Which attribution model works best for me is a logical question that often pops up when you first start using Attribution.
Unfortunately the answer isn't as clear as the question.
The reason for this is because the attribution model you choose is situational, as in there are certain times you'll want to use specific models oppose to others.
Luckily with Attribution you can easily adjust your model on the fly and use the date picker to evaluate your campaigns & channels.
Attribution model dropdown:
It is a generally accepted principal that when you want to do top of funnel analysis to find the return on ad spend for subscribers or leads you'll want to use a first-touch attribution model, and when you are looking to see bottom of the funnel analysis like new paid subscribers or purchases you'd want to use the last-touch attribution model.
eCommerce example - last touch:
Now say for example you are an eCommerce company and you're entering a big selling month like December, in this case you may want to adjust your date range to December 1st - 31st and set your model to last touch, this would provide insight to your return on ad spend for bottom of the funnel events like purchases and allow you to analyze the channels and campaigns that are leading to purchases.
SaaS example - first touch:
In contrast to the example above say you are a SaaS company with a trial. In this case you may want to generate sign ups or new trials for the month of January, you're planning a big Google ad words push and want to see which keywords are drove the most sign ups. In this scenario you may want to consider a date range set from January 1st - 31st with the model set to first-touch, here you would get detailed analysis on which keywords drove the most trial started events.
eCommerce & SaaS - Linear & Position based (U shaped):
Linear and Position based are two models that can be used more generically when you want to look at your overall marketing performance in a multi-touch environment.
In this case you may want to see your return on ad spend for sign ups in Q1 so you'd set your date range from January 1st - March 31st and your event to sign up. With the linear attribution model applied each touch would get an equal amount of credit for a conversion within the date range, with position based you can be more particular with how much credit each touch gets depending on its place in the visitors journey within the selected date range. You're welcome to adjust these values in Settings.
If you'd like to learn more about the specifics of each attribution model please check out our article here.
If you have any questions or concerns about this please don't hesitate to reach out by starting a chat!