Most of the money the third-party cookie story has cost advertisers was spent getting ready for it. Readiness audits. Cookieless identity contracts. A line item on the agency invoice called Privacy Sandbox migration, billed monthly, against a date that kept moving. I sat through a lot of those pitches between 2022 and 2024. The honest ones told you to fix your conversion tracking, which you needed to do regardless. The rest sold insurance against an event with no confirmed date, and charged like the date was next quarter.
Here is the part of the record that is not in dispute, because Google published it. In January 2020, the Chromium blog said Chrome would make third-party cookies obsolete within two years, with Privacy Sandbox as the replacement: browser APIs meant to handle interest-based targeting, remarketing, and conversion measurement without letting anyone follow a user from site to site. Two years came and went. In January 2024, Chrome switched on Tracking Protection for 1% of users, restricting third-party cookies by default for that slice. That 1% is where most of the "cookies are dying" coverage came from, and it is worth being precise about what it was: a test population, not phase one of a rollout. The plan has been revised more than once since, which is why I now send people to Google's own Privacy Sandbox site for the current status rather than to a vendor deck or a trade-press headline written eighteen months ago.
So I am not going to relay Google's latest position second-hand and pretend it is settled. What I can tell you, from accounts I have run and watched through the whole saga, is which parts of this actually touched performance and which parts never did. The short version: Chrome was rarely the reason your remarketing pool shrank, the preparation work still pays even though the deadline evaporated, and search budget is the least exposed money in your account no matter what any browser does next.
The reason the deadline kept sliding is simple if you ever tried to buy against the replacement. The Sandbox APIs were available for testing for years, and in that whole time I never saw an advertiser or an agency I worked with move real budget into Topics-based targeting and keep it there. Nobody switches off a cookie that still works for a browser API that returns coarser signal and reports fewer conversions. For deprecation to happen, the replacement had to be good enough that buyers would accept a short-term loss on the way through, and it never got there. Once I understood that, I stopped reading the announcements as plans and started reading them as weather. My rule with clients from mid-2024 on: no line in the budget gets approved because of a cookie deadline, only because it improves measurement in the browsers people already use.
This is the part both camps miss. The advertisers who ignored the saga entirely were right about Chrome and wrong about everything else, because third-party cookie coverage has been eroding for years without Chrome's participation. Safari has blocked third-party cookies outright by default since 2020, and Safari is a large share of US traffic, disproportionately the iPhone-owning, higher-intent share. Firefox blocks them too. Apple's App Tracking Transparency cut off a chunk of in-app signal on top of that. And in the EEA, Google will not use your traffic for personalisation or remarketing unless your consent signals are being passed correctly, which means a misconfigured banner can do more damage to your data than any browser policy on this list. None of that reversed when Google's cookie plans changed. If your remarketing pool has been thinning and your view-through numbers have been drifting downward, look there first.
Here is the odd consequence: most teams did the right work for the wrong reason, and they get to keep the benefit. First-party data collection, server-side tagging, Enhanced Conversions, offline conversion imports, a consent setup that actually transmits signals instead of quietly swallowing them. Every one of those improves measurement in a browser that still allows third-party cookies. Enhanced Conversions is the clearest example I have. In accounts with a clean checkout or lead form, turning it on has recovered somewhere in the range of 8 to 15% of conversions that were previously going unattributed, and none of that recovery has anything to do with cookie policy. It works by matching hashed first-party data you already collect against signed-in Google users. The cookie deadline was the excuse a lot of teams needed to get budget for tracking work. The excuse expired. The tracking didn't.
What I would stop paying for is anything whose pitch depends on the deprecation date. If a vendor opens with the end of third-party cookies, ask two questions: which Sandbox API have you implemented for a client, and what did it return compared to the cookie-based setup it replaced. I have asked that maybe a dozen times. I have never once received an answer with a number in it. Same test for an agency carrying a Privacy Sandbox readiness workstream on a monthly retainer: ask what shipped last month. If the answer is a document, you are funding a research project about a schedule that Google keeps revising.
I should say plainly that I write for groas, which runs Google Ads accounts autonomously, so weigh the next few sentences accordingly. The reason this saga never worried me on the search side is that the optimisation loop doesn't need cross-site identity to work. It reads the query, the match type, and the conversion data from your own account, then moves bids, budgets, negatives, and the landing page against it. When someone types "emergency furnace repair 2am", that string tells you more about intent in one instant than a 90-day behavioural profile ever did, and no browser setting can take it away from you. Display prospecting and audience-based buying are genuinely exposed to signal loss. Search is the least cookie-dependent money in most accounts, which is one more reason I would concentrate budget there before I would fund a readiness project.
So the next time a cookieless strategy deck lands in your inbox, check the date on the file and then check whether the argument survives without a deadline. If the whole case rests on cookies going away by a particular quarter, it rests on a date Google has already revised more than once. Take the hour you were going to spend on Sandbox planning and spend it verifying that your conversions are firing. That one pays this month, and it pays again in whichever browser breaks next.
