Analytics and Tracking

ITP

Also called Intelligent Tracking Prevention

Safari's built-in rules that block cross-site cookies and cut short the life of many identifiers set by scripts.

Quick facts: ITP

Category
Analytics and Tracking
Also called
Intelligent Tracking Prevention
Level
Intermediate
Affects
Returning visitor counts, attribution windows, retargeting audience size
Where to see it
Safari developer tools, GA4 browser reports, Meta Events Manager
In this article4
  1. How ITP works
  2. Why ITP matters
  3. Where ITP is misunderstood
  4. What to do about it

How ITP works

Intelligent Tracking Prevention is a set of rules built into Safari that decide how long a cookie is allowed to live and whether one domain may recognise a browser on another. Cookies set by a domain other than the site being viewed are blocked outright. Cookies set under your own domain survive, but not always for as long as you asked.

The distinction that catches people out is how the cookie was written. A cookie created by JavaScript running in the page is capped at seven days regardless of the expiry date it was given, and shorter in some circumstances. A cookie set by your server in the response header is treated more generously. Same site, same visitor, same cookie name — completely different survival, decided entirely by which side of the connection wrote it.

Why ITP matters

Anything that depends on recognising a returning visitor is affected. Analytics identifiers expire, so a customer who researches you over several weeks arrives each time as a new user and the channel that originally introduced them gets no credit. Retargeting audiences fill more slowly. Conversion windows in ad platforms quietly become shorter in practice than the setting suggests.

How much this costs you depends on who your customers are. If you sell mainly in Australia, the UK, North America or the Gulf, Safari is a large part of your audience and the effect is significant. Selling mostly within Nepal, where Android is far more common, the same restrictions bite less — but the international side of a business usually feels it hardest, which is exactly the side where the customers are most valuable.

Where ITP is misunderstood

The first myth is that cookies no longer work at all. First-party cookies still work in Safari; they simply have shorter lives when scripts create them. Advice to abandon cookies entirely is wrong and expensive.

The second is believing a vendor who claims to bypass it. Nothing restores cross-site tracking in a browser that has decided against it. Server-side tagging and first-party endpoints make cookies live longer and reduce blocking, which is real and useful, but they are not a way around the rules — and a tool sold as a way around them is a reason to walk away.

The third is not testing. Owners check their own site in Chrome, see healthy numbers, and never look at Safari, where a large share of their reporting is being quietly truncated.

What to do about it

Start by finding your exposure. Look at the browser breakdown of your traffic, then compare returning-visitor rates and conversion rates between Safari and Chrome. A wide gap that has no business explanation is usually measurement loss rather than customer behaviour.

Where the numbers justify the work, move identifiers to being set by the server rather than by a script, and consider a first-party endpoint so tagging requests travel through your own domain. Where they do not, adjust how you read the data instead: watch trends rather than individual journeys, use shorter reporting windows consistently, and check the platform against something you can count by hand, such as enquiries in your inbox. Getting this right is ordinary analytics and tracking work, not a lost cause.

Do and do not

Do

  • Compare Safari and Chrome results before blaming performance
  • Have the server set identifying cookies where you can
  • Keep the whole customer journey on one root domain

Do not

  • Assume all cookies stopped working in Safari
  • Buy a tool that promises to bypass browser rules
  • Test measurement only in the browser you happen to use

Questions people ask about this

Does ITP block Google Analytics?

It does not block the tool, but it shortens the cookie that identifies a browser when a script sets it. The practical result is that returning visitors are counted as new ones more often, multi-visit journeys break apart, and direct traffic looks larger than it is. Totals stay usable; individual user journeys become much less reliable.

Can I stop ITP affecting my site?

You cannot switch it off — it is the browser's decision, not yours. You can reduce the damage by having your server set identifying cookies rather than JavaScript, sending conversions from your server to ad platforms, and keeping your whole customer journey on one root domain. Anyone selling a complete bypass is describing something Safari does not allow.

Do other browsers do the same thing?

Firefox applies similar protections by default, and Chrome blocks cross-site cookies in its private mode while offering users their own controls. Safari is the strictest, so it is the one worth testing against. If your measurement holds up in Safari, it will generally hold up everywhere else.

Related terms

Found this useful?

Share it, or ask an AI to summarise it

Back to the glossary

Knowing the term is the easy part

Applying it to your own site and budget is the work. Book a call and I will tell you what actually applies to you.