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.
