By the Digital Empire Regulatory Research Team (PixelProof Analysis Team) · Reviewed by Andy Gaber, Founder, Digital Empire Holdings LLC · Published August 30, 2026 · Last updated August 30, 2026
Meta attribution windows decide which ads get credit for which conversions, and the rules have moved several times since 2021. This guide walks the current 2026 state end to end for Shopify DTC advertisers: what each window means, where they live in Ads Manager, why the pre-iOS-14.5 28-day-click default is gone, how iOS App Tracking Transparency and Aggregated Event Measurement cap the longer windows, why Meta, GA4, and Shopify never quite agree, and how attribution changes reshape ROAS math in ways that look like campaign performance changes but are actually reporting changes.
An attribution window is a look-back period. When a conversion happens on the merchant's site — a Purchase, an Add to Cart, a Complete Registration — Meta looks backward from that moment across the selected window and asks whether the same user clicked or viewed a Meta ad inside that window. If yes, the ad gets credit; if no, the conversion is unattributed in Meta reporting even though it happened. The window is not a data collection setting; it is a reporting and optimization setting applied to events that were already collected via the browser Pixel and the server-side Conversions API.
The setting itself lives at the ad set level in the campaign create-or-edit flow inside Meta's attribution setting documentation. Since the 2021 unification, the choice made when the campaign is built determines how Meta's auction bids that campaign — the auction optimizes toward conversions attributed inside that window, not conversions attributed inside whatever window the advertiser happens to be viewing in reports later. That distinction is load-bearing: changing the reporting-side comparison window in the Ads Manager column picker does not change what the auction optimized for.
1-day click. Meta credits a conversion if the user clicked a Meta ad within 24 hours before the conversion. This is the tightest causation window and the closest to same-session attribution. It is the standard choice for advertisers who want conservative credit and are willing to trade off delayed-conversion volume.
7-day click. Meta credits a conversion if the user clicked a Meta ad within 7 days before the conversion. This is the default click-side setting and captures multi-session consideration cycles common in retargeting and mid-to-high AOV categories. Any conversion attributed to a click 2 to 7 days before it happened is inside 7-day but outside 1-day, and that gap is where most of the retargeting delta shows up.
1-day view. Meta credits a conversion if the user viewed a Meta ad within 24 hours before converting, without clicking. This is view-through attribution, and it is the most contested column in the report because the causal link between an unclicked impression and a same-day purchase is weaker than for a click. It is meaningful for high-frequency brand campaigns and video, and often noisy for cold-prospecting static image sets.
7-day click plus 1-day view. The default combined window since Meta's 2021 unification. This is what Ads Manager shows by default when no override is set, and what the auction optimizes toward on standard conversion campaigns. It combines a moderately generous click window with a tight view-through window.
Engaged view. A video-specific window: Meta credits a conversion if the user watched at least 10 seconds of a video ad, or 97 percent of a video shorter than 10 seconds, and then converted within the window. This is meaningful for accounts running Reels and Feed video creative with real thumb-stop performance and has no signal for accounts running only static images. Meta's comparison window documentation walks through how each of these appears in the report.
Before Apple shipped App Tracking Transparency in April 2021, Meta's default attribution window for standard events was 28-day click plus 1-day view, with account-wide selection of 1 day, 7 days, or 28 days available in reports. A retargeting campaign that closed a sale 20 days after a click was fully credited under the default. That world made intuitive sense for higher-consideration DTC categories where multi-week research cycles were normal.
The ATT rollout collapsed the fidelity of tracking on iOS-opted-out users beyond a short horizon. Meta responded by moving the default to 7-day click plus 1-day view for standard events, deprecating 28-day click as an option for standard events entirely, and shifting the selection point from the account level to the campaign level so each campaign carries its own window. That is the world we still live in as of 2026 — the 28-day-click column has not returned for standard events, and none of the Advantage+ surfaces expose it as a choice.
The direction of travel across 2024, 2025, and into 2026 has been consistent: shorter deterministic windows, more emphasis on modeled conversions to fill the gap on iOS-opted-out cohorts, and more automated allocation via Advantage+ Shopping Campaigns. Meta has not reintroduced 28-day click for standard events. Advantage+ Shopping continues to default to the platform-recommended window and to hide the per-campaign window override that manual campaigns still expose. Advertisers who want the older behavior of long-look-back credit for retargeting have to accept the reporting rather than reset it.
The practical 2026 posture for a Shopify DTC advertiser is: leave the ad-set-level window on the default 7-day click plus 1-day view for standard conversion campaigns, use the comparison-window column in reports to see how different windows would credit the same period, and treat any large gap between windows as diagnostic information about the consideration cycle rather than as license to change the auction-side setting. Meta's attribution comparison guidance is the canonical source for how the comparison view differs from the attribution setting.
