Skip to main content
Part of Digital Empire
Methodology

How PixelProof detects broken tracking — and what it does not do

By Andy Gaber, Founder · Published August 20, 2026 · Last updated September 11, 2026

PixelProof helps Shopify teams detect broken Meta Pixel and GA4 tracking after theme, checkout, consent, or campaign changes — free storefront scan first, then paid monitoring. Detect and monitor come first. Fix snippets on paid plans are guidance you apply, not a headline “concrete fix path,” and not a magic auto-install.

1. What PixelProof is

Most “pixel tools” sell implementation, attribution dashboards, or a Shopify app install. PixelProof sells a narrower job: tell you whether the tracking loaders that should be on your public storefront are still there after something changed, then keep watching. Shopify already retired Additional Scripts and legacy checkout.liquid tracking for non-Plus stores on August 26, 2026. Theme edits, consent apps, campaign tags, and GTM publishes still break pixels after that cutover — usually without a storefront error.

We are not listed on the Shopify App Store, not Built for Shopify, and not a Meta or Shopify partner program. We do not publish fake install or review counts. Support is support@pixeluptime.com only.

2. What the free storefront scan checks

You paste a store URL. Our server fetches that URL's public HTML (the same document a logged-out shopper's first request can see) and runs a deterministic regex ruleset against it. Same HTML in, same findings out. No LLM is called. The current storefront rules look for:

  • Meta Pixel presence — fbq('init'), the fbevents.js loader, and the standard facebook.com/tr noscript fallback. When the ID is in the HTML, we capture the 15–16 digit pixel ID.
  • GA4 presence — gtag('config', 'G-…'), gtag/js?id=G-, and quoted measurement IDs, including some proxied hosts.
  • GTM container presence — gtm.js loaders and GTM- container IDs, including IDs that only appear as query params on a first-party proxy.
  • Other storefront-visible loaders when they appear: TikTok Pixel, Snapchat Pixel, Pinterest tag, and common affiliate scripts (Rewardful, Refersion, ShareASale, Impact).
  • Shopify Web Pixels / Customer Events wiring markers — signals that the store loads pixels through Shopify's sandboxed manager even when individual fbq snippets are not in the raw HTML.
  • Duplicate Meta-family loaders on the same page (a heuristic: more than one ad-platform pixel family on one response, which can inflate attribution).
  • An always-on checkout limitation finding on Shopify storefronts: Additional Scripts, Thank You, and Order Status pages are not visible from this fetch. That finding is labeled as a scan limit, not as a confirmed broken pixel.

If Shopify is detected but no storefront pixel/analytics markers appear at all, we add a separate headless / client-hydrated warning. Many Hydrogen and SPA themes only attach pixels after JavaScript hydrates — a server HTML fetch will honestly say it saw nothing.

3. What the free scan does not cover

  • Does not authenticate to Shopify admin, Meta Events Manager, or a GA4 property.
  • Does not open checkout, place a test order, or read customer records.
  • Does not prove a Purchase, AddToCart, or InitiateCheckout event fired — only that a loader signature was (or was not) present in that HTML.
  • Does not inspect consent-mode configuration as a legal audit.
  • Does not write to your theme, install an app, or push pixels via the Web Pixels API.
  • Does not call Meta's Graph API to confirm last-event timestamps. If an older page on this site implied that, it was ahead of the live scanner. The live scan is HTML presence plus the honest checkout-unknown finding.
  • Does not crawl four extra collection pages from sitemap.xml. One requested URL, one HTML body, 8-second timeout, 2 MB cap.

A green-looking score means the loaders we know how to see look present on that page. Only a real test order (or Meta Test Events / the optional Chrome extension inside your own session) confirms the events fire end-to-end. We say that on the free-scan landing too.

4. How a scan actually runs

  1. Normalize and fetch. The URL is forced to http(s), private/loopback/link-local hosts are rejected (SSRF hardening, DNS pinned per hop), then we GET the document from our server — never from your customers' browsers.
  2. Grep public loader patterns. scanHtml() in the rules engine applies the storefront rule list above and records captured IDs. Findings are reproducible.
  3. Label what we cannot see. Checkout / Additional Scripts stay canVerifyFromStorefront: false. Absence of evidence is not scored as a critical break.

The same fetchStorefrontHtml() + scanHtml() pair is what the landing scanner, /meta-monitor/free-scan, and the paid Monitor cron use. Paid does not secretly switch to a different, deeper crawler.

5. What Monitor watches over time

A free scan is one-shot. You leave with a result. Paid Monitor stores the snapshot and re-runs that same storefront scan on a recurring schedule for subscribed stores, then emails when findings change (theme deploy, consent app, GTM publish, campaign tag). Marketing copy calls this a weekly monitor; the important contract is recurring + alert, not a promise that we watched checkout or Ads Manager overnight.

Monitor-Only is $29/mo for 1 store: alerts, no snippets, no PDF.Starter is $99/mo for 3 stores and adds paste-ready snippets for detectable storefront findings. Agency is $149/mo for up to 15 stores, then $8/store/mo overage. Full grid: /meta-monitor/pricing. The soft-paywall CTA deep-links Monitor-Only checkout at /meta-monitor/pricing?checkout=monitor_only&utm_source=site&utm_medium=methodology&utm_campaign=pp_monitor&utm_content=pp_methodology.

6. What “fix snippets” mean

Starter and Agency can attach paste-ready guidance from a build-time snippet library (public Shopify / Meta / Google patterns, acceptance criteria included). You copy, paste, and verify. We do not push the snippet into your theme, we do not create a Custom Pixel for you, and Monitor-Only does not include snippets at all. A snippet that maps to a storefront-visible loader is still not a fix for checkout surfaces this scan cannot see. That is guidance, not an auto-install, and not the headline claim of the product.

7. Soft paywall and fail-open

Free scans are budgeted in this browser's localStorage (3 scans, +2 if you leave an email, then a ~24h cooldown). When the budget is used, we ask you to start Monitor-Only or come back later. We do not hard-lock the product, invent a countdown of remaining seats, or call the next scan “premium locked.” If storage is missing (private mode), the gate fails open so you can still scan. The standalone free-scan page also uses a short-lived session cookie (one scan per hour) as a separate abuse control — still not a paywall brick.

8. What we will not claim on this page

  • No clearance, compliance certification, or guaranteed ROAS recovery.
  • No Shopify App Store / Built for Shopify / official partner status.
  • No white-label reseller program and no fabricated review counts.
  • No “we walked your checkout like a shopper” for the free HTML scan.

Privacy basics: we store the URL you scanned and the loader findings; email if you give it; Stripe if you subscribe. We do not scrape shopper PII off the storefront. Privacy · Cookies · FAQ · Terms.

Run the free storefront scanStart Monitor-Only — $29/mo

PixelProof home · Monitor-Only · FAQ · Agency workflow · Docs · Glossary

Cookie settings