PixelProof Roadmap

PixelProof exists because Shopify's Aug 26 2026 checkout-extensibility upgrade removes the Additional Scripts box that thousands of merchants have relied on to fire Meta pixels, GA4 tags, and Google Tag Manager containers — and because Meta's own API deprecation wave (47 Commerce endpoints blocked Jul 29 2026, Graph API v20 and Marketing API v24 sunsetting Sep 24 and Oct 6) is landing on a completely independent timeline. This page tracks, honestly and in order, what has actually shipped toward solving that problem, what's actively being built, what's planned next, and what's still just an idea. It is generated from a source file any future engineer or Andy himself can edit directly (`content/roadmap/meta-monitor/roadmap.mdx`), then rebuilt with `node scripts/build-roadmap.mjs` — nothing here is aspirational marketing copy dressed up as a changelog. If a line is under "Shipped," it traces to a real commit in this repository's own history, most of which are also visible in more granular form on the PixelProof changelog.

• Aug 5 — Core scan engine + landing page live at /meta-monitor (commit `c936601`+): free Shopify storefront scan for Meta pixel, GA4, and GTM tracking gaps ahead of the Aug 26 2026 Shopify checkout upgrade.
• Aug 6–8 — Rewardful affiliate program wiring, dunning sequence, lifecycle emails, and the incumbent price-watch monitor (Elevar / Northbeam / SegmentStream tracked weekly).
• Aug 8 — Security hardening pass (Hostile-Agent Audit H1–H8/H10/H11): CSRF origin allowlist on every state-changing POST, sanitized error responses, typed request-body guards.
• Aug 9 — `/products` portfolio page tying PixelProof to EntryProof and TariffWatch; `empire.customers` shared cross-pick identity table; portfolio footer strip cross-linking all three picks.
• Aug 9 — First-60-seconds pass: welcome page with cross-sell/share/confetti, self-serve account deletion, deferred Sentry init off the LCP-critical path.
• Aug 13 — Free Meta Pixel Debugger tool (no signup required), public `/meta-monitor/status` status page backed by hourly uptime snapshots, and `/meta-monitor/changelog` with RSS.
• Aug 13 — Cookie consent banner wired to actually gate PostHog session replay and Sentry client-side error capture behind the visitor's accept/decline choice, with an EU-default-to-declined branch.
• Aug 14 — Landing A/B testing framework: the hero headline now runs a real 3-variant experiment (control / urgency / outcome framing) via Vercel Edge Config, with sticky per-visitor bucketing and exposure logging to `public.ab_experiment_exposures`.
• Aug 14 — Daily data-retention enforcement cron: free-scan and lead-capture PII older than the 30 days promised on `/meta-monitor/privacy` is now automatically anonymized, not just promised.

• Aug 21–24 — Customer-facing webhook integrations (Slack, Microsoft Teams, PagerDuty, Discord, and generic HMAC-ready) with per-severity filter, per-event filter, 10-fire/hour rate limit, and paste-ready setup guides in the help center covering all five channels.

• M2 revenue gate check (`/api/cron/pixelproof-m2-gate-check`, daily 1 PM UTC): tracking real MRR against the M2 milestone threshold defined in `PICK_1_PRODUCT_BUILD_PLAN.md` before unlocking the next pricing tier's paid features.
• Session Replay Verifier: weekly PostHog recording review (`/api/cron/session-replay-verify`) flags UX-blocking issues found in real user sessions — currently building out the reviewer queue UI at scale.
• GSC verification: no pick has a formally verified Google Search Console property yet (tracked honestly in the `content-refresh-audit` and `sitemap-indexing-check` crons' own code comments) — refresh/sunset triggers run in "unconfirmed, age-only" mode until this lands.

• Named credentialed reviewer for any page making substantive regulatory claims, per Google Perfection Standard v2 §16.3 — currently an open, tracked gap (org byline only, no fabricated review claims) pending an Andy cost decision.
• Chrome extension continuous-monitoring mode (referenced in the free-scan FAQ as an existing capability path) getting its own public roadmap entry once the packaging work is scheduled.
• Expanded incumbent comparison pages beyond the current five (Elevar, Rockerbox, SupaEasy, Vibhora, WeltPixel) as new competitors enter the Shopify tracking-monitor space.

• A public API for agencies managing tracking health across multiple client stores from one dashboard.
• Auto-resolve for PagerDuty incidents on subsequent healthy scans (today customers manually resolve open incidents in the PagerDuty console).

Every entry above is either a direct citation of a git commit in this repository, a currently-running scheduled job (verifiable in `vercel.json`'s `crons` array), or a plan explicitly labeled "Planned" or "Considering" so it can never be mistaken for something already live. There is no separate "roadmap team" maintaining a different source of truth than the code — the person or agent who ships a feature is the same one responsible for adding its line here, in the same pull request where practical. When that discipline slips (a shipped feature not yet reflected here), treat this page as lagging reality by at most a few days, not as authoritative marketing.

This page itself is not exempt from the honesty standard it describes: nothing under "Shipped" links to a feature that doesn't exist in production today, and nothing under "Planned" or "Considering" is described with more certainty than it deserves. If a competitor or a curious prospect is reading this specifically to gauge how seriously we take this product, that scrutiny is the point — a roadmap that only ever says "everything is great, more is coming" is not a roadmap, it's marketing copy wearing a roadmap's clothes, and this one is deliberately built not to be that.

*This roadmap is generated from `content/roadmap/meta-monitor/roadmap.mdx` and this repository's real commit history — Andy or any future agent can edit the source file directly and re-run `node scripts/build-roadmap.mjs`. Nothing on this page is aspirational marketing copy; every "Shipped" line traces to an actual commit.*