App Tracking Transparency, per Apple's privacy and data use policy, requires apps to request explicit permission before tracking users across other companies' apps and websites. A large fraction of iOS users decline that prompt inside the Facebook and Instagram apps. For that opted-out cohort, Meta cannot use the ATT-restricted signals to attribute a conversion back to a user-level ad click.
Meta's answer is Aggregated Event Measurement (AEM). AEM caps attribution for iOS-opted-out traffic at a 7-day window, limits view-through attribution severely for that cohort, and requires the merchant to configure up to 8 prioritized conversion events per verified domain. Practically, this means the longer the reporting window an advertiser selects, the smaller the incremental data it captures from iOS-opted-out users — a 7-day-click window is at the AEM ceiling; a longer view-through window mostly picks up non-iOS traffic.
The domain-verification part of AEM is where the silent-loss failure mode lives. If the domain listed in Meta Business Manager under Brand Safety then Domains is a legacy vanity domain or a myshopify.com subdomain that now 301-redirects to the current storefront, AEM does not automatically follow the redirect, and iOS-opted-out attribution silently degrades regardless of which window is selected in the report. This is one of the checks a monitoring tool such as PixelProof surfaces on scan.
Meta, GA4, and Shopify almost never agree on the same order count for the same day. The differences are architectural, not defects. Meta credits conversions inside its own window using its own event stream (browser Pixel plus CAPI, subject to iOS ATT and AEM). GA4 by default uses data-driven attribution across the channels it observes and has its own set of session-timeout and channel-grouping rules. Shopify records the raw order and attributes the session via UTM parameters, landing page, and its own last-non-direct logic.
On a healthy stack, Shopify sees the largest number (every order), GA4 sees fewer than Shopify (session drops, ad-blocked GA4 loads, iOS Safari ITP session boundaries), and Meta sees fewer than GA4 for Meta-attributed sales specifically (iOS-opted-out cohort, view-through opt-in, ad-blocker impact on the browser Pixel). Any DTC founder who tries to reconcile these to the same absolute number is chasing a target that does not exist. The useful comparison is trend agreement: if all three move together, the stack is healthy; if one flattens while the others move, that platform is broken or misconfigured.
Prospecting campaigns. Default to 7-day click plus 1-day view. Prospecting depends on Meta's auction feeding a strong conversion signal to the optimizer, and the default window is what the auction is best tuned for. Do not shorten to 1-day click on prospecting unless a specific test justifies it — the shorter window can starve the optimizer of the multi-session conversion signal that is common in cold traffic.
Retargeting campaigns. The default 7-day click plus 1-day view is usually still the right pick. Some accounts benefit from 7-day click alone (no view-through) on retargeting if view-through is inflating credit relative to the observed lift — a common pattern when retargeting audiences are heavily overlapping with high-frequency organic touchpoints. A tight lift test on view-through inclusion inside retargeting is worth running once per year on any account north of five figures in monthly retargeting spend.
Brand and video campaigns. Engaged view is the argument for running these on Meta at all — if the credit for a viewed-video-then-later-converted path is not captured, the reporting under-credits video creative. Engaged view has real signal for accounts with real video creative and no signal for accounts running only static image ads.
Advantage+ Shopping campaigns. The advertiser does not choose. Advantage+ uses the platform default and optimizes across placements and audiences automatically. The practical implication is that any comparison between an Advantage+ Shopping campaign and a manual campaign has to hold the attribution window constant in the comparison-window report column, otherwise the comparison is unfair.
The default Ads Manager report shows Results, Cost per Result, and Purchase columns computed against the campaign's attribution setting. To see how a different window would have credited the same period, open Columns then Customize columns then Compare attribution settings, and pick up to three windows to display side by side. This adds columns such as "Purchases (7-day click)" and "Purchases (1-day click)" next to the default. Nothing about spend, delivery, or the underlying event stream changes — only the reporting-side attribution is being re-run against the same events.
The most common mistake operators make is treating the comparison-window column as a proxy for what the campaign would have delivered if the auction had optimized against that window. It is not. The auction bids toward the campaign's attribution setting; changing the comparison-window column just changes the credit assignment on already-delivered events. Actually optimizing toward a different window requires editing the ad set attribution setting, which resets the learning phase for that ad set.
Advantage+ Shopping Campaigns, per Meta's Advantage+ Shopping documentation, automate audience selection, placement, and creative rotation. They apply the platform-default attribution window and do not expose the per-ad-set attribution override that manual campaigns do. From an advertiser perspective, that means the auction-side attribution question is settled — the choice is whether to run Advantage+ Shopping at all, not what window it optimizes against.
The reporting-side comparison-window columns are still available for Advantage+ Shopping campaigns. That is often the more useful lens on Advantage+ performance than the default view, because it lets an operator ask: how much of Advantage+'s reported Purchase count is inside the 1-day-click window versus the full 7-day-click plus 1-day-view default? A campaign whose reported ROAS collapses when the comparison window narrows to 1-day click is one whose reported ROAS is being carried by longer-look-back credit and view-through, and that is diagnostic even when the operator has no auction-side control to change.
