Most "setup guides" stop at installing the Pixel. That gets you tracking, but it does not get you trustworthy tracking — and it never touches the question that actually pays your rent: which of these ad-reported sales made money. This checklist covers the full stack, from plumbing to profit.
The 8-step Facebook Ads conversion tracking setup checklist
1. Connect the Facebook & Instagram sales channel
In Shopify admin, go to Settings → Apps and sales channels → Facebook & Instagram. Log in and connect your Business Manager, ad account, Page, and Instagram profile.
This is the foundation. If the wrong ad account or Business Manager is linked here, every downstream event routes to the wrong place and nothing below will save you.
2. Turn on Pixel + Conversions API (set data sharing to "Maximized")
Under Data sharing settings, choose Maximized. This enables both the browser-side Meta Pixel and the server-side Conversions API (CAPI) at once.
You want both on purpose. The Pixel fires from the shopper's browser and gets blocked by ad blockers, Safari/Firefox tracking prevention, and closed tabs. CAPI sends the same event from Shopify's server, recovering the ones the browser drops. Field estimates put ad-blocker- and consent-affected traffic at roughly a tenth to a quarter of users, according to Audiense/Elevar — that is how many purchase signals a Pixel-only setup can silently lose.
3. Set one shared deduplication key
This is the step nearly every checklist skips, and it is the one that corrupts your data fastest. When the Pixel and CAPI both send a Purchase for the same order without a shared key, Meta counts it twice.
Meta deduplicates on a matching event_id plus event_name (primary) or falls back to the fbp cookie, and only within a 48-hour window, per Meta's developer docs. Shopify's native integration handles this when both channels run through the official app — so do not bolt on a second Pixel via a theme snippet or a third-party app. A store showing Meta purchases at roughly double its Shopify orders almost always has a dedup misconfiguration, not real demand.
One more matching rule: hash PII (email, phone) with SHA-256, but never hash fbp or fbc — hashing those breaks Event Match Quality and your matching quietly degrades.
4. Add UTM parameters to every ad
Meta auto-appends fbclid, but that only helps Meta. To make Shopify's and GA4's channel reports readable, tag your ad URLs with utm_source, utm_medium, and utm_campaign by hand.
Without UTMs, paid-social traffic scatters into "direct" and "other" in your store-side reports, and you lose the ability to cross-check Meta's self-credited numbers against anything.
5. Confirm your attribution window
Check your attribution setting at the ad-set level. The current default is 7-day click plus 1-day view, per Jon Loomer and Foreplay. The 28-day windows were removed after iOS 14.
Know what this does before you touch it. The 1-day view half credits Meta for sales where the shopper only saw an ad and never clicked — Shopify has no concept of a view, so those never appear as Facebook sales in your store. This is the single biggest reason Meta out-reports Shopify. Tightening to 1-day-click can cut reported conversions by around forty percent for the same real sales, per TrackBee — narrower credit window, not fewer orders.
6. Run test purchases and verify
Place a real test order (you can refund it after). Then confirm the event landed:
- Meta Pixel Helper (Chrome extension) shows which events fire on each page.
- Meta Events Manager should show the Purchase arriving from both the Pixel and the Server, deduplicated into one, with a high Event Match Quality score.
- The revenue value should match what you intend to pass (subtotal vs. subtotal-plus-shipping-plus-tax — pick one and be consistent).
7. Set your maintenance cadence
Tracking rots. Theme updates, app conflicts, and Meta changes all break events silently. Put these on a recurring calendar:
- Weekly: compare Meta-reported purchases against Shopify orders for a trailing 7–14 days (never single days — see below).
- Monthly: review event-parameter completeness in Events Manager.
- Quarterly: audit UTM consistency across live campaigns.
8. Reconcile against real profit
Here is where the standard checklist ends and the useful part begins. Getting the Pixel to fire tells you a sale happened. It tells you nothing about whether that sale made money after product cost, print fees, shipping, and payment processing. For print-on-demand and dropshipping margins, an order can track perfectly and still lose you cash.
If your Facebook conversions look wrong in a way tracking fixes cannot explain, our guide to Facebook Ads showing zero conversions but getting sales walks the diagnostic tree. The same class of problem hits other channels too — see Google Ads not tracking Shopify purchases and GA4 not tracking Shopify purchases.
Why your numbers still won't match — and why that's fine
Even with a flawless setup, Meta and Shopify will report different purchase counts. CAPI recovers lost events; it does nothing about the structural reasons the two systems measure different things:
- View-through — Meta credits views; Shopify credits only clicks that led to checkout.
- Modeling — Meta estimates iOS opt-out and blocked conversions; Shopify only reports real, completed orders.
