions more credit.

These insights show which streams are most (or least) valuable, allowing brands to allocate their marketing budget accordingly. Funnelling money into the right campaign for your specific business can improve ROI and increase online revenue, without the need for additional advertising budget.

In this blog, we review each of the Google products that has DDA available.

Attribution models:

Last click: full credit is given to the last clicked ad and relevant keyword
Last non-direct click: full credit is given to the last non direct click (this is only available in GA and is the default attribution model)
First click: full credit is given to the first clicked ad and relevant keyword
Linear: credit is equally distributed across all clicks (organic and paid) in the customer journey
Time decay: more credit is given to clicks that happened closer to time of conversion
Position based: the first and last clicked ads are both given 40% credit, and the remaining 20% is split equally across all other clicks
Custom model: you are able to build a completely bespoke model based on any of the predefined models available
Data-driven attribution: credit is distributed based on past data for this conversion.

Google Ads

By default, Google Ads uses a last click attribution model – however, we are able to set up data-driven attribution modelling as long as the following campaign requirements are met:

15,000 clicks and 600 conversions per conversion type in 30 days
10,000 clicks and 40 conversions in 30 days ongoing.

The Google Ads’ DDA only takes paid campaigns into consideration when attributing credit, so other marketing channels that have been involved will not get any recognition for the part they played in the conversion.

While this means the data-driven attribution model in Google Ads does not necessarily provide a holistic overview of your customer journey, it is still useful for optimising keywords and paid campaigns.

Search Ads 360

SA360’s data-driven attribution model can encompass both display clicks, paid clicks, paid social clicks and natural search clicks. To use DDAM in Search Ads 360, you’ll need: 15,000 clicks and 600 Floodlight conversions in the last 30 days. The initial learning period takes 24 hours, so ideally data-driven attribution would be set up at least 24 hours before you want to use it.

To set up data-driven attribution in Search Ads 360: Create Channel Groupings (e.g. generic keywords, brand keywords) > Select which Floodlight activities you want to be analysed in the DDAM > Apply Channel Groupings > Create columns in the interface.

Things to consider:

SA360 ignores search and display impressions, so the DDAM only considers clicks that lead to conversions
You will need to be set up on Campaign Manager and a common set of Floodlight activities to track display clicks
You will need to use Search Ads 360 Natural Search reporting and have Equal attribution option chosen to track natural search clicks
Offline conversions can be used to generate a model if the click ID for conversions is less than 30 days old.

Campaign Manager

Data-driven attribution in Campaign Manager incorporates the same metrics and channels as Search Ads 360, but also includes impression data, offering a slightly more comprehensive view of the customer buyer journey.

To use the Campaign Manager Attribution Modeling Tool:

Go to Reporting > Attribution
Select a Floodlight configuration
Click Attribution Modeling Tool
Click the dropdown next to the default model, and select Create new data-driven model
Name the model
Select all Floodlight activities that represent conversions
Choose Basic Channel Grouping or select a custom channel grouping
Set a custom lookback window.

It takes up to 48 hours for a data-driven attribution model in Campaign Manager to ‘train’ before it is usable.

Google Analytics

There are seven attribution models to choose from in the free version of Google Analytics, but you will need to be an Analytics 360 customer to access data-driven attribution modelling.

GA360 offers a multitude of features and benefits that aren’t available in Google Analytics; you can see all the differences between Google Analytics and Google Analytics 360 here.

Google Analytics 360

Unlike Google Ads and Search Ads 360, Analytics 360 has a holistic overview of all user actions from all digital marketing activity, so creating a data-driven attribution model in GA360 can provide in-depth insight to how each channel contributes towards conversions.

Data-driven attribution models (DDAM) are only available in GA360 if you meet the following initial requirements:

Google Analytics 360 customer
Have either Ecommerce Tracking or Goals set up
Have a Google Ads account with at least 15,000 clicks on Google Search and a conversion action with at least 600 conversions in the last 30 days.

You will also need to meet the ongoing requirement of 400 conversions per conversion type with a path length of 2+ interactions and 10,000 paths in the selected reporting view, over a 28 day period.

Because DDAM in Analytics 360 provides a rounded view of how each channel contributes to conversions, the insights can be used to inform your budget strategy as it highlights how investing in particular channels may provide a higher ROI.

Data-driven attribution is a valuable tool and can provide greater insights into how each channel and marketing campaign impacts the customer buyer journey.

As Google Analytics 360 Resellers we are able to help you on your own attribution journey – contact our expert team for more information.