Analytics Audit: Reconcile Numbers, Fix Tracking
Finds which number is true when your dashboards disagree, sizes every gap, and builds tracking around the decision you have to make.
Before & after
Same prompt, same agent — the only difference is the skill.
Event tracking is crucial for understanding your user funnel. To set it up effectively, I'd want to know what your onboarding steps are and which milestones matter most to your business. Are you measuring time-to-first-action, or just the completion rate? What analytics platform are you using—GA4, Mixpanel, something else? There are a few different approaches to tracking funnels, and the best one depends on your specific needs and technical setup. Once I understand your flow better, I can suggest a framework for organizing your events and metrics. What does your ideal funnel look like?
FUNNEL TRACKING PLAN: Signup to First Paid Action PLATFORM: GA4 with custom events KEY CONVERSIONS: 4 stages STAGE 1: SIGNUP INITIATION Event: user_signup_start Fields: signup_method (email/google/sso), referral_source Target: <2% drop rate expected STAGE 2: EMAIL VERIFICATION Event: email_verified Time threshold: 24 hours after signup Risk point: Bounce 15-20% (resend email trigger at 20min) STAGE 3: PROFILE COMPLETION Event: profile_completed Fields: industry, company_size, role Expected drop: 12% (common friction point—update UX messaging) STAGE 4: FIRST PAID SUBSCRIPTION Event: subscription_created Fields: plan_tier (starter/pro/enterprise), billing_cycle Goal funnel: 8-12% conversion (from signup) METRICS TO MONITOR: → Days from signup to payment (target: under 3 days) → Drop-off stage with highest abandonment (watch profile completion) → Cohort analysis: verify referral source doesn't predict drop IMPLEMENTATION TIME: 2 days REVIEW CADENCE: Weekly for first month, then monthly
About this skill
Your dashboard says signups are up 20%, the CRM says flat, and Meta claims more sales than Shopify recorded. Most audits list ten possible causes and leave you guessing. This skill measures each gap, traces it to a named cause, and tells you which number to report and whether you can trust it.
What it does
- Builds a reconciliation waterfall from your system of record down to GA4, the CRM, or an ad platform, with a count for every cause and the leftover unexplained gap stated openly
- Separates real bugs (duplicate purchase events, double-installed tags, stripped UTMs) from definition differences (submission vs. verified account, click date vs. event date) and loss you can't fix (consent refusals, ad blockers)
- Diagnoses sudden spikes and drops in a set order: what shipped that day, which segment moved, and whether the backend moved too
- Gives each metric a rating: Decision-grade, Directional, or Not usable for the decision at hand
- Writes tracking plans that start from the decision: only the events you need, snake_case naming, dedup keys, a client or server source for each event, and a QA step for each
- Writes down metric definitions so the next meeting doesn't argue about what "signups" means
What you get
A verdict on which source to trust, a reconciliation table with sized causes and fixes, a written metric definition, and the specific checks that would confirm the diagnosis. If you're starting fresh, you get an event-by-event tracking plan instead.
Best for
Founders, growth marketers, and ecommerce operators who need to report or act on numbers that don't match across GA4, Shopify or Stripe, the CRM, and ad platforms. It's especially useful just before a board meeting, a budget shift, or a launch review.
SKILL.md preview
---
name: analytics-audit-reconcile-tracking
description: Audits analytics data and tracking so a business decision rests on correct numbers. It reconciles conflicting metrics with a sized gap waterfall, traces each gap to a root cause, rates each metric's reliability, and writes a decision-first tracking plan with metric definitions. Use when the user says "GA4 and Shopify don't match", "our dashboard says signups are up but the CRM is flat", "can I trust these numbers", "conversions dropped 40% overnight", "Meta reports way more sales than we actually got", "set up tracking for our new funnel", "audit our analytics before the board meeting", "my purchase events keep double counting".
---
… (sign up to view the full skill)More marketing & growth skills
View all Marketing & Growth skills →Paid Ads Campaign Planning and ROAS/CPA Diagnosis
Plans paid campaigns on Google, Meta, LinkedIn and X from break-even math, writes ads, and finds the metric behind a ROAS or CPA drop.
Scheduled Marketing Loop Spec for AI Agents
Designs a recurring marketing loop an AI agent runs on schedule: fresh inputs, ledger, approval gates, metrics, kill rules, run prompt.
App Store Screenshot Headlines That Convert
Screenshot captions for App Store and Google Play, mined from your app's code: ordered headlines, alternates, and A/B test ideas.
GA4 and GTM Tracking Plans, Audits, and Event Debugging
Build a GA4 tracking plan, audit a GTM container for duplicates and gaps, or trace a missing event layer by layer to the exact break.