Choosing the right attribution model for your PPC ad is extremely important for tracking goals and conversions. However lots of people aren’t even aware of what these models are, how they can be useful and when they should be using them. Here I will be teaching you everything you need to know about attribution models.
What are attribution models?
An attribution model is the rule, or set of rules, that determines how credit for sales and conversions is assigned to touch points in conversion paths. Without attribution models you won’ t be able to accurately give credit to the correct marketing channels. All marketing channels should be assigned an attribution model, but I will namely be focusing on Google Ads.
When you set up a conversion goal Google Ads will automatically set your attribution model to the default, which is last click. This setting isn’t particularly visible so changing the default setting isn’t something that people think of.
There are 7 common attribution models that we will be looking into.
- Last non-direct click
- Last Google Ads click
In this attribution model, the marketing channel that was the first touch point will get 100% of the credit for the conversion.
For example, imagine you are searching for a new dining table so you type ‘shabby chic dining tables’ into Google and you see an image that you like at the top of the page and click through to the website via a Google shopping ad.
But of course, this is an expensive buy and you want to shop around a bit before you make a solid decision, so you go away and do some research into some more dining tables. After a while you decide that actually, the first table you saw is the one you want. So you go back to the website but this time you type the websites domain directly into your browser and make the purchase.
With first click attribution 100% of the credit will go to the Google Ads shopping because this is the first marketing channel you came through.
First click attribution model is perfect for those wanting to identify what methods and strategies drive new customers.
This is Googles default setting and it means that when you get a conversion 100% of the credit will go to the very last marketing channel the prospect came through.
For example, imagine you are searching for some new running trainers, so you make a search on Google. You click through to the first website that is under the paid ads (so the top organic result). After browsing a while you decide that you can probably wait until pay day to make a purchase.
However, a few days later you’re at home scrolling through Facebook and you see an ad advertising the pair of trainers you liked with a 10% discount, so you click through to the same website via Facebook and purchase them whilst the offer lasts.
With last-click attribution model 100% of the credit will go to paid social (Facebook) because this is the very last marketing channel the prospect came through before they made the purchase.
Last-click is great for those who want to find out what channel is providing most of their conversions. In other words, last-click allows you to see where your bottom-of-line funnel transitions are happening.
Last Non-Direct Click Attribution
Last non-direct click attribution model is pretty self explanatory. This model ignores all direct traffic and will give 100% of the credit to the last marketing channel the prospect came through that wasn’t direct.
For example let’s go back to the dining table. You first came through on Google Shopping. Then let’s say you then see an ad on Facebook advertising the same dining table you were looking at the previous day so you click through to the website again. You take a browse but decide you better discuss the purchase with your partner/housemate later on when they are home.
You decide you want the table so go back to the website directly (by this point you know their domain pretty well) and make the purchase.
100% of credit will go to Facebook ads. Why? Because this was the last non-direct marketing channel you came through.
Last non-direct click attribution is great for those who want to measure your marketing success but don’t want to account for direct traffic. You may want to do this often, as direct conversions are simply the result of people who have already seen a number of your marketing campaigns.
Linear Attribution Model
Linear attribution model is when each marketing channel that was involved in the conversion path gets an equal amount of credit.
For example, if you type a search query into Google for new shoes and click on the first paid ad but decide not to buy and a day later click on an ad through Facebook and then later on that night type the URL into your browser to actually make the purchase, credit will be split equally between the three.
So in this case Google PPC will receive 33.33%, Facebook advertising will receive 33.33% and direct will receive 33.33% of the credit.
Linear attribution model is great if you want to measure your marketing channels holistically.
This is where attribution modelling can get complicated. Position-based attribution will assign 40% of the credit to the first-click and another 40% to the last-click. The remaining 20% will be distributed evenly between the interactions in between.
Let’s say you make a search on Google and organically land on a company’s blog post and subscribe to their newsletter. Then a few days later you see an Instagram ad that encourages you to click through to the website again but you still don’t purchase.
A few days later, browsing the web you notice a display ad for the same company, you click through and finally decide that now is the time to make a purchase.
In this instance, 40% of the credit would go to organic search and another 40% would go to display ads because this is the very first and last interaction the prospect had before purchasing. The remaining 20% would be split evenly between email and Instagram as these were the interactions in between.
Position-based attribution is great if you want to know which marketing channel works best for acquiring new audiences and which is best at converting audiences.
Time-decay attribution is when the interactions closest in time to the sale or conversion gets most of the credit. Essentially, the last interaction gets most the credit and the first gets the least.
For example, if a customer comes through display, then email, Facebook Ads, direct and finally organic the credits would be distributed something similar to:
- Display – 5%
- Email – 10%
- Facebook Ads – 15%
- Direct – 30%
- Organic – 40%
This attribution model is effective for determining which channels regularly drive conversions and which are primarily top-of-funnel channels.
Last Google Ads Click Attribution
Last Google Ads click, again, is pretty simple. 100% of the credit will go to the very last interaction that came from Google Ads.
So let’s say you search on Google and click on a paid ad. Over the next couple days you then return to the website via other ads and Twitter but still don’t make a purchase. The next day you decide that you want to take another look so you visit the website again through another paid ad, you still don’t buy anything, but later on that night you go to the website directly to make the purchase.
100% of the credit will go to the SECOND time you clicked on a paid ad, right before you visited directly.
Last Google Ads click attribution is perfect for those who want to determine which Google Ads keywords are driving most revenue for your business.
So there you have it, the 7 best attribution models and when you should be using them.
If you’d like to know more, contact us!