Meta Pixel will measure some purchases on customers' devices from November 12
According to advertising trade outlets, Meta will turn on a setting from November 12, 2026 that lets Meta Pixel measure some website purchases on the shopper's own device as well, for advertisers who take no action. This summary rests on reports by PPC Land, MetricMaven and Stape, not on Meta's own text. The setting can be turned off per dataset.
A default-on measurement setting starts on November 12
As PPC Land reports it, from November 12, 2026 Meta Pixel will measure some website purchases on the shopper's device and not only on Meta's servers. The setting is labelled "Measure conversions on customer devices" and is already on for advertisers who do nothing. In the Pixel settings panel of Events Manager, a grey line shows that measurement starts on November 12, 2026. MetricMaven notes that the November 12 date comes from Meta's email to advertisers and is not repeated in the help article.
According to Meta's notice as PPC Land relays it, no code changes are needed. The setting applies to datasets that already send events through Pixel or the Conversions API, and advertisers can switch it off for a dataset at any time. This article rests on what the three trade sources relay rather than a primary Meta publication; the body of Meta's own help article could not be read on our side. Every claim below is attributed to the outlet that makes it.
Matching happens on the device, and the limits come from help articles
According to PPC Land and MetricMaven, the match draws on details a shopper has already given the business, an email address for example, combined with short-lived device data that does not identify anyone. All Meta gets back is word that a sale took place and which campaign earns the credit. MetricMaven and Stape both relay that Meta says its attribution window and attribution model stay the same. Stape adds that the feature complements Pixel and the Conversions API instead of replacing them, and that no new CAPI integration or manual switch-on is expected. The technical mechanism is not documented by Meta, and Stape's explanation is its own guess.
MetricMaven, citing Meta help articles, lists three limits: people in the EU or the UK are outside the feature's reach, the Pixel has to sit directly in the page code (an iframe or an image-only pixel does not qualify), and the site has to require sign-in by email address or mobile number. By the same source, the opt-out route is Events Manager, Data sources, the relevant Pixel, Settings, then the toggle for the setting. PPC Land notes that these conditions do not appear in the settings panel text itself.
Reported sales can rise while actual sales stay flat
Stape warns that attributed conversions and reported conversion value may rise, which could lower reported CPA and lift ROAS even when real sales and spend are unchanged. MetricMaven likewise says counts can move for reasons unrelated to performance because attribution rules stay the same; that CPA and ROAS would shift accordingly is MetricMaven's own inference. More attributed conversions do not necessarily mean more real ones. The reports also list topics Meta has not explained publicly: whether recovered conversions feed bidding and optimisation, whether purchase value and currency are included, how consent signals are handled, and how Meta avoids counting one order twice when both a recovered purchase and a Conversions API event report it.
PPC Land points out that Black Friday falls on November 27, 15 days after the change. MetricMaven recommends recording a baseline before November 12 and holding out a region or a campaign as a control before attributing any uplift to Meta. Yiqi Wu of Aimerce, a Pixel tracking vendor for Shopify merchants, posted on X on October 26 that he would not opt out by default, and suggested comparing Meta-attributed purchases with Shopify paid orders day by day over the last month. Wu has a commercial interest in the topic.
The sources say nothing about Türkiye, so this section is commentary
None of the three sources mentions Türkiye. They do not say Türkiye is excluded or included; the EU and the UK are the only places named as excluded, and no list is given for other countries. This section is therefore not a fact drawn from the sources but explicitly UNALSOFT commentary, and it makes no claim about whether accounts in Türkiye will receive the setting.
The unknowns, one by one: any Türkiye-specific treatment, how consent and KVKK would apply locally, whether the setting will arrive switched on for Pixels used in Türkiye, and the legal nature of the data exchanged. We assert no KVKK fact. Our view is only that a measurement setting that matches against personal data merits a review by legal counsel, and MetricMaven likewise frames the setting as a legal decision for regulated clients.
The UNALSOFT view
We read this as a question of how to read the numbers in an ad report. When measurement coverage widens, the sales shown in a dashboard may not grow at the same pace as real business results. The practical consequence is plain: record current conversion and order ratios before November 12, leave the setting on or off as a deliberate choice, and compare any shift with real orders. That is why on the ad management side we treat reading the platform report next to business outcomes as its own check. Because the sources are trade outlets, anyone needing certainty should follow Meta's own statements.
Shall we read your numbers together before November 12 and Black Friday?
A short conversation is enough to record your current conversion baseline and manage the setting deliberately.