Find the growth problem. Keep the evidence. Measure the next action.
Run the Growth Fix Queue, AI Referral-to-Conversion Tracker, Distribution Proof, Crawler Evidence Card, Buyer Question Source Map and Signup Handoff Verifier in one reviewable workflow.
Local-analysis pilot: no account connection, network submission or automatic deployment. Imported numbers are unverified until reconciled to their source. Empty or missing metrics are never treated as zero.
Import aggregated outcomes
Export a GA4 acquisition/landing-page report or prepare the same columns in a spreadsheet. Use one row per site, source, medium, campaign and page. GA4 key events should be mapped to conversions only if they represent the outcome you intend to measure. Do not paste identifiers or individual sessions.
Sites represented
—
Reported sessions
—
Reported conversions
—
Metric definitions and validation
A referrer classification is not proof that an AI answer cited the site. Sessions in an exported report are not unique people. Summing conversions is meaningful only if the event definition is consistent. A valid signup conversion rate needs both a comparable visit/session denominator and verified signup events. Reconcile this worksheet to the original GA4 property, date range and event settings before publishing claims.
Growth Fix Queue
Five deterministic checks across imported sites: low conversions, weak engagement, missing outcome measurement, AI referral handoff gaps, and campaign attribution gaps. These are diagnostic flags, not AI predictions or proven causes.
Import report data to generate ranked flags. For validation, use three sites and manually classify each flag as useful, false positive or not enough evidence.
AI Referral-to-Conversion Tracker
Roll up recognizable AI referral domains by landing page, sessions, engagement and measured conversions. Google AI Overviews and AI Mode often cannot be split reliably from Google Search traffic using referrer alone. Unclassified referrals stay unclassified.
Import a report containing source, landing page, sessions and conversions.
Distribution Proof
Compare posts and channels using the campaign/source/medium dimensions you already tag in Campaign Studio. A campaign name should identify an individual post or creative where that granularity is needed.
Import dated campaign rows to compare two channels over a bounded 14-day period. Incomplete dates or attribution will be flagged.
Decision rule: continue a channel only with adequate qualified outcomes and effort context; otherwise adjust the message and tagging or stop the experiment. Do not infer causation from last-click labels alone.
Crawler Evidence Card
Paste an already sanitized request-log aggregate. Required: user_agent, request_path, status, count. Optional: observed_at. Do not include IP addresses, query parameters, cookies or tokens. User-agent strings are self-declared and cannot prove a verified crawler identity.
No request evidence imported.
Buyer Question Source Map
Record a real buyer question, the answer page, a primary source and the date it was checked. This map documents research opportunities; it does not claim that an AI engine cited anything.
No questions recorded. Prompt selection and buyer intent remain estimates without independent research.
Signup Handoff Verifier
Read-only checklist. It never submits a form or enrolls an address. End-to-end delivery requires a separately authorized test recipient and inbox access.
Run the experiments without overclaiming results
Growth Fix Queue
Three sites, five rules, record accepted flags and false positives. A flag is not a conversion win.
AI referrals
Match one source report to GA4 and verify that actual conversion events exist before reporting a rate.
Distribution Proof
Two channels, 14 calendar days, tagged posts, comparable qualified outcomes.
Suggested research review: 14 days after the first valid import. No testing has been performed on your traffic by this page.