Life (and data analytics) would be a lot simpler if a user immediately purchased or filled out a lead form immediately after engaging with a paid media ad or organic post.
Unfortunately, this is rarely the case.
Most people will visit your site several times prior to converting. They’ll engage with many touchpoints before purchasing -- from reading blog posts to engaging with paid YouTube video campaigns.
So which interaction deserves the credit? The blog posts or your video campaign?
With multiple touchpoints in a user’s journey, it’s difficult to showcase to your clients and higher ups the value of each channel. These discussions are important to deciding budgets and importance of upcoming projects.
This is what attribution modelling is trying to answer.
What are marketing attribution models?
Attribution modeling allows a business to analyze which touchpoints, or marketing channels, should receive credit for a conversion. Each model distributes the value of a conversion across each touchpoint differently.
Google Analytics has a tool that allows businesses to compare how different models distribute the value of a conversion. Here are some of the more common attribution models:
First Interaction
Last Interaction (most commonly used by businesses and individual marketing channels)
Last Non-Direct Click
Linear
Time-Decay
Position-Based
Data Driven (recommended model, but this requires enough data to work)
By comparing each attribution model, you can get a better idea of the ROI for each channel.
The most recommended model is Data Driven as this model relies on historical data to calculate the value of each channel. Unfortunately, this model is not available to all businesses as it requires a certain amount of data to work.
Let’s take a look at the different models to help you determine which is best for you.
First Interaction Model
What is it?
One-touch model
Gives 100% of the credit to the first action the customer took on their conversion journey
Ignores any following engagements the customer may have had with other marketing channels prior to converting.
Let’s look at an example.
You see an Facebook ad for Hootsuite and are immediately intrigued, so you click the ad.
You click around on through the site, but decide it needs more research and team discussion.
You sign up for their email newsletters and click on a paid search ad. Later, you click through an organic listing to look through Hootsuite’s available resources - blog post, webinars, etc and believe the tool is a good fit for you. You finally decide Hootsuite would be an excellent tool for your team and convert by going directly to the site.
So who gets the credit?
Facebook does.
Using a first touch model, Facebook would get the entire conversion credit, regardless of the other marketing channels touched.
This is a powerful attribution model when you’re trying to discover the methods and strategies that drive the most new customers to your business.
You can use this model to increase your ad spend and focus on these channels to generate new customers at a faster rate.
Last Interaction Model
What is it?
One-touch model
100%all the credit goes to the customer’s last touchpoint before converting
Ignores all previous engagements the customer may have had with other marketing channels prior to converting
Most common model used by businesses and Marketing Solution interfaces
So let’s say you’re scrolling through Facebook again and see Hootsuite’s Ad. You sign up for their email list, click through a paid search ad, then on an organic listing to the site’s resources, but finally convert by going directly to the site.
Who gets the credit?
Direct gets 100% of the credit.
This model is useful as it lets you know how users are finally converting. This model may skew more towards brand search or direct to site as it does not take into account the other touchpoints a user engaged with to get to know your brand.
Last Non-Direct Interaction Model
What is it?
One-touch model
100% of credit goes to the last channel clicked prior to Direct
Ignores all previous engagements the customer may have had with other marketing channels prior to converting and Direct last interactions
This attribution model takes into account that the last interaction model tends to skew more heavily towards direct, as direct traffic is often influenced by your other marketing campaigns.
So compared to the first two models, now who gets the credit?
Organic gets 100% of the credit.
If you are more interested in the effectiveness of your marketing strategy than in how many people know your business’ domain, measuring the last click non-direct can be more useful to you.
Linear
What is it?
Multi-touch model
Gives equal conversion credit to all channels in the conversion path
Linear ensures that all channels that the user engages with receive some sort of partial credit.
Who gets the credit?
They all do.
This model allows you to see each channel’s part in the conversion path and measure your marketing strategies holistically.
Time-Decay
What is it?
Multi-touch model
Gives the touchpoints that occured closer to the time of the conversion more credit than touchpoints further back in time. The closer in time to the conversions, the more credit a touchpoint receives.
Who gets the credit?
Technically, they all do but the Direct (the last touchpoint in our example) receives a majority of the credit.
This model is effective for determining which channels regularly drive conversions and which are primarily prospecting channels.
Position-Based
What is it?
Multi-touch model
The first and last click matter most and receive the majority of the credit. Everything in between only gets a little bit of the credit.
In Google Analytics, the first and last clicks are each given 40 percent of the credit and the other 20 percent is distributed equally across the middle interactions.
Who gets the credit?
Again, all channels technically get some credit. In our example Facebook Ads (the first touchpoint) and Direct (the last touchpoint) get the majority of the credit.
This model ensures that you are focusing on two very important touchpoints -- the first point that acquired the customer and the last point that converted the customer.
This model tells you which marketing channels are best for acquiring an audience and which are best for converting that audience.
Data Driven
What is it?
Multi-touch model
Distributes credit for the conversion based on past data for the conversion action. I
Uses account’s data to calculate the actual contribution of each interaction across the conversion path
Requires a certain amount of data to work
Based on an Algorithm, instead of assumptions.
While most other channels fall to extremes or assumptions, Data Driven relies on the algorithm and the weight you give a conversion to determine which channels are the most efficient.
So which model is right for you?
Go into Google Analytics and compare models. You can find the tool under "Conversions" > "Attribution" in your Google Analytics account.
Compare models and choose which one is right for your business. If you don’t see Data Driven as an option in your model comparison tool, it means that you do not have access yet to this model.
Not finding one that works for you? It is possible to custom models in GA that are based off of one of the core models (First Click, Last Click, Last Non-direct, Linear, Time Decay and Position Based).
For a client where we didn’t have access to Data Driven Attribution, I created models in GA based off of the Position Based model, but these models gave less credit to the last click if it was direct or a brand search click. The model allowed us to understand the conversion path without favoring direct or brand searches too heavily. You can follow this guide here to set up your own customer GA modeling.
Which model works best for you? Comment below with your thoughts!
Comments