# Attribution When the Customer Buys In-Store: The Offline Conversion Gap

Source: https://attributionplaybook.com/blog/2026-11-15-attribution-in-store-purchase-offline-gap.html

November 15, 2026

*The Meta Ads Manager shows 45 conversions. The Shopify backend shows 68 orders. The gap is not a rounding error. It is the customers who clicked an ad and bought in the store, and Meta never saw them, because the register has no browser and the Pixel never fired.*

## The register has no browser, and that is the whole problem

Retail POS hardware operates completely outside browser cookies. A cash register checkout has no \_fbp cookie, no browser session, and no Pixel event, so the standard tracking that works online simply does not exist in the store. The customer can click an ad on Monday, walk into the store on Thursday, and buy, and Meta will never connect the two, because the only thing that connects them is the customer, and the customer was never identified at the register.

The consequence is not just a reporting gap. It is an optimization gap. Meta's algorithm builds its audience model from the conversion signals it receives, and when the in-store conversions are invisible, the algorithm optimizes for the customers it can see, which are the online buyers. The store's best customers, the ones who research online and buy in person, are missing from the model, and the budget moves toward the channels that report well rather than the channels that sell well. The [server-side tracking post](2026-11-14-server-side-tracking-attribution-gap.md) is the same gap in a different shape: the Pixel misses conversions the browser never sees, and the fix is to send them from the server instead.

The scale is not small. The Pixel misses 30 to 40 percent of conversions on average, and the attribution gap that sat at 30 to 40 percent right after iOS 14.5 has expanded to 50 to 70 percent for most ecommerce accounts, because ad blockers, Safari's Intelligent Tracking Prevention, and iOS opt-outs have all grown since. The in-store gap sits on top of that, and for an omnichannel brand it is the larger of the two.

## The fix: send the register to Meta

The setup is straightforward, and it starts at the register. Send in-store Shopify POS purchases to a Meta Offline Event Set, formatted as Purchase events, with the customer matched by email or phone hashed with SHA-256. Upload the data either by manual CSV or by an automatic Conversions API sync, and wait up to 72 hours to judge match quality. The CAPI sync takes more setup, but it cuts out repeated manual work, and for a multi-store setup it is usually the better choice.

The key field is the click id. If you capture the fbclid at landing and persist it, the offline event can be matched deterministically to the ad click. Without the click id, the attribution becomes probabilistic rather than deterministic, because Meta has to match on hashed email or phone alone. Facebook matches offline conversion events to ad impressions and clicks within 28 days after a click and 1 day after an impression. Events sent outside those windows will not be attributed to a specific ad, but they will still appear in reporting as general offline events, so the revenue is not lost, only the attribution.

The failure modes are specific, and they are all worth checking. No email or phone at checkout means weak matching, and the event is unlikely to be attributed. Late uploads mean missed or uneven attribution, so a regular automated sync beats a sporadic backfill. Duplicate transaction ids mean inflated conversion counts, because the same sale gets counted twice. And a CAPI setup without deduplication, the same event\_id through both the Pixel and CAPI, means the online sale gets counted twice as well. The [7-day click post](2026-11-10-meta-7-day-click-1-day-view-vs-ga4.md) is the companion check: the window that decides what Meta counts is the window the offline setup has to respect, and the offline event that arrives after the window closes is an event that never gets attributed.

## What to do on Monday

Turn on Meta data sharing to Maximum in Shopify, which enables the Conversions API, and create an Offline Event Set in Events Manager named for the store and the year. Start collecting email or phone at the POS checkout, because that identifier is the whole setup, and without it the matching is weak. Then run a test: make a test purchase in store with a known email, send the event, and check the Event Match Quality score and the transaction count against Shopify POS after 72 hours.

The test is the point. A setup that is never validated is a setup that silently fails, and the failure looks exactly like a channel that does not work. The store that sends its register to Meta learns which ads drive foot traffic. The store that does not keeps optimizing for the online buyer and wondering why the store is quiet. The difference is a CSV upload and an email field at the register, and it is the cheapest attribution fix in the whole stack.

## Frequently asked questions

### How does Meta attribute an in-store purchase to an ad?

Send the in-store purchase to Meta as an offline event, with the customer matched by hashed email or phone and the original click id where you captured it. Meta matches the event to the ad click within a 28-day click window and a 1-day view window. Events sent outside those windows still appear in reporting as general offline events, but they will not be attributed to a specific ad.

### Does the Meta Pixel track in-store purchases?

No. A POS checkout has no browser, no \_fbp cookie, and no browser session, so the Pixel never fires. The sale is invisible to Meta unless the register data is sent back through the Conversions API or an offline event set. That is the gap, and it is structural, not a bug.

### What customer data does Meta need to match an offline sale?

Email and phone, hashed with SHA-256 before upload. Without an email or phone at checkout, matching is weak, and the event is unlikely to be attributed. Collecting the identifier at the register is the whole setup, and it is also the part most stores skip.

### How long before I know the offline setup is working?

Up to 72 hours to judge match quality. Check the Event Match Quality score, the upload timing, the transaction count, and the revenue against Shopify POS. Delayed uploads hurt match quality, so a regular automated sync beats a sporadic backfill.
