A complete guide to understanding attribution in Google Ads

Gotraffic author

Matea Viher

12 October 2021 | Google Ads

A complete guide to understanding attribution in Google Ads

Attribution models are designed to help you understand the path a user is taking before making a conversion. How, when, and why users converted through your PPC ads.

On their way to performing the desired action, users can run multiple searches and thus interact with more than just one ad. With attribution models, Google Ads gives you the ability to assign a value to each of these interactions.

Zašto su modeli atribucije toliko bitni?

Because they take you away from the one-dimensional thinking that only a certain keyword and a certain ad are responsible for the conversion. Also, they give you an insight into what else played a role in the user’s path to conversion.

Available attribution models and when they can be useful

Google Ads currently offers 6 attribution models.

  1. Last click

The last click is the default attribution model in the Google Ads system, and the credit for the conversion is entirely attributed to the last-clicked ad and the corresponding keyword.

This is the simplest attribution model. It means that even if you show 10 different ads to one user, only the last one gets credit for the conversion.

Last click attribution model gives you the information you want, but it doesn’t tell the whole story because the user path isn’t always that simple.

Rarely will a user convert right after the first time seeing your ad. It is more likely that they will first find you accidentally on a search engine, then check on social networks and only after that conduct specific search to perform the conversion.

The only case where this attribution model can be useful is when the user’s decision-making process is fast and impulsive (for example, plumber services, ordering pizza, etc.)

Advantages: A simple and direct approach that provides insight into the resource that is the last driver of a conversion.

Disadvantages: Does not show the full picture of the user’s path to conversion, which can result in pausing individual campaigns just because they do not get credit at first glance.

  1. First click

The first click attribution works on the same principle as the last click, except in this case the credit for the conversion is entirely attributed to the first click – the first ad that the user clicked on and the corresponding keyword.

How does it work?

Let’s say you have a standard search campaign. The user sees your ad and clicks on it, but does not choose to purchase. After that, he or she sees your remarketing image ad, clicks on it again but still doesn’t place the order. Finally, the user enters the brand keywords and performs the conversion.

But all the credit for the conversion goes to that first ad user clicked and keyword on the search network.

Two problems are immediately clear with this attribution model:

  • the ad with the first click caught user’s attention, but did not achieve the desired action
  • if more than one keyword was included in the conversion process, it means that the user did not yet have a clear intention when searching for the first one, to which the credit is attributed.

On the other hand, this attribution model can be very good for brand awareness campaigns and attracting customers in the research phase.

Why? Because it shows you that the first keyword or ad was compelling enough to drive the click. It shows you that you have targeted a good audience, you have a good call to action, and that you have chosen good keywords for your goal.

Advantages: Lets you see how users discover your brand.

Disadvantages: Does not show the full picture of the user’s path to conversion, which can result in pausing individual campaigns just because they do not get credit at first glance.

  1. Linear model

The linear attribution model evenly distributes the credit to each interaction the user had prior to conversion.

For example, if you had a complex strategy that involved different types of campaigns (one could only contain a brand keywords, the other general ones), Google Ads would give equal credit to each of the campaigns.

This model can give you a good idea of what forms of advertising are paying off for you. Why? Because you’ll clearly see which of the selected channels don’t generate clicks and aren’t crucial for conversions. 

This attribution model is very useful if you have a clear plan for attracting customers and guiding them through the buying funnel.

Advantages: Provides a more complete picture, which allows you to analyze all interactions on the conversion path.

Disadvantages: Since all interactions are given equal credit, it can distort some data and give individual clicks too much or too little credit, making optimization more difficult.

  1. Time Decay

With this Google Ads model, more credit is given to clicks that occurred closer to the conversion.

For example, if a user had 7 different interactions within a month before the conversion itself, the last two interactions would get the most credit.

This attribution model can be useful if you have long and complex campaigns and the user needs more time to decide to purchase a product (for example, cars or work machines).

On the other hand, for industries where the conversion is quite fast, such as e-commerce, this attribution model is not usable.

Advantages: Gives a good insight into the whole process of selling a product or service.

Disadvantages: This model gives merit based on time and can distort the data. That very first click can be in fact crucial for the user.

  1. Position based

In this model 40% of the credit goes to the first and last ad interactions and the corresponding keywords while the remaining 20% ​​is allocated to other interactions with the ads along the conversion path.

With this attribution model, you can get a slightly better insight because it gives you information about:

  • which keyword or ad first caught the user’s attention
  • which keyword and ad the user clicked on before the conversion
  • which clicks between the first and last one supported the user’s decisions.

By using this, you give the most attention to the first and last clicks that are responsible for the initial interest and the conversion itself, while reducing the importance of consecutive interactions in the research phase.

How can this model be useful? To help you find the best combination of initial and final interaction that will surely lead to a conversion.

Advantages: Recognizes that the most important clicks are the first and last, but at the same time validates clicks in between.

Disadvantages: May underestimate the interactions that take place between the first and last click.

  1. Data driven model

As its name suggests, a data driven model is used based on previous data for a particular conversion action. It differs from the above models because it uses the actual data from your account to calculate the contribution of each interaction on the conversion path.

Google uses its algorithms to study all user interactions with your account and to assess which of them have the greatest impact on conversion.

However, in order to use this attribution model, you will need to meet the requirements set by Google because this model requires a certain amount of account data:

  • at least 3000 ad interactions in supported networks within 30 days
  • the conversion action must have at least 300 conversions in 30 days

Once the conditions for using the models are met, you will need to maintain them from month to month:

  • at least 2000 ad interactions in supported networks each month
  • the conversion action must have at least 200 conversions each month.

Advantages: The most accurate attribution model because it takes actual data into account.

Disadvantages: Strict requirements for its use.

How to set up an attribution model?

The attribution model is set for each conversion separately. Go to Tools & Settings – Measurements – Conversions.

You can choose to create a new conversion where you will select the desired attribution model among the other data required to create the conversion.

If you already have set conversions and want to change their attribution model, select the conversion and click on it. In the new window, select Edit Settings.

Click on the last item called Attribution Model and select the new one.

If you are unsure of the existing attribution model and would like to see what you can get by choosing another model Google Ads gives you the opportunity to do that as well. Go to Tools & Settings – Measurements – Attribution.

In the menu on the left, select Model comparison. At the top of the report you can select:

  • dimension (campaign, ad group, keyword)
  • conversion action (all actions or a specific one)
  • lookback window
  • two attribution models for comparison.

Now you can see how changing the attribution model would affect conversions and the percentage of that changes.


Attribution models may seem quite complicated at first, but we hope we have helped clarify them with this article. They are intended to clarify how and why a user made a conversion.

Be careful when choosing an attribution model because it can have a big impact on your data and future campaigns. A bad model for your business also means bad decisions in the future. On the other hand, with the right model, you get a powerful tool to further develop digital marketing.

Finally, we will once again repeat the brief guidelines for selecting an attribution model:

  • last click for bargain products where the purchase decision is made quickly
  • first click for campaigns aimed at raising brand awareness
  • a linear model for well-planned campaigns
  • time decay for long and complex campaigns where the user needs a long time to decide to buy
  • a position based model for finding the best combination of initial and final interactions
  • a data driven model is always the best choice if you have enough data.

Share this article

Latest from our blog

Want to get more leads and sales?

Tell us more about your business and let's see how we can help you achieve your goals.

Get free proposal