For most advertisers outside the EU and UK, leave Meta's "Measure conversions on customer devices" on, but take a baseline before November 12, 2026 and stop reading Ads Manager ROAS as if nothing changed. The setting lets Meta credit some website purchases it could not connect to an ad before, so reported conversions and ROAS can rise while real sales stay flat. Turn it off for now if client billing, bonuses or automated rules depend on Ads Manager numbers and you cannot re-baseline them before Black Friday. Adstudio makes the before/after check one question: it puts Meta-reported purchases, split by attribution window, next to your GA4 and store revenue for the same dates.
TL;DR
- What it is: a per-pixel setting in Events Manager that matches an ad interaction to a later purchase on the shopper's device, using details the customer already gave you, such as an email address (Meta Business Help Center).
- When: the Events Manager panel reads "Measurement starting November 12, 2026", and it is on by default for eligible pixels (PPC Land).
- Who: not active for people located in the EU and UK; reported eligibility also needs the pixel on the page itself and a site where customers log in with an email or phone number (Relevant Audience).
- What changes: which purchases Meta can count, not your attribution window or model. Expect more attributed purchases, lower reported CPA and higher reported ROAS (Stape).
- What to do: export a 30 to 60 day daily baseline now, track Meta purchases against backend orders, and annotate November 12 on every chart.
What is "Measure conversions on customer devices" in Meta?
It is a new toggle under each Meta Pixel's settings in Events Manager. Meta's own description in the panel is that, with it on, conversion measurement happens on the customer's device instead of exclusively on Meta's servers (PPC Land).
Meta's help article, About measuring conversions on customer devices, describes the idea in plain terms. Today, some purchases on your website never get tied back to the ad that started them, for example when someone sees an ad, comes back days later and buys. With this setting, the match is made on the shopper's device, using information the customer already shared with your business (such as an email address) plus temporary, non-identifying device information. Meta then receives confirmation that a purchase happened and which campaign to credit, and says the identifying details used for the match are not sent to it in that step (Stape's summary of Meta's article).
Two things it is not:
- It is not a new attribution window. Meta says the setting changes which conversions can be measured, not how they are credited. Your 7-day click / 1-day view (or whatever you use) stays the same.
- It is not a replacement for the Conversions API. It works alongside your existing Pixel and Conversions API setup, and no code change is needed to turn it on (Relevant Audience).
When does it start, and does it apply to my pixel?
The grey note under the toggle reads "Measurement starting November 12, 2026", and trade press reports that it switches on by default for pixels already sending events, unless you opt out (PPC Land). That date is 15 days before Black Friday on November 27.
Whether it does anything for you depends on eligibility. As reported from Meta's help pages:
| Condition | What it means for you |
|---|---|
| Location | Not active for people located in the EU and UK. If most of your buyers are there, expect little change. |
| Pixel installation | The Meta Pixel must sit directly on the page, not inside an iframe and not as an image-only pixel. Embedded or third-party checkouts are where to look first. |
| Login | The site must require customers to log in with an email address or mobile phone number. A store where most people check out as guests will see less effect. |
| Default state | On for eligible pixels. It can be turned off at any time, per pixel. |
Sources: Relevant Audience, Follow the Spend, both citing Meta's help article.
If your business runs several pixels (one per brand or per site), check each one. The setting lives on the pixel, not on the ad account.
Will it change my ROAS and CPA in Ads Manager?
Probably, and that is the whole reason to plan for it. If Meta connects purchases it used to miss, the purchase count and conversion value in Ads Manager can go up while ad spend and real orders stay exactly where they were. Reported CPA falls and reported ROAS rises from the counting change alone (Stape).
That shift matters in three places:
- Week-over-week and year-over-year comparisons. November 2026 is now measured differently from November 2025. A "ROAS up 15%" headline after November 12 says nothing about performance until you check it against orders.
- Automated rules and scripts. A Meta automated rule that raises budgets when CPA drops below a threshold will read a counting change as a performance win. Review rules keyed to CPA or ROAS before the date.
- Client and finance reporting. If an agency fee, a bonus or a budget decision is tied to Ads Manager ROAS, the target needs a note or a new baseline.
What Meta has not said publicly, as of November 2026: whether purchase value is always included for these matches, whether the recovered conversions also feed bidding and delivery or only reporting, and whether they get their own label in reporting (PPC Land). Plan as if you will not be able to separate them in Ads Manager.
How do I turn off Measure conversions on customer devices?
The toggle is per pixel, in Events Manager:
- Open Meta Events Manager from Business Settings or the Ads Manager menu.
- Click Data sources and select the pixel you want to change.
- Open the Settings tab.
- Scroll below the Meta Pixel settings section to Measure conversions on customer devices.
- Switch it off, and repeat for every other pixel your ads optimize for.
Steps as described by Relevant Audience and Stape. You can switch it back on later, so turning it off for the holiday period is a reversible decision. Write down the date you changed it, because any switch, on or off, creates its own break in the numbers.
