Attribution is how Converge assigns a session or event to a specific touchpoint—like a channel, campaign, or referrer. That event could be a conversion event like Placed Order, or something earlier in the customer journey like Viewed Product. Converge processes attribution in two steps:
  1. Session attribution: assigning a session to a channel, campaign, adset or ad.
  2. Event attribution: assigning credit for an event to the different sessions a user had.
This gives you a clear view of which channels are driving customers and how they contribute to your business success.

​
Session attribution

Session attribution is how Converge determines where a visit on your site comes from based on the landing page URL, referrer and certain Click IDs or Tracking parameters.

​
What is a session?

Converge defines a session as a single visit to your website from a specific source and withing a certain time frame. We use the following rules to define when a session starts and ends:
  • A session has a time-out of 60 minutes since the last observed event. If a users visits your website, leaves the window open and continues more than 60 minutes later a new session will be started.
  • A visit from a different source than the current session will start a new session. For example: a user lands on your website from a Facebook ad, leaves and comes back through a Google Ads ad 30 minutes later.

​
Example

Let’s say two users search for your company on Google. One clicks on a paid ad from Google Ads, the other clicks an organic search result. They land on: Converge sees both users coming from www.google.com, but based on the gclid parameter that Google Ads attaches to the URl for each click, we know the first one came from a Google Ads campaign. The second visit has no gclid, so we attribute it to Google’s organic search. So we can attribute to the following Channels:
  • Paid click: Google
  • Organic click: Google (Organic Search)
To go any deeper—like understanding which campaign, ad set, or ad drove the click—we need tracking parameters.

​
Tracking Parameters

Tracking parameters are extra fields in the landing page URL that provide context about the session source. For example, someone clicks your search ad from Google Ads. The URL might look like: https://www.example.com/?gclid=123456789?cvg_source=google&cvg_cid=98765&cvg_adid=12345 Here’s what we can extract:
  • gclid: 123456789
  • cvg_source: google
  • cvg_cid: 98765
  • cvg_adid: 12345
If you’ve added Google Ads as a Marketing Source
, Converge fetches your campaigns, ad sets, and ads—including their unique IDs. We then matches those IDs with what we find in the landing page URL. So we can attribute:
  • Channel: Google
  • Campaign: Campaign A
  • Ad set: Ad set 1
  • Ad: Ad variant D
You can find more details on how Converge attributes sessions here.
Converge relies on correct tracking parameters to correctly attribute your sessions. Learn more about setting up the tracking parameters here.

​
Event attribution

Event attribution assigns credit for a specific event, like a Placed Order or Started Subscription, to the different sessions within a customer journey. Here’s an example journey:
  1. The user sees a Facebook ad but doesn’t click.
  2. The user clicks a Google Search ad and lands on your site, browses your products but doesn’t buy something.
  3. The user returns via organic search, adds a product to the cart, but doesn’t complete the checkout.
  4. The user gets a cart abandonment email, clicks it, and completes a purchase with a value of $ 1000.
The sum of all these interactions (our touchpoint) a user has with your brand, either on your website of anywhere else, is called a customer journey. Event attribution is the process applying credit to each of these touchpoints to determine how much they contributed to the user performing a specific action, like making a purchase.
Converge does not mark specific events as a Conversion. You can apply attribution to any event you track in Converge.

​
Converge vs In-Platform attribution

Converge will try to attribute your events to as many touchpoints as it can observe but this has its limits. As Converge can only track how a user interacts with your website we can’t track if a user only viewed an ad or received an email. Converge will only know if someone viewed the ad or received an email if the user clicked on a link and landed on your website. For the example above, Converge will only be able to observe the following three sessions the user has had on your website as we can not track that a user viewed an ad on Facebook as he did not visit the website by clicking a link in the ad. In many cases, your ad platforms will also try to attribute your events. We call this In-Platform Attribution and can possibly show some discrepancies when compared to Converge. This doesn’t mean that either Converge or the ad platform is wrong. The all just have a limited perspective of the truth. In our earlier example:
  • Converge attributes the purchase to any of the three sessions it observed according to the attribution model.
  • Facebook attributes the purchase to the ad view (even without a click). It will ignore all other touchpoints.
  • Google Ads attributes the purchase to the search ad click. It will also ignore all other touchpoints.
Each platform has limited visibility on how a user interacts with your brand and all have their own attribution model. This can cause the sum of in-platform attributed purchases to exceed your backend totals, since multiple platforms may each claim full credit for the same purchase.

​
Attribution models

Converge supports different attribution models to apply credit to the different sessions in a customer journey. See the following three attribution models where the value of $ 1000 from our example is attributed to the different session Converge could observe. Last Touch All credit goes to the last session Converge observed. First Touch All credit goes to the first session Converge observed. U-Shaped Credit is shared: 40% to the first and last sessions, 20% to the ones in between. See all attribution models Converge offers and in which use cases they could be applied.