Skip to main content

Ad spend interpolation

What is ad spend interpolation?​

In some scenarios, the tracking script is not loaded by the visitor's browser.

That is the case, for example:

  • if you add the tracking script through a tag manager (GTM, Tag Commander in CDN mode) and the user's browser or a browser extension blocks such scripts.
  • if you operate in a country whose regulation is less permissive than the French CNIL when it comes to collecting browsing information without consent.

When the tracking script is not loaded, it is impossible for Webmarketer to attach a session and to link a session to an ad node.

This can skew the performance figures displayed by:

  • reducing the number of events attributed to an ad node
  • reducing the number of sessions attributed to an ad node
  • inflating your CPA, or any other data computed from ad spend

To correct that bias, ad spend interpolation lets you compute the gap between theoretical sessions and the sessions actually measured.

A concrete example​

A Meta Ads campaign is clicked 128 times and has spent €64. Webmarketer, however, only measured 112 sessions through the tracking script.

This means 16 sessions happened in a browser, or with an extension, blocking tracking.

Since the events produced by visitors during those 16 sessions carry no tracking identifier, Webmarketer classifies them under the "Non-attributable" category. Those events are therefore not attributed to the ad nodes they originated from, which can skew the CPA or any other indicator based on ad spend.

To address this, Webmarketer provides an Interpolated ad spend column for each node.

In our example, that column shows €56 (87.5% of €64) for our Meta Ads campaign, so the data computed from that column accounts for the exclusion of non-trackable traffic.

The remaining €8 (€64 - €56) is counted under the "Non-attributable" category, and can therefore be compared with the events generated by those non-trackable sessions.

The "Non-attributable" section​

The "Non-attributable" category gathers all the ad spend that led to non-trackable sessions and all the events produced during those sessions.

While it does not let you break down the performance of the campaigns that acquired that traffic, it does let you keep an eye on the overall KPIs for all of that non-trackable traffic.

How Webmarketer counts​

Most ad platforms report the number of clicks on an ad:

  • Google Ads
  • Microsoft Advertising
  • Meta Ads
  • LinkedIn Ads

This lets Webmarketer know the theoretical number of expected sessions, and therefore compute the interpolated ad spend for each ad node of those integrations.

Unfortunately, some ad platforms do not report the number of clicks on an ad link:

  • Snapchat, for instance, only returns the number of swipes, which does not match the number of outbound clicks to your website

Another possible reason for the theoretical session count being missing: you set up a custom integration, paid monthly, and you do not know in advance how many clicks that traffic source will deliver.

To cover that need, Webmarketer provides proxy URLs that measure those clicks for you.

Those tracking links look like this: https://{tracking-subdomain}.{your-domain}.tld/p/?projectid={project-id}&wmktid={YdScfsiViXKgmsCY}&targetUrl={destination-url}&cid={campaign-id}&[...]

When a visitor opens that URL, they are sent to the page given in the targetUrl parameter through a 302 redirect.

Webmarketer then counts one theoretical click, and can compare the number of theoretical clicks with the number of measured sessions.

Note: you do not need to add the wmktid, cid and other tracking parameters to the targetUrl. They are copied from the proxy URL into the destination URL.