When we have an online store we very often launch advertising campaigns on different platforms. Our thinking is that the more the better, the more social platforms or possible locations where we can appear the better. And we are really not far from this being the correct way, but what happens when it comes to measuring these campaigns or measuring which of them is providing the best results , not only in terms of attracting traffic, but also in terms of the participation that have had in the final conversion or sale .. How can we know for the future in which campaign or in which traffic channel to invest more or less.

For this problem , the solution is found in attribution models . But this issue of attributions is more complex than it seems and that is why the vast majority of platforms specialized in e-commerce do not have it incorporated and you have to turn to specialists for it. But is it worth investing in applying attribution models to our online store?

The answer is yes. Traffic attribution is much more complex than simply looking in Google Analytics at how many users have entered through a social, paid search campaign...

When the client is impacted with a paid campaign and they access our website through this link, in principle we should have that link marked to know where that user enters. We can do this link marking through UTM . With this we will see the source of traffic through which a potential client enters . However, this will only give us a view of the volume of visits, not the performance with respect to our final objective, which is that the user ends up purchasing.

This is because a user or visitor does not have to buy when entering the website . That impression has served to generate a visit, but not a conversion.

In order to measure the degree of involvement that a traffic source has had in the final purchase , attribution models are used .

What is an attribution model?

Let's start by knowing what attributions are. Attributions are known as a set of rules that distribute the value of a conversion between the different parties or contact points that were involved in the path that the user followed.

But attribution isn't just that, it's also about understanding the context behind the data, it's about understanding the consumer and understanding the full journey and the full value of how the consumer engages with the brand.

One of the keys when working on attribution models in an online store is that everyone uses the same formula. The approach should be the same as that used by the rest of the team, as it will be the way we understand the impact on consumers.

To know if our e-commerce needs the attribution application, we must first consider what data we want to know with the attributions.

And many brands do not understand how their consumers interact with them , or what the consumer thinks when they begin the interaction with that brand. Therefore, the attributions will give them a vision of how the user moves within our website and their behavior.

The average number of visits for many e-commerce businesses is four visits before a person converts. So looking only at the last access point is not enough , since we will not know the impact that each of those previous visits to the website has generated on the user.

Another point to take into account when we talk about attributions is the attribution window. The attribution window is the time frame in which the attribution will be applied . If more sources are involved in a conversion than appear in that time window, attribution will not consider them. It will only assign a value to the sources that intervene within the window.

For all this, it will be very important to understand what models are used within the process.

Types of attribution models

Last interaction , the last point of contact would receive 100% of the sale value. PPC works very well in last-click conversion because that type of activity will drive the user who has a very high awareness of the brand and will be ready to convert at that moment.

Last indirect click , all direct traffic is ignored, and all credit for the sale goes to the last channel the customer clicked on before the conversion. It is the model that Google Analytics uses as the basis of its conversions.

With last interaction or non-direct last click models, you would be missing the opportunity to understand what activities take that user from awareness to behavior to conversion. We need to have that level of understanding and this is one of the reasons attribution exists in the first place. That is why the last interaction and the last non-direct interaction are not so good for understanding consumer behavior.

First interaction , in this case the first point of contact will receive 100% of the conversion credit. This model will be useful for users who did not know the brand and who, through an online advertising campaign, have begun their first interactions with the business.

Linear Attribution, each touchpoint in the conversion path will share the same value for the conversion. One of the inefficiencies of this model is that it does everything equally. It doesn't show which channels really impacted the user the most. Whether I enter once through Facebook ads and then through Google ads where I finally purchased, they will all be given the same credit. It would only be useful if the campaigns are designed to maintain constant contact with the user.

Time decay , touchpoints closest in time to the time of sale or conversion get maximum credit. The last interaction will be the one that takes the most value in this model, but what we will also look at is how far back in the user's journey the previous interaction occurred.

Base position . Google Analytics popularized it as the most accurate. Credit for the conversion will be given to the first point of contact and the last point of contact. The former is considered to be where the user discovers the brand and the latter is where the user is ready to convert.

One of the key things that attribution can help us understand is return on investment or return on ad spend. Therefore, attribution will give us an analysis of which point of contact with the user is most important, which will allow us to make decisions about optimization or marketing investment based on the data obtained.

What is attribution for?

As we have seen so far, attribution is the way in which the value of a conversion can be distributed between the different points of contact that an e-commerce has had with the customer in the purchasing process. This way we can know the relevance of the source of the campaigns and how they directly affect them.

For an e-commerce, having a single attribution model is important since it allows us to combine data from different platforms and not duplicate metrics such as sessions, sales or income from our campaigns .

For all these reasons, this branch of digital marketing is one of the most complex and, as we mentioned previously, it is often necessary to resort to the help of a professional. Defining which model will be the most suitable for our business is going to be one of the keys that will allow us to guide our marketing and focus it on the specific objectives that we want to achieve in the online store.

Because there is no specific business model for one type of model, it is necessary to analyze the interactions and behavior of our audience to establish objectives. With this analysis we can see how we want the traffic sources to interact with the client and thus establish one of the different models.

In web analytics we can obtain real results of how the campaigns are working not only for the acquisition of traffic, which is also important. but by the degree of participation they had in the user's purchase decision.

Remember that implementing an attribution model to your marketing strategies can help you improve the performance of the online store , achieve a greater volume of transactions, since we will optimize the campaigns according to the data obtained, and thus improve the total conversion rate of your e- commerce.

It may interest you:

• CRO methodology and conversion optimization for your online store.
• What useful metrics does Google Analytics give us by default to start optimizing our website?
• 13 tips to optimize the speed of your online store and reduce its bounce rate.