These guides explain what MonitorAI check for, what each result means, and the rules used to decide pass, warning, or error. They are written for merchants, customer success, and operations — not for engineers.
How to read audit results
| What you see |
What it means |
| Error |
Something likely hurts data quality, attribution, or reporting — prioritise fixing. |
| Warning |
Worth investigating; may be expected for some stores. |
| Info |
FYI only; no urgent action. |
| Success |
No issue detected for this rule. |
| Unknown / not enough data |
Too little traffic, new property, missing connection, or data not ready yet. |
Start here — value and ROI
How Littledata quantifies its own worth to a store, and where that copy ends up:
- Value banner glossary — Plain-English dictionary of every dollar and percentage figure in the value banner, with examples. The fastest way to understand what a customer is looking at.
- Value metrics — how they’re calculated — Full mechanics: comparison windows, formulas, suppression rules, edge cases.
- Customer emails (Intercom) — Monthly ROI and trial “days to value” emails: company fields,
Monitor: <destination> <check> audit-check company tags for audience filtering, trigger events, template tips, troubleshooting without engineering tools.
What each audit means (for merchants)
The day-to-day audit results, by channel:
- Google Analytics 4 (GA4) — Traffic and orders, Shopify vs GA4 order counts, links to Google Ads and BigQuery, custom dimensions, and more.
- Meta (Facebook) pixel & Conversions API — Event Match Quality (EMQ) by funnel step, Events Manager / dataset diagnostics, benchmarks, multiple pixels on one ad account.
- Klaviyo flows — Littledata vs native abandonment triggers, competitor flows, exclusions, revenue and attribution windows, and spam profiles injected into the list.
- Shopify store — bot checkouts creating invented shoppers, Markets, Amazon traffic.
Agency API
- Agency API — Per-user read-only tokens for pulling agency dashboard data into your own tools; each token sees the same stores as the user who created it. Try-it-out browser at /api; AI agents start at llms.txt or the raw OpenAPI spec.
For other Littledata apps
- Customer API — Load value messaging and open audit findings (info / warning / error) for a store by hostname, for use in other Littledata products (server-side only; shared key).
- CIP contract — The customer-success endpoints behind cip.littledata.io, for Littledata admins and the agents acting for them (admin token;
write scope for updates). Machine-readable: openapi-cip.yaml.