Optimize the effectiveness of your marketing mix.
List the problem you’re trying to solve, the information that you need from your attribution model, and how you’ll use that information to solve the problem.
Your goals and the data you need will determine the complexity of your attribution model.
If your marketing mix is limited to a few channels, you can probably just suffice with a basic analytics setup and last click attribution. However, things get more complicated if you’re working with a large number of marketing channels, media platforms, and offline channels.
Choose an attribution model that will help you solve your problem by gathering the specific information that you need.
The default conversion attribution models built into Google Analytics are:
- Last Click Attribution: A transparently simple model that weighs all conversion credit to the absolute last interaction.
- Last Click Non-direct Click Attribution: Same as last click attribution, except it will avoid crediting direct more often than not.
- Last Ads Click Attribution: Gives all the credit to your Google Ads campaigns.
- First Click Attribution: The opposite of last click attribution; it weighs brand awareness interactions much more heavily than intent-based or action-based interactions.
- Linear Attribution: Gives the same amount of credit to every touch point rather than favoring a particular one.
- Time Decay Attribution: Gives the most weight to the channel closest to the conversion.
- Position-based Attribution: Assigns greater weights to the first and last interactions, also sometimes known as the bathtub model.
In Google Analytics, go to Conversions > Attribution, click on Model Comparison Tool, and select an attribution model from the Select Model drop down list.
Make sure that you’ve already set up goal conversion or ecommerce tracking in Google Analytics, you’re only tracking relevant goals, and you’ve imported cost data for all marketing campaigns and channels into Google Analytics.
To get more accurate results, or to evaluate your marketing from different perspectives, use Google Analytics or R to build custom attribution models.
The baseline attribution models found in Google Analytics will certainly give you an answer, but their accuracy can be questionable at times. If you want the most accurate results possible, you’ll need to build your own custom attribution models.
Luckily, Google Analytics allows you to easily shift channel credits around, change time-decay and the lookback window, and create custom credit rules. If you have a good analyst, they can develop custom attribution models using Markov Models in either Google Analytics or R.
Looking at time-based cohorts can help you determine the correlative effectiveness of certain marketing actions. For example, comparing the number of conversions within a specific time-based cohort, such as January to March, against other cohorts can help you better determine the effectiveness of any changes you made in that time frame.