[kkstarratings]
All of us who are familiar with it know that the Ads interface provides us with information on many different KPIsnot only through the dashboard, but also through its default reports. These can be found in the reportsBut in this post we are going to focus on other more hidden documents. In case you have ever wondered what you can find in that mysterious section called search attributionIn this section we are going to present a small guide about the different reports offered and the data that can be extracted from them.
Attribution reports
Let's start at the beginning. Attribution reports show information about the routes followed by users in their interactions until they generate the conversion action. It is important to consider two factors:
- These reports are not available for App download conversions, physical store visits or goals imported from Analytics.
- If the conversion action to be analyzed is configured at the MCC level, the data in these reports will only be available at this level, not for individual accounts.
Now it is, first thing what we are going to find is a general view on some of the reports that we will see later on in the different sections of this report. search attributionsuch as path length, time lapse or multi-device conversions.
In addition, during this process we can modify different variables such as the conversion action we are interested in at any given moment, the conversion window, the analysis period or whether we prefer to focus on clicks or impressions.
Conversions
Here we can distinguish between main and assisted conversions. The main ones break down the conversion actions configured in the account or in the MCC, with their corresponding conversion count and value.
On the other hand, assisted conversions show the conversions that our ads have contributed to. Within this report, we can see the last click conversions (those that occurred immediately after the ad click) and the click/impression assisted conversions (those that occurred after clicks/impressions for each keyword, ad group, campaign... not counting the last click). After these metrics, a relationship between the two is displayed KPIswhich can be considered a ratio that classifies the keyword, ad group or campaign as an attendant or converter.
The data in these reports are of interest for identify those keywords that generate the highest number of conversions.
Multi-device activity
These reports show information about the ads that a user interacts with before making a conversion, but with the particularity of being able to view through which devices this interaction has taken place.
In this way, we can detect certain patterns that give importance to one device or another, obtaining information that allows us to optimize campaign management.
There are three types of reports:
- Devices: here we will see the conversions made in each device category, differentiating between the one that generated the interaction and the one that finally closed the conversion.
- Assisted devices: similar to the assisted conversions report, but this time by device. The metrics shown are last click conversions, assisted conversions and a relationship between the two that we can define as the frequency with which assisted conversions from one device, have been closed on another. This is easier to understand with an example:
In the image above, the ratio of assisted conversions on mobile is 0.90, which means that for every conversion closed on mobile, 0.90 conversions closed on other devices were assisted by a mobile device. From this we can deduce that mobile is more of a converter than an assistant.
- Device paths: presents a breakdown of the user's path through the different devices they interact with until the conversion is made.
Routes
The routes section allows us to see the paths that users have taken to complete conversions. Here we will be able to perform three types of path analysis:
- Main routes: shows the most frequent routes when completing a conversion. This report allows us to choose any of the structural components and the different path lengths, depending on the analysis we are looking for or the adjustments we intend to make.
- Time lag: provides information on the time it takes for users to convert from the first impression of the ads, or from the first click. This will help us to understand how long the reflection period prior to purchase lasts, both in days and hours.
- Path length: this analysis is simple, as it allows us to visualize how many clicks are necessary to close the conversion. It is important to note that we will only see the path through the Ads ads, not through all channels (for this we will go to Analytics).
Click analysis
The reports in this section allow us to know through which campaigns, ad groups or keywords users have entered our website (first click analysis) or have performed a conversion (last click analysis):
- First click analysis. We will be able to see what is generating traffic to our website for the first time, as well as the conversions and their value with a first click attribution model.
- Last click analysis. The real difference with the previous section is that, by analyzing the last click made before the purchase is generated, we are talking about terms or campaigns that will be more transactional.
Attribution models
As we already know, an attribution model is a set of rules through which the value of a conversion is distributed among the different channels with which a user interacts before making that conversion.
This report allows compare data between different attribution models to see the variation in moving from one to the other. This can be very useful for us to substantiate the propositions of changing the customer attribution model. It is also useful to see if the variation in information is significant when the conversion window is changed.
And so we would have finished the route through the information that we can find within the search attribution reports.
Do you use them often and do you think they offer interesting information? Let us know your experience in the comments.
