
Funnel Analyst Companion
Analyze conversion funnels and identify optimization opportunities with data-driven insights
What You Can Do
Upload your funnel metrics and this skill will analyze conversion patterns, identify drop-off stages, and generate prioritized experiments to improve your conversion rates. You get actionable recommendations backed by statistical rigor, including A/B test designs, cohort segmentation insights, and impact projections for each optimization.
Features
Automatically identify which funnel stages have the highest abandonment rates and where your biggest conversion leaks occur.
Segment users by acquisition source, device, geography, and behavior to uncover where specific cohorts struggle in your funnel.
Get structured experiment designs with clear hypotheses, success metrics, required sample sizes, and expected lift calculations.
Determine if conversion changes are statistically significant and calculate confidence intervals for experiment results.
Receive recommendations for how to present your funnel data visually to communicate insights to stakeholders.
Get AI-generated hypotheses about why users drop off at each stage, based on common UX and behavioral patterns.
Rank optimization opportunities by potential impact (expected uplift × effort) to focus your team on high-ROI experiments.
Example Output
Funnel Analysis Summary
- 📊 Drop-off Analysis:
- Landing → Signup: 45% conversion (acceptable)
- Signup → Trial: 62% conversion ⚠️ (below benchmark)
- Trial → Paid: 28% conversion 🔴 (critical)
- Paid → Retained (30d): 71% conversion (good)
Top 3 Optimization Opportunities:
-
Trial Onboarding Flow (Expected: +8-12% conversion)
- Hypothesis: Users abandon because onboarding is too complex
- Experiment: Simplified 3-step tutorial vs. current 6-step flow
- Sample size needed: 2,400 per variant (80% power)
-
Pricing Page Transparency (Expected: +5-7% conversion)
- Hypothesis: Users leave due to unclear value proposition
- Experiment: Add feature comparison table and social proof
- Sample size needed: 1,800 per variant
-
Checkout Friction (Expected: +3-5% conversion)
- Hypothesis: Multi-step checkout creates abandonment
- Experiment: One-page checkout vs. current multi-page
- Sample size needed: 1,200 per variant
What's Included
- Funnel Audit Framework: A structured methodology for analyzing your funnel data, identifying anomalies, and benchmarking against industry standards.
- A/B Test Design Templates: Ready-to-use templates for hypothesis statements, success metrics, statistical power calculations, and experiment roadmaps.
- Cohort Analysis Toolkit: Techniques for segmenting users by multiple dimensions and identifying which cohorts contribute most to conversion problems.
- Prioritization Scorecard: A framework for ranking experiments by expected impact, implementation effort, and strategic alignment.
- Statistical Analysis Guides: Explanations of confidence intervals, statistical significance, p-values, and how to interpret experiment results correctly.
Who It's For
- Product Managers
- Growth Marketers
- Data Analysts & Analytics Engineers
- E-commerce & SaaS Product Leads
- Conversion Rate Optimization Specialists
Best For
- Identifying conversion bottlenecks in multi-stage funnels
- Designing statistically rigorous A/B and multivariate tests
- Prioritizing feature experiments by expected ROI
- Segmenting users to understand cohort-specific behavior
- Interpreting experiment results and generating actionable insights







