How should we handle tracking loss from privacy changes and ad blockers?
Our answer
Send every ad platform a server copy of each sale. Compare each platform's purchase count with store orders monthly and after each privacy change. Set budgets on store revenue and geo holdout tests, which need no user tracking.
- Confidence
- Medium
- Strongest proof
- 6 official rules
- 2 datasets, 1 practitioner, 8 opinions or claims
- Sources
- 16 sources
- 17 statements
- D
New here? Take the 1-minute tour
Why this is the answer
3 reasons and our judgmentServer events get past ad blockers. Shopify and Pinterest say browser ad blockers can't block them. Browser-only tracking loses 20 to 40% of purchase events by Jetfuel's estimate and 30 to 40% by Trackbee's, though neither shows data.
Apple's iOS 27 releases block third-party identity services in Safari. Common Thread Collective (CTC) saw Meta conversion rate fall about 15% over 11 days in October, and still advises checking your setup before blaming Apple.
You can measure sales without following users. Mix models and geo holdouts do this, as Fospha argues. Wicked Reports advises basing budgets on store revenue rather than platform-reported conversions.
Our judgment
- Run a server copy of each conversion alongside the pixel on every platform you spend on. Platform docs say server events escape browser blocking, and seller estimates put browser-only loss high.
- Reconcile each platform's purchase count with store orders monthly and after each privacy update. Treat a gap beyond a few percentage points as a setup or blocking problem first.
- Set budgets on store revenue, new customers and geo holdouts. These need no user tracking.
- Treat Apple's block on third-party identity services in Safari as a reason to audit your tracking, not rebuild it. CTC says to check the plumbing before blaming Apple.
Our judgment, built from Shopify, Pinterest, Meta, Jetfuel Agency, Trackbee, Eric Seufert, Ruler Analytics, Wicked Reports, Common Thread Collective, Fospha and Apple: our reading of these sources, not a finding from any one of them.
What the sources say
17 statements from 16 sourcesPinterest recommends using both its tag and its Conversions API for full coverage. The API is not blocked by browsers or ad blockers but needs backend work, hashing and user consent.
“we recommend that you use both to ensure maximum event coverage, eliminating any data gaps that may occur from using one method only.”
Triple Whale, a measurement vendor, says Safari 27 blocks a private list of identity domains, so UID2, ID5 and LiveRamp IDs are lost. Its own pixel is not blocked, and Apple can change the list without an iOS update.
“Check Safari when something looks off: Split the metric by browser and OS version, then compare it with your store's orders.”
Shopify's Meta app offers Standard, Enhanced and Maximum data sharing. Enhanced and Maximum add the Conversions API, which sends purchases server to server so browser ad blockers cannot stop them.
“Data sent from server to server can't be blocked by browser-based ad blockers.”
CTC saw Meta conversion rate down about 15% over 11 days in October, as iOS 27 reached about 26% of users, but notes Prime Day timing and says to check Shopify against Meta purchase counts first.
“check the plumbing before blaming everything on apple.”
Google says consent mode modeling needs 700 ad clicks over 7 days per country and domain group. Its example of a 50% consent rate losing only 19% of conversions is illustrative, not measured.
“You have a daily ad click threshold of 700 ad clicks over a 7 day period, per country and domain grouping.”
Stape, which sells server-side tracking, relays that Meta's on-device conversion matching starts November 12, 2026, on by default outside the EU and UK, and may lift reported conversions without real sales changing.
“advertisers may see an increase in attributed conversions and reported conversion value.”
Apple says Safari in its version 27 releases blocks network access to identity resolution services that collect emails to track people across sites, while first-party personalization is not blocked.
“First party personalization is not tracking, and is not what our technologies seek to prevent.”
Meta says third-party tools can under-count Meta conversions because they rely on cookies and referrer URLs and miss view-through and cross-device sales. Meta's explanation partly favors itself.
“may under-report conversions on Meta technologies”
Trackbee, which sells a Shopify tracking app, recommends running the TikTok Pixel and Events API together with deduplication, then checking TikTok's purchase count against Shopify orders.
“Without deduplication, running both Pixel and Events API inflates your conversion count in TikTok Ads Manager.”
Eric Seufert says brands send full server-side conversion data only to their biggest platforms, so a smaller channel may need better results to win budget. Equal ROAS does not mean equal value.
“Equal campaign ROAS therefore doesn't imply equal commercial value”
Wicked Reports, an attribution vendor, says Conversions API purchases should match Shopify orders within a few percentage points, with matching event IDs, and to decide on your own revenue data.
“within a few percentage points”
Fospha, which ranks its own mix model first, defines privacy-safe measurement as mix models, geo lift studies and aggregate first-party data rather than user-level path attribution.
“What privacy-safe measurement is not: a multi-touch attribution tool with a cookie consent banner added on top.”
Jetfuel, an agency, estimates that browser-only tracking loses 20 to 40% of purchase events, with no data shown, and says wiring AppLovin's Conversion API before launch shortens the warmup.
“Wiring CAPI before launch compresses the warmup phase materially.”
Didomi, which sells consent tools, found European cookie banner opt-in rates from 49.8% to 64.4% across 16 industries in its 2025 client data, with fashion and jewelry at 55.4% and beauty at 54.6%.
“Some configurations may correlate with better performance in our data but do not necessarily reflect sound privacy practice.”
WeltPixel, which sells a Shopify tracking app, says a browser pixel alone suffices under $1K a month of ad spend; add a server path around $2K to $10K, and custom servers fit 10,000+ orders a month.
“A gap above 10% is worth investigating.”
Didomi, which sells consent tools, says US cookie banners use opt-out, and in its 2025 client data acceptance averaged 94.2% nationally, from 90.6% to 99.2% by state, with California at 91.3%.
“U.S. banners operate under an opt-out model”
Ruler Analytics, an attribution vendor, says Shopify is the source of truth for orders and revenue and GA4 for traffic. Shopify almost always shows more, and gaps above 10 to 15% usually signal a setup problem.
“Shopify is the reliable source of truth for transaction and revenue data. GA4 is better suited for traffic and on-site behaviour.”
In the full answer
For readers who sign in- What to do this week3 steps
- The numbers behind it3 figures from cited sources
- Where each source standsA chart placing each source on the question
- The strongest disagreementThe best case against the answer, from a cited source
- How sure we areWhat its confidence rests on, and what would change the answer