- Click-date reporting — Meta files a conversion on the ad-click date, not the purchase date, so daily numbers desynchronize even when totals agree. Always compare trailing windows.
- Last-click vs. window — Shopify's default is last non-direct click; Meta claims anything inside its window.
A 20–35% gap between Meta-reported purchases and Shopify orders on the default window is normal, per Vaizle and TrackBee. Chasing exact equality is a fool's errand. For the full framework on which number to trust for what, start with the cluster hub on reconciling your ecommerce data.
Worked example: one week, four different "sales" numbers
Say you sell mugs and run Meta ads for a week. Assume a hypothetical average order of $40 subtotal plus $5 shipping plus $4 tax, so $49 total per order, and 100 real orders — 55 clicked a Meta ad, 15 only saw one, 10 came from Google, 20 from organic or direct, and 8 later refunded. Here is what each system shows you (the numbers are illustrative; the arithmetic is exact).
Meta Ads Manager: ~78 purchases. It counts 55 click-through + 15 view-through = 70, adds roughly 8 modeled conversions, reports them on the click date, and does not subtract the 8 refunds. At subtotal-only revenue: 78 × $40 = $3,120.
Shopify Analytics: 100 orders. Last-click attribution credits ~55 to Facebook, ~10 to Google, and files the 15 view-through buyers under their real last referrer — not Facebook. Total sales before refunds: 100 × $49 = $4,900; after 8 refunds of $49, about $4,508.
GA4: ~72 purchases, split fractionally under data-driven attribution, so no single channel matches Shopify's last-click view.
Your bank payout: less than all of them. The deposit is captured charges minus fees, refunds, and disputes. On a Basic plan, US cards, processing runs about 2.9% + 30¢ per transaction, per Webgility:
- Captured: 100 × $49 = $4,900.00
- Processing fees: (2.9% × $4,900) + (100 × $0.30) = $142.10 + $30.00 = −$172.10
- Refunds issued: 8 × $49 = −$392.00
- One chargeback fee: −$15.00
- Net deposited: $4,320.90
Four numbers — 78, 100, 72, and $4,320.90 — for one week of 100 real orders. None is wrong. Shopify's order count is the truth for how many sales happened; Meta's is the truth for how many its ads plausibly influenced; the payout is the truth for cash in the bank.
Where PodVector fits
PodVector connects your Shopify, Meta Ads, Google Ads, Printify, and Printful accounts into one live data warehouse and computes true per-order profit — the number none of the four dashboards above shows you. It is not a dashboard; it reads your ad data and your real costs together so "78 conversions" becomes "here is what each order actually earned after print cost, shipping, and fees."
Victor, PodVector's AI operator, analyzes that reconciled data and proposes moves, executing approved changes on the Shopify side — he reads your ad numbers but does not touch your ad account. Connect your stores and see per-order profit.
FAQs
Do I need both the Meta Pixel and the Conversions API, or just one?
Both. The Pixel captures rich browser signals but gets blocked; CAPI sends server-side events that survive blockers and closed tabs. Running both with proper deduplication recovers lost purchases without double-counting. Meta itself recommends the redundant setup.
Will setting up CAPI make my Facebook and Shopify numbers match?
No. CAPI recovers events lost to tracking gaps, but it cannot close methodology gaps like view-through credit, modeled conversions, and click-date reporting. Expect a structural gap of roughly 20–35% on the default window even with perfect tracking, per Vaizle. Aim for a stable ratio, not equality.
My Meta purchases suddenly doubled after I added CAPI — is that good?
No — that is almost always a deduplication failure, not extra sales. A correct redundant setup keeps your count stable while recovering blocked events; it should not inflate it. Check that a single shared event_id and event_name are sent on both copies and that you are not running a second Pixel from a theme snippet.
Why does my Shopify payout not match my reported sales?
Because a payout is a batch of balance transactions — captured charges minus processing fees, refunds, and chargebacks — not a day's orders, and it excludes third-party gateways like PayPal entirely. On a Basic plan, US cards, fees run about 2.9% + 30¢ per transaction, per Webgility. "Gross sales minus fees equals payout" is the classic reconciliation error.
How often should I check my conversion tracking after setup?
Weekly, at minimum. Compare Meta-reported purchases against Shopify orders on a trailing 7–14 day window (never a single day, because Meta reports on click date). Review event completeness monthly and audit UTMs quarterly. Investigate immediately if the gap jumps outside its normal range.
Should I change my attribution window to make numbers look better?
No — change it to match how you actually think about credit, not to flatter a report. Tightening to 1-day-click can drop reported conversions by around forty percent for identical real sales, per TrackBee. The window changes the credit story, not the orders in your bank.