Real-time alerting & observability
- Automated crawls: daily, weekly, and after every deploy.
- Instant alerts to Slack, Microsoft Teams, email, webhooks and PagerDuty.
- Threshold alerts, e.g. tell me if checkout events drop more than 15% hour-over-hour.
AuditTag Guardrails · continuous tracking monitoring
Your tracking works today. What happens after the next deploy?
A theme ships, a checkout is rebuilt, a consent banner changes, a new tag goes in. One of those can silently break your measurement. Guardrails watches the tags that carry your revenue and tells you the moment one stops firing, instead of a mystery dip in next month's numbers.
Start with a free audit See a sample →
Read-only, no write access
Alerts, not surprises
Starts from your free audit
Tag Health
87/ 100
▲ +4 this week · dipped to 71 after a theme update, caught & fixed
Alerts
- Meta Purchase stopped firing on /checkout · caught, fixed
- dataLayer ecommerce.value became a string on 3 templates
- GA4 purchase firing with a valid payload
Monitored checks
- Purchase event firing
- GA4 collect firing
- Product schema present
- Meta CAPI dedup concierge review
CI/CD guardrail
- Deploy #488 passed
- Deploy #482 blocked: purchase missing on staging
A tracking failure doesn't stop your sales. That's what makes it dangerous.
Your campaigns keep running. Your dashboards keep loading. Your team keeps making decisions. But the data underneath may no longer be telling the truth, and nothing on the page looks broken. The gap between the break and the day someone notices is where the money goes: every algorithmic bid that depends on the missing signal degrades for as long as it runs blind.
-
01 · Detect
Know when something changes
Guardrails re-checks your tags on a schedule and after every deploy, reading the same served pages a crawler does, plus the runtime state where your setup allows it.
-
02 · Compare
See exactly what moved
Each run is diffed against the last known-good snapshot: a tag that stopped firing, a
dataLayervalue that changed type, a payload that lost a field. -
03 · Alert
Hear about it the same day
When something revenue-critical drifts, you get an alert with the detail, in Slack, Teams, email or a webhook, not a surprise in next month's numbers.
-
04 · Protect
Catch it before it ships
Where your stack supports it, the same checks run against staging on a pull request and fail the build before a broken tag ever reaches production.
What Guardrails watches
Everything here is scoped to the tracking layer that actually carries your revenue signal. Each capability is labelled with where it honestly stands today.
Developer & CI/CD pipeline guardrails
- Staging integration via GitHub Actions, GitLab CI and Vercel.
- Block a deploy when a critical tag or a dataLayer schema fails validation.
- Headless rendering (Playwright) to audit SPA state changes and full funnels.
Data layer & payload audit engine
- Schema-drift detection: validate
window.dataLayeragainst a defined JSON schema. - Catch the classics: a value that became a string, an empty items array, a missing currency.
- Server-side CAPI validation: cross-check client pixels against Meta CAPI and GA4 Measurement Protocol for clean deduplication.
Governance, privacy & financial intelligence
- Revenue-loss estimates that map a missing trigger to wasted ad spend and unrecorded checkout value.
- Pre-consent leak detection against consent platforms like OneTrust and Cookiebot.
- Vendor discovery: surface fourth-party scripts piggybacking on your tags.
Straight answer: what is built, and what isn't
The free audit and the remediation packs are live and automated today. Run one now and see. Continuous Guardrails is in early access. We onboard your first domains and watch them by hand while the self-serve dashboard and the automated alerting ship, so you get the monitoring outcome immediately and the automation lands over the following weeks. The dashboard above is what the self-serve view is being built to look like, not a screenshot of a finished product. We would rather tell you that than pretend a mature platform exists that doesn't, and we will never sell you a capability that isn't real yet. When something goes wrong, the person who reads your email is the person who wrote the code. Reach me any time at hello@audittag.com.
What does a broken pixel actually cost?
A deliberately conservative estimate from your own numbers. A broken conversion event doesn't just lose reporting. It can degrade the optimisation signals your advertising platforms depend on, for as long as it goes unnoticed.
Ad spend flying blind per break
£0
Efficiency wasted while blind
£0 to £0
An estimate from your own numbers and a conservative 10% to 25% waste band. On-page and tracking fixes improve how reliably your spend is measured and bid on; they are not a traffic machine, and we won't pretend otherwise. Your real result depends on your platforms and how they rebid.
Guard, from £69 a month
Every plan starts from the free audit that's live today. Early access means we onboard your first domains hands-on while the self-serve dashboard ships. Prices in GBP; the full ladder and add-ons are on the homepage.
Starter
For one business
£69/mo
- 3 domains
- 10,000 URLs audited / mo
- Weekly automated scans
- Email alerts
- Core Tag Health
Agency / Growth
For agencies & growing teams
£259/mo
- 25 domains
- 100,000 URLs audited / mo
- Daily scans
- White-label PDF & web reports
- Slack / Teams / GTM integrations
- Consent-leak audits
Enterprise
For large & multi-brand organisations
from £990/mo
- Unlimited domains
- 1M+ URLs / mo, continuous scan loops
- CI/CD staging pipeline blocking
- Server-side CAPI validation
- Multi-tenant RBAC
- Dedicated SLA & account manager
Questions worth asking
Do you need access to my site to monitor it?
Only a read-only foothold: a scheduled crawl of your public pages, or one lightweight beacon you can remove any time. There is no login to hand over, no plugin with write access, and nothing that can change your site. Never write access.
Is Guardrails actually built, or is this a landing page?
The free audit and the remediation packs are live and automated today. Continuous Guardrails is in early access: we onboard your first domains and watch them by hand while the self-serve dashboard and the automated alerting ship. You get the outcome immediately; the automation lands over the following weeks. We would rather tell you that than pretend a dashboard exists that doesn't.
What exactly does it watch for?
The tracking layer that carries your revenue signal: whether your GA4, Meta and Google Ads events fire on the pages that matter, whether your ecommerce dataLayer still matches its schema, and whether a deploy or a theme change has stripped a tag. When something breaks, you get an alert with the payload, not a mystery dip in next month's numbers.
How does the CI/CD guardrail work?
It runs the audit against your staging environment on a pull request and fails the build if a revenue-critical event goes missing before it ever reaches production. It is the same discipline you already apply to code, applied to your tags. We run it on our own repository.
How do the alerts reach me?
Slack, Microsoft Teams, email, a generic webhook or PagerDuty, whichever you use. You set a threshold, for example tell me if checkout events drop more than 15% hour over hour, and you only hear from us when something crosses it.
How do I start?
Run the free audit. It costs nothing, needs no account, and shows you the state of your tracking today. If it is worth watching, we onboard your first domains from there. No sales process.
Don't wait for your reporting to tell you something broke
Start with the free audit. It shows you what's happening with your tracking today, and Guardrails takes it from there.