How to use Attribution Modelling in Google Analytics

How to use Attribution Modelling in Google Analytics

Anna Kowalska

Attribution modelling in Google Analytics is a powerful tool which allows you to determine the value of a marketing campaign or channel. Use it correctly and you’ll be able to view the entire conversion path, answering those all-important questions:

Are your marketing efforts worth the outlay? Is paid search working despite the low conversions? And what keywords are best for conversions?

Given the insightful results, attribution modelling is often overlooked. If you’re not familiar with this tool and never thought to give it a go, it could be worth some of your time. Here’s how to get started with some of the main features

Do you need attribution modelling?

Before we get any further, it’s important to ask yourself whether you really need attribution modelling as it won’t work for everyone. As such, you’ll need to first look at a multi-channel report of your site.

In the overview section, you’ll be able to determine exactly how much ‘overlap’ there is between each channels – whether it’s organic search, paid or direct. This is shown in the overlap between the circles (each circle representing a different channel).

Only when multiple channels show a significant overlap can attribution modelling be used. In this instance, this tool is invaluable as it helps determine or attribute a defining cause to the conversion. There are other ways to access whether attribution modelling is required but this is by far the simplest, and quickest.

How to use Attribution Modelling

Always start by assessing whether the default attribution model is accurately representing the channel performance. There are several model metrics in attribution modelling used to analyse the change in value from one channel to the next.

This includes ‘Last Interaction Model’ which shows all information of the last channel interaction along the conversion path. ‘Last Non-Direct Click Model’ is similar but places all credit on the last channel along the conversion that is not direct.

As the name would suggest ‘Last Adwords Click Model’ gives credit to the last AdWords click while ‘First Interaction Model’ determines which channel initially brought visitors to your site. The ‘Linear Model’ equally shares credit across all channels meaning if they’re was five channels, each would receive 50% credit.

Finally, ‘Time Decay Model’ places more credit on channels that are closer to the end of the path while ‘Position Based Model’ puts most emphasis on the first and last channels before splitting the rest of the credit equally to the channels in between.

Create Custom Channel Groupings

While the above can provide plenty of meaningful results, sometimes it’s important to create custom channel groupings. That may be useful for those who use channels such as email marketing, affiliate networks, unique search engines and so on.

It’s simple and quick to create your own by clicking on the ‘Channel Groupings in the menu drop down. From here you can create a new channel grouping. It’s an effective way to filter data according to your specific business set up. As such, custom groupings are one of the best ways to generate more insightful results. These can also be compared against your sites most important channels.

Have a play around with attribution modelling and you may discover a handy new tool to improve marketing expenditure, saving you time and money.