Attribution Models: unmounting the Last Click

The last click is the most widespread model of attribution in the sector and at the same time one of the worst represents the real performance of our campaigns. This post is a reflection on this model in which the striker "pacifier" that pushes the ball to the bottom of the goal takes all the credit.

First of all, if you need to refresh concepts, take a look at this post of my colleague Veronica explaining what are and how the main works Attribution Models.

My brand campaigns work great!

Brand campaigns have exceptional performance compared to generic campaigns, they have a return on low investment. This is quite logical. But This performance gap is artificially enlarged when we use a last-click attribution model, passing the metrics to reflect a not-so-real reality as it should.

And if my brand campaigns work so well, so I was going to invest in generics, right? Try to pause all your generic campaigns to see what happens. Surely your brand campaigns will worsen their performance immediately.

How do you explain this?

It would be like playing a game without midfielders, the ball would get less to the striker, receive fewer passes, mark fewer goals. What is happening is that your brand campaigns are "stealing" conversions to the other campaigns. There are previous interactions to click on the brand campaign that the system is not taking into account when making the attribution of the sale to be using a model of last click.

Let's go on a trip...

Let stand a second and think about how the consumer of our productbehaves. Let's say summer is coming and I want to go on vacation to Majorca (what original). In this type of products whose purchase is very reflective and requires a long decision time, the "deception" is more clearly seen.

The decision process is much more complex than it seems

The decision process is much more complex than it seems

The complex decision-making process

Sometimes users directly perform brand searches and enter the Web through our ads. This can happen if the user is already clear that he will buy in your e-commerce. If not the case, which is the most common in a purchase of this type, the search will probably start with a generic concept like "hotels in Mallorca". You look for, you see that there is a lot of offer: checkers, travel agencies, the hotel chains themselves, offer aggregators, discount pages.

Everyone fights for taking the cat to the water. You don't want to pay more and you compare who doesn't match in these cases? Something more expensive, something cheaper, free cancellation, free access to the spa, room upgrade. The conditions and prices are different from one page to another; you should be well informed.

The documented user

And you compared, you compared... Compare! Without realizing it you have clicked on all the ads of the search and in much of the organic results. You've spent 2 hours browsing a lot of pages and you still don't have your clear decision. In a moment you made him earn a bit more to Google. I'll keep looking...

Search Results for "Hoteles Mallorca"

Search results for «Hoteles Mallorca»

The internet user today is intelligent, documented very well and their decision making process is generally long for this type of products. Since you start with the process of finding a vacation until you make the reservation may have been weeks.

Probably ends up in the brand

When you finally have it clear your last search is not generic. You don't look for "hotels in Mallorca" again. You know where the perfect offer for you is. You know you're going to go to that four-star Alcudia with that great offer you saw on the website of the Red Logo travel agency. You're looking directly for the brand name. It is easier to search the Web through the Google search engine itself than to type the full URL in the address bar. Surely you click on AD to access the web and make the reservation.

A road that began with a generic search "hotels in Mallorca" has ended in a purchase that has entered by a key word of brand; The name of the travel agency. According to the model of last click, which is the one that uses AdWords by default, the merit of this sale has been of the brand campaign. You forget who put the ball in the game.

The Last Click Lie

For me this model of attribution is misleading. It corrupts the data and does not reflect the actual performance of the campaigns. It would make sense to give most of the merit to the search that started the interest. The one through which your client, already real, found out about your offer. There are other attribution models for this. After all, if the ball is not put into play, there is no goal, no matter how good the striker is.

¿ What's the alternative?

There are many other attribution models. Personally, I prefer the attribution model known as "prefer first attribution" (a attribution model that neither Adwords nor analytics integrate). This model attributes most of the conversion to that search that made you start to exist for your client, and reduces the importance of progressively in the following impacts.

"Prefer first interaction" attribution Model

"Prefer first interaction" attribution model

In Digital Menta , we have tools that allow us to use advanced attribution models like this one. Thus, the process of optimizing the campaigns fits much more to the reality and you can focus your work on increasing the number of new customers while you keep the existing ones.

Written by


ANALYTICS · 27/09/2016

Did you like it?

Get a Quote

Joan Buigues

ANALYTICS · 17 / 06 / 2019

Dashboard for Facebook and Instagram: Do you know how to prepare yours?

If you work in digital marketing, you've probably ever needed to have a dashboard for your Facebook and Instagram profiles, with all the data well integrated and with the information you need to make the right decisions.

Joan Buigues

ANALYTICS · 12/09/2018

Google Sheets: 7 formulas to squeeze your data

Google Sheets or Google spreadsheets have become today an essential tool for many companies. And It's that they offer most of the features of traditional spreadsheets with the advantage of being in the cloud. In this way, we can work by teams making changes [...]


ANALYTICS · 22/03/2018

Custom Reports in Google Analytics

Personalized reports help us to concentrate and summarize the data, so that they help us to analyze the data that really interest us. Stop using the reports you find by default, and tell Google Analytics what metrics you need to work with at a glance.

Send this to a friend