Safari ITP (Intelligent Tracking Prevention) is a tracking-prevention system Apple built directly into the Safari browser (iOS and macOS). It restricts cookie lifespan and third-party cookie access to make it harder for sites to track users across other sites.
This restriction is designed for user privacy, but it also affects the conversion counts shown in ad dashboards. Tools like Meta Pixel, Google Ads conversion tags, or GA4 can miss some events in Safari traffic. The issue isn’t that the campaign stopped working — it’s that the measurement appears incomplete.
WebKit’s tracking prevention documentation explains that ITP blocks third-party cookies without exception, and that cookies created via JavaScript are deleted within 7 days without user interaction. WebKit’s Private Click Measurement announcement describes the on-device alternative measurement model Apple proposes instead of cookie-based tracking. Meta’s Conversions API documentation and Google Ads’ Enhanced Conversions documentation show that it’s possible to measure using server-side data or encrypted first-party data, without depending on browser cookies.
What Is Safari ITP?
ITP is a browser privacy layer designed to prevent sites from combining a user’s behavior across other sites. It has existed in Safari since 2017, and its scope has expanded with every release.
ITP is a browser setting, not an add-on
Users don’t turn ITP on separately; it’s enabled by default in Safari. That means a site can’t “turn off” ITP — it can only adapt its measurement approach around it.
The goal is privacy, not opposing advertising
ITP’s purpose isn’t to block ads — it’s to make it harder for a user’s behavior on one site to be carried over to other sites without their permission. Ad measurement becomes harder as a side effect of this restriction; the real target isn’t advertising, it’s cross-site identity matching. Overlooking this distinction can lead to chasing the wrong solution.
How Does ITP Limit Cookies?
ITP’s restriction isn’t a single rule — several mechanisms work together.
Third-party cookies are blocked entirely
Safari grants no access, without exception, to cookies belonging to another domain. This largely disables the classic tracking method that relies on identifying a user with the same identity across multiple sites.
First-party cookies also have a lifespan
Cookies and storage created via JavaScript are deleted within days without user interaction. Scrolling doesn’t count as interaction; a click, tap, or keyboard input is required.
Link decoration is detected
Parameters added to a URL, like a click ID, are recognized by ITP. The validity period of cookies created through this method is additionally shortened.
How Does It Affect Meta Pixel and Google Ads Measurement?
ITP doesn’t stop the campaign, but it can create a gap between the dashboard and the actual result.
Pixel-based events can be lost
Since Meta Pixel runs in the browser, event matching can weaken in the parts that depend on the cookies ITP restricts. This loss can grow especially when several days pass between the first ad contact and the conversion.
Google Ads and GA4 carry the same risk
As long as conversion tracking works on a cookie basis, a similar risk of under-reporting exists in Safari traffic. This doesn’t mean the campaign is performing poorly — it means measurement can’t keep track of the user.
The gap varies by device and browser
The size of the impact depends on how much of the traffic comes through Safari/iOS. This gap can be more pronounced for campaigns targeting an iOS-heavy audience.
For example, if a campaign generates 80 requests in the CRM within a week but the dashboard shows only 55 conversions, and the missing 25 requests largely come from Safari traffic, the problem isn’t with the campaign — it’s in the measurement chain.
Ways to Reduce the Impact
The goal isn’t to bypass ITP — it’s to make measurement less dependent on the cookie chain.
Server-side measurement
Methods like Conversions API on Meta’s side and Enhanced Conversions on Google’s side deliver data from the server or through encrypted first-party information instead of the browser cookie. These methods don’t replace Pixel on their own — they’re meant to work together with infrastructure like server-side tagging.
Matching with first-party data
Methods that match encrypted customer data such as email or phone number are less affected by cookie loss. This approach is especially valuable for form and lead-focused campaigns; for more detail, the article What Is a First-Party Data Strategy? can help.
Apple’s own measurement model
Private Click Measurement is a model that processes, on-device and without carrying user identity, “which ad click led to which general outcome” and reports it with a delay. It may not overlap one-to-one with dashboard data, but it provides an additional reference point.
Pre-Launch Checklist
A gap caused by ITP shouldn’t be confused with an actual measurement error. The following checks help clarify the distinction:
- Is the gap between dashboard conversions and CRM/sales records being examined broken down by browser or device?
- Is server-side measurement (Conversions API, enhanced conversions) set up, or is the browser cookie the only data source?
- Is the gap concentrated only in Safari/iOS traffic, or is it seen across all traffic?
- Is the gap larger in campaigns where more time passes between the ad and the conversion?
If the gap is concentrated only in Safari, the problem should be looked for in the measurement chain, not in the campaign.
A Common Mistake
The most common mistake is interpreting a lower-than-expected conversion count in Safari traffic as “this audience doesn’t buy.” The actual result may have happened — it may simply not be fully reflected in the dashboard.
Summary
Safari ITP restricts cookie lifespan and third-party access to protect user privacy. This restriction can cause cookie-based measurement tools like Meta Pixel and Google Ads to miss some conversions in Safari/iOS traffic. The problem can’t be eliminated entirely, but the gap can be reduced through server-side measurement and first-party data matching.