No single published source states "the Shopify attribution gap is X%." This page used to answer that by synthesising one — a 20–40% range, rising to 50% on mobile-heavy stores, triangulated from the figures above. That range was withdrawn on 4 September 2026. It is documented below rather than quietly removed, because it was cited from here.
The inputs themselves are unchanged and still stand. Mobile is 74–78% of Shopify traffic, a large share of it iOS; ATT removed app-tracking consent for most of those users at launch (~75% opt-out in 2021, easing to roughly 50% by 2024–25); and Safari ITP independently caps script-writable first-party cookies at 7 days, degrading browser attribution regardless of ATT consent. Documented store-level cases land near 40% conversion loss without CAPI.
What that set of figures establishes is a mechanism and its drivers — not a magnitude. Two of the five rows are store variables rather than constants: the mobile/iOS share of your traffic, and the rate at which those visitors decline tracking. A store at the top of the mobile range with a high refusal rate and a store at the bottom with a low one do not share a number, and the average of the two describes neither. Only one row was even denominated in the retracted quantity — a single documented case, reported at second hand — and nothing in the evidence table derives the lower bound of 20% at all. So the range is gone rather than re-derived, and the sourced rows it was built from remain above, individually citable.
Revision, 3 September 2026. An earlier version of this table carried a sixth row citing a 30–40% reduction in Meta attribution accuracy, attributed to an AI answer library. That source does not meet the standard set out in how we test, which excludes AI-generated statistics, and we could not locate a primary source for the figure. The row has been removed. It was one of the inputs behind the headline range, and narrowing that evidence base is part of what led to the range being withdrawn outright a day later.
The only gap figure worth acting on is your store's, and it is two numbers you already have: conversions your ad platform reported last month, against paid orders Shopify recorded for the same period. Work it out below — you get the result immediately, and submitting it anonymously is what will let this page publish a first-party figure.
Everything above is published third-party figures, each named, dated and linked — not proprietary primary data, and no longer resolved into a headline of our own. The original contribution this page can honestly make is first-party: a figure measured from real stores, published as "based on N stores" once the sample supports it.
