The Attribution tab lets you compare your different attribution models, the active model being the one with a green dot next to its name. You can view sub-levers and details by clicking on the zoom icon next to a lever's name. Active filters are also applied to this tab, and to all the Marketing Mix tabs.
When you enter a lever associated with Google Ads, Meta or Snapchat, a new conversions table will be displayed under the attribution models table. These are the conversions and revenue attributed to each advertising network, according to the attribution window active in the network (Google Ads and Snapchat) or all those available (Meta). These numbers should be compared with participation in ThankYou Analytics, which counts the number of conversion paths in which the lever has participated.
SMART (Social Media Ads Recovered Touchpoints) technology was created to reduce the gaps between social media platforms and our analytics solution. In an analytics solution, Meta is only present when it is at the end of the path, and alone.
This is because when a web user clicks on an ad on Meta, the advertiser's site opens on Meta's internal browser, on which a cookie is stored. When the user opens the same site on their external browser, or another device, a new cookie is stored, and two separate sessions are calculated from an analytics solution point of view. Only Meta knows how to match paths between different browsers and devices, because it uses client information parameters to optimise the quality of event matching. This can be a combination of different types of data:
Last name
First name
Telephone number
Email address
IP address
Source : Meta documentation
Our solution for reducing the gaps between the Meta vision and ThankYou Analytics' first-party data is therefore to combine these two pieces of data to find cross-browser, cross-device conversions, while including post-click and post-view attribution. Using an API, we retrieve Meta impressions, clicks, costs and conversions so that they can be automatically synchronised with our first-party data in real time. There are a few things you need to do to ensure that your data is properly synchronised:
Synchronise the Meta connector
Have the same conversion goals between ThankYou Analytics and Meta
Add an amount to conversions
If the uplift of the social ads lever is low or if there is no amount on conversions, it will be necessary to add a random amount to at least two decimals and to create a new dimension for this new amount.
The SMART vision can be added to a linear model, and can be based on the following attribution windows (possibility of associating post-view and post-click):
Post-view 1 day
Post-view 7 days
Post-view 28 days
Post-click 1 day
Post-click 7 days
Post-click 28 days
Why do I have discrepancies between the clicks from the advertising networks and the entries in ThankYou Analytics?
The entries correspond to the first pages viewed by users when they arrive on your site, and should correspond to the outgoing clicks from the advertising networks. However, there may sometimes be discrepancies, for a number of reasons:
Synchronisation between the ad network campaigns and the traffic is not complete (absence of utm in the ad network).
The clicks observed in the advertising network are not outbound clicks (link click, interaction with a post, etc.)
Clicks are deduplicated by the advertising network (double click by the same person, bot traffic, etc.)
Internet users closed the page too quickly and the entry did not have time to be calculated
What's the difference between a mono-touch and multi-touch model?
A mono-touch model allocates 100% of the conversion to a single lever. A multi-touch model will divide the conversion between all the levers, and therefore take into account all the touchpoints.
How can I find out how the attribution models are set up?
To find out how the attribution models are set up, simply go to Settings > Attribution > Model configuration.
How do I change the active allocation model?
To change the active attribution model, go to Settings > Attribution > Default attribution model.
This table shows you the participation of each lever, as well as their position in the conversion paths.
A participation corresponds to the number of conversion paths in which the lever has participated. This corresponds to the sum of deterministic (with consent) and probabilistic (without consent) participation.
The percentage represented in the "Alone" column indicates the number of times the lever was the only touchpoint in a conversion path. "Beginning of journey" corresponds to the number of times the lever was the first touchpoint in a conversion journey involving several touchpoints. The "Middle" column indicates the number of times the lever was present between other touchpoints in a conversion journey, and "End" indicates the number of times the lever was the last touchpoint in the journey.
You can see the sub-levers and details by clicking on the zoom icon next to the name of a lever. Active filters are also applied to this tab, and to all the Marketing Mix tabs.
How is lever participation calculated?
Participation does not depend on the active attribution model. If a lever is present in the conversion path, it will be counted once. If two identical URLs are present several times in the same path, it will only be counted once.
This tab allows you to analyse the interactions between your levers, sub-levers and details, according to deterministic participation (with consent). Clicking on the zoom icon next to a lever will take you to its sub-levers and details.
You can display the data in percentage or volume, by clicking on the three dots at the top right of the table.
To go further in the analysis and compare more precisely the interactions between two levers, you can click directly on the box in the table that cross-references these two levers, or click on "Advanced analysis" and enter the two levers to be compared. You'll find a representation of the interactions between your two levers, as well as the place of the selected lever in the conversion paths, in relation to the lever being compared.
You can also do this analysis by selecting several levers, sub-levers and details. If several elements are selected, the levers will not be added together, but compared with the one opposite.
Note: the lever on the left is not necessarily represented in the bubble on the left. The bubble on the left will be the one with the most interactions.
How do I interpret the data in the interaction table?
The table is read in rows, starting with the levers in the first column, and displays the percentage of interaction of the chosen lever with each of the other levers. For each box, hover over the elements to see how they should be read.
This tab contains an analysis of the repetitions of your levers, sub-levers and details. Deterministic participations represent the number of conversion paths with consent in which the lever has participated, while touch points with conversions represent the total number of touch points in the conversion paths: a lever can therefore have several touch points in a path. The average number of repetitions shows you the average number of touches present in a conversion path, and the repetition rate shows the percentage for which a lever has had more than one touch in a conversion path.
If you wish to display your sub-levers, simply click on the zoom icon next to your lever. You'll also have access to the details.
This tab allows you to analyse your customer journey as a whole, or according to your levers, sub-levers and details, using the filters. All filters can also be applied to this tab.
You can display the data in cumulative or non-cumulative form, as a percentage or by volume.
How do I interpret the data in the "Time lag" table?
These are conversion times. The second line reads as follows: "X% of conversions were completed in less than 10 minutes" if the display is cumulative, or "X% of conversions were completed between 5 and 10 minutes" if the display is not cumulative.
How should the data in the "Path lenght" table be interpreted?
This is the number of touchpoints present in the conversion path. The second line reads as follows: "X% of conversions had 2 touchpoints or less in the path" if the display is cumulative, or "X% of conversions had strictly 2 touchpoints in the path" if the display is not cumulative.