Should I keep it on or turn it off?
Our view: the setting does not change what your ads do, it changes what Meta can see. Seeing more is useful as long as you are not fooled by it. Here is how we would decide:
| Your situation | Our recommendation |
|---|---|
| You judge Meta against backend orders or GA4 already | Keep it on. Re-baseline and move on. |
| Client billing, bonuses or targets use Ads Manager ROAS | Off until you agree a new baseline with whoever reads the number, then on. |
| Budget rules or scripts trigger on CPA or ROAS thresholds | Keep it on, but tighten the thresholds or pause the rules for the first two weeks. |
| Health, finance or another sensitive category | Ask your legal or privacy owner first. The setting uses customer-provided details on the device, and how consent signals apply is not documented yet. |
| Most buyers are in the EU or UK, or checkout is guest-only | Leave the default. Little is likely to change, but annotate the date anyway. |
Two things we would not do. We would not change the Conversions API, Advanced Matching or the checkout app in the same week, because then no one can tell which change moved the numbers. And we would not credit a rise in Meta-reported purchases as real lift without checking orders in your own system.
How do I tell real sales lift from extra attribution?
The test is simple: if Meta reports more purchases but your store did not take more orders, the difference is attribution. Set it up before November 12:
- Export a baseline. Daily Meta-reported purchases and purchase value for the last 30 to 60 days, with the attribution setting written next to it (Follow the Spend).
- Pull the same days from your backend. Orders and revenue from your store or CRM, and purchases from Meta traffic in GA4.
- Track a ratio, not a total. Meta-reported purchases divided by backend orders, by day. Before November 12 it should be fairly steady. A step up right on the date, with orders flat, is the setting at work.
- Split Meta's number by attribution window. Compare click-through and view-through results separately before and after the date. Our guide on why Meta Ads and GA4 numbers don't match explains the windows and the March 2026 link-click change that already moved them once this year.
- Annotate November 12 on every dashboard and report that uses Meta conversions, so a reader in January knows why the line moved.
- For proof of incremental sales, run a holdout. Pause Meta in a set of regions, or keep a control group, and measure orders in your own backend. Follow the Spend suggests watching the ratio for four to six weeks after the switch before drawing a conclusion (checklist).
If you are also getting ready for the holiday peak, our Black Friday checklist for Google and Meta ads covers tracking checks, review lead times and budget changes in the right order.
How Adstudio helps you check the numbers before and after November 12
Adstudio connects Meta Ads, Google Ads, TikTok Ads, Google Analytics 4 and Shopify to one workspace, so the comparison above is a question, not a spreadsheet.
- Meta results split by attribution window. Ask the AI assistant for Meta Ads "Compare my Meta purchases and purchase value for the 14 days before and after November 12, split by 1-day click, 7-day click and 1-day view" and you get each window side by side, counted for what each ad set optimizes for.
- Store revenue next to Meta's ROAS. With Shopify connected, ask "Put my Shopify revenue next to Meta's reported ROAS for the last 30 days" and you see whether reported return and real takings moved together.
- GA4 for the same dates. Purchases and revenue by source, medium and campaign from your GA4 property, so Meta's number sits next to an independent one. More on the AI for Google Analytics page.
- A pixel health read. Ask "Check my Meta pixel" to see each pixel's last event, events by name, browser versus Conversions API events and how many events carry customer information, which is the kind of detail that decides how much this setting can match.
- Alerts on the numbers that matter. Set an alert on Meta cost per conversion or ROAS so a sudden step after the date reaches your inbox instead of waiting for the monthly report; see how to set up ad spend alerts.
You can connect Meta Ads, GA4 and your store from $19/month.
FAQ
What does "Measure conversions on customer devices" mean on Meta? It is a Meta Pixel setting that lets Meta match an ad interaction to a later website purchase on the shopper's device, using information the customer already gave the business plus temporary device information. Meta receives which campaign to credit rather than the identifying details.
Is Measure conversions on customer devices on by default? Yes, for eligible pixels. The Events Manager panel shows measurement starting November 12, 2026, and you can turn it off at any time per pixel.
Does it change my Meta attribution window? No. Meta says the attribution window and model stay the same. The setting changes which purchases can be measured, which is why reported totals can move without any change in your settings.
Do I still need the Conversions API? Yes. The setting works alongside the Pixel and the Conversions API and does not replace either. Keep your server-side events running and avoid changing them in the same week.
Why did my Meta ROAS go up in November without more sales? If it jumped around November 12 while your store orders stayed flat, the likeliest cause is that Meta began crediting purchases it could not connect to ads before. Compare Meta-reported purchases with backend orders by day to confirm.
Does it apply in the EU and UK? No. Meta says the setting is not active for people located in the EU and UK, so advertisers selling mainly there should see little change.
Platform behaviour described as of November 2026, per Meta's Business Help Center and the linked trade reports of Meta's Events Manager notice.