PixelProof does not change attribution windows — window selection is an advertiser decision. What PixelProof does is monitor the event stream the windows are computed against, so that when the numbers move, an operator can tell whether the move is real, is a reporting-side window artifact, or is a broken pixel producing bad inputs to the attribution.
The scan detects browser Pixel event loss on pages where the Pixel should be firing, CAPI event-share degradation via the Conversions API deduplication contract, event_id mismatches that produce duplicate Purchase counts and inflate reported conversions across every attribution window at once, and AEM-verified domain drift that quietly collapses iOS-opted-out attribution regardless of which window is selected. The output is a pre-vs-post comparison: what the event stream looked like before the change, what it looks like now, and which of those two states is producing the numbers currently showing up in Ads Manager under whatever window is selected. That is the reason a monitor exists at all — the window selection is downstream of whether the events firing into it are correct.
For the iOS 14.5+ App Tracking Transparency deep dive with the aggregate impact numbers Meta itself published, see iOS 14.5 ATT Impact on Shopify Tracking (2026). For the full server-side Conversions API walkthrough on Shopify 2026, including deduplication and consent gating, see Meta Conversions API Setup for Shopify (2026 Deep Dive). For how Klaviyo and the Meta Pixel co-exist on Shopify with dual-fire and monitoring, see Klaviyo + Meta Pixel Integration Guide 2026. For the companion tracking-event schema map on the Shopify checkout upgrade, see Shopify 2.0 Tracking Events Guide 2026.
Not sure whether your reported numbers reflect a real change or a broken pixel? Run a free scan of your tracking stack → or read the iOS 14.5 ATT impact deep dive for the constraint that shapes every window above.
An attribution window is the period of time Meta will look backward from a conversion to decide which ad, if any, should get credit for causing it. If a shopper clicks a Meta ad on Monday and buys on Wednesday, a 7-day-click window credits Monday's ad; a 1-day-click window does not, and the sale shows up as unattributed in Meta reporting even though the shopper reached the store from an ad. The window is set per campaign in Ads Manager and applied at report time, which means changing it changes historical numbers without any new spend actually moving.
For standard optimization events on Meta as of 2026, the default is 7-day click plus 1-day view. Options available at the campaign level are 1-day click, 7-day click, 1-day view, 7-day click plus 1-day view, and engaged view for video. The pre-iOS-14.5 28-day-click default is no longer selectable for standard events. Advantage+ Shopping campaigns use the default and expose fewer window controls to the advertiser by design.
Each platform uses a different attribution model on the same set of orders. Meta credits the last ad clicked or viewed inside its own window; GA4 by default uses data-driven attribution across all channels it observes; Shopify records the raw order and attributes the session via its own landing-page and UTM logic. Reporting drift between the three is the expected state, not a defect. The right comparison is trend agreement, not absolute-number agreement.
ATT restricts third-party tracking on opted-out iOS traffic, and Aggregated Event Measurement caps that cohort at a 7-day window with view-through attribution highly restricted. In practical terms, longer windows still exist as report options but capture progressively less iOS-opted-out data the longer they run, so a 7-day-click Purchase count on an iOS-heavy DTC store is closer to complete than the same 7-day-view number, and both are less complete than what a purely first-party integration such as Shopify or Klaviyo sees.
A 1-day-click window is closer to real-time and tighter to actual causation for high-consideration or impulse purchases where the shopper buys the same session they click, and it is generally more useful for prospecting-heavy accounts optimizing for volume. A 7-day-click window captures multi-session consideration cycles more common in retargeting, upsell, and higher AOV categories. Neither is universally right; the tradeoff is between how much delayed-conversion revenue you accept as ad-driven and how much bid signal you feed the auction.
Engaged view credits a conversion when a viewer watched at least 10 seconds of a video ad, or 97 percent of a video shorter than 10 seconds, and then converted inside the window. It is only meaningful on video creative and is most useful for accounts running Reels and Feed video with real thumb-stop performance. Turning it on for accounts running only static image ads adds no signal.
No. PixelProof monitors whether the events the windows are computed against are firing correctly — browser Pixel, server-side Conversions API, deduplication via event_id, and AEM-verified domain integrity. Window selection is an advertiser decision made inside Ads Manager, and it does not fix a broken pixel; a 7-day-click window on a stack that has stopped firing Purchase events still reports zero. The two work at different layers, and both matter.
No. This is a technical explainer on how Meta's attribution windows work and how iOS ATT constrains them. Revenue recognition for accounting purposes is governed by GAAP or IFRS and the merchant's own auditor, not by whatever number Ads Manager displays. Digital Empire Holdings LLC is a monitoring and data vendor, not a law firm or an accounting firm.
Attribution windows are downstream of whether the events firing into them are correct. PixelProof scans your Meta Pixel, CAPI, and AEM domain integrity continuously.