Przejdź do treści
Biblioteka skilli
Dane i analityka · Analiza

funnel-analysis

/funnel-analysis

Analizuje lejek konwersji: wskazuje miejsca odpadania, liczy konwersję etap po etapie, formułuje hipotezy przecieków i proponuje eksperymenty poprawiające wyniki.

Repozytorium
pm-skills ↗
Licencja
MIT
Ostatnia zmiana repozytorium
23.08.2026

Ocena i bezpieczeństwo

100/ 100Klasa: dobry
Format25 / 25
Frontmatter z nazwą i opisem, poprawna nazwa, opis do 1024 znaków, nazwa zgodna z folderem.
Odkrywalność25 / 25
Opis mówi, kiedy użyć skilla, ma sensowną długość i jest w 3. osobie.
Treść35 / 35
Co najmniej 150 słów i do 500 linii, nagłówki, listy kroków, przykłady albo kod.
Higiena15 / 15
Bez zaślepek typu TODO, bez ścieżek Windows, z licencją.
Ryzykobrak sygnałów w skanie
Pliki w paczce
1
Skrypty
0
Komentarze HTML
0
Adresy URL w SKILL.md
0
Przestarzałość

Brak sygnałów. Repozytorium zmienione 23.08.2026, bez odwołań do wycofanych modeli i API.

Licencja

MIT, otwarta: można używać i zmieniać, z zachowaniem informacji o autorze.

Co robi i kiedy

Analizuje lejek konwersji: wskazuje miejsca odpadania, liczy konwersję etap po etapie, formułuje hipotezy przecieków i proponuje eksperymenty poprawiające wyniki.

Oryginalny opis (EN)

Analyze a conversion funnel — identify drop-off points, calculate stage-by-stage conversion rates, generate leakage hypotheses, and recommend improvement experiments. Use when diagnosing funnel performance, prioritizing optimization work, or designing experiments to improve conversion.

Jak działa

  1. Purpose
  2. Domain Context
  3. Instructions
  4. Notes
Sekcje z nagłówków SKILL.md, w oryginale · 123 linii, 881 słów
Podgląd SKILL.md
---
name: funnel-analysis
description: "Analyze a conversion funnel — identify drop-off points, calculate stage-by-stage conversion rates, generate leakage hypotheses, and recommend improvement experiments. Use when diagnosing funnel performance, prioritizing optimization work, or designing experiments to improve conversion."
---

# Funnel Analysis

### What You'll Need
| Input | Required? | Example |
|-------|-----------|---------|
| Funnel stage names and conversion data | ✅ Required | Sign-up → Activation → First Purchase (1000 → 420 → 180 users) |
| Time period for the analysis | ✅ Required | Q4 2025, last 30 days |
| Product / feature being analyzed | ✅ Required | Onboarding flow for B2B SaaS |
| Historical baseline or benchmark | 🟡 Recommended | Industry average activation rate ~40% |
| Qualitative user feedback about drop-off points | ⚪ Optional | Session recordings, support tickets |

> **Don't have everything?** Start anyway — the skill will work with what you provide and flag where richer input would improve the output.

## Purpose
Systematically analyze a conversion funnel to find the highest-impact drop-off points, generate hypotheses for why users are leaving, and recommend concrete experiments to improve flow.

## Domain Context
Funnel analysis answers the question: "Where are we losing users, and why?" The most valuable insight isn't just *where* the biggest drop occurs — it's *why* it happens and *what to test next*. This skill combines quantitative analysis (conversion rates, leakage volume) with structured hypothesis generation to turn data into action.

**Key principle:** Fix the biggest leaks first, but consider absolute volume, not just percentage drop. A 5% drop at 100,000 users/month is more impactful than a 30% drop at 1,000 users/month.

## Instructions

You are analyzing a conversion funnel for **$ARGUMENTS**.

Work through each section systematically.

### Phase 0: Context Confirmation

Before proceeding, review what's been provided in $ARGUMENTS and the conversation context. **Only ask about what's missing — don't re-ask what's already clear.**

1. **Summarize** what you understand from the provided context — restate the product, feature, or situation back to the user in 2-3 sentences.
2. **Identify gaps** — if any of the following are unclear, ask:
   - What conversion funnel are we analyzing?
   - What are the funnel stages from entry to completion?
   - What time period should we examine?
3. **Confirm**: _"Here's my understanding: [summary]. I plan to [brief description of what the skill will produce]. Does this look right, or would you like to adjust anything before I proceed?"_

If the user provides additional context, incorporate it. If the user confirms, proceed.

### Step 1: Map the Funnel

For each stage provided, calculate:
- **Users entering**: count or percentage
- **Users exiting**: count or percentage
- **Conversion rate**: users who proceeded / users who entered × 100
- **Drop-off rate**: 100 - conversion rate
- **Drop-off volume**: absolute number of users lost at this stage

Present as a table:

| Stage | Users In | Users Out | Conversion Rate | Drop-off | Drop-off Volume |
|-------|----------|-----------|----------------|----------|----------------|
| [Stage 1] | [N] | [N] | [%] | [%] | [N] |
| ... | | | | | |

**Overall funnel conversion**: [first stage users → last stage users] = [%]

### Step 2: Identify the Biggest Leaks

Rank stages by drop-off volume (not just percentage). Highlight the top 2-3 stages where the most users are lost.

### Step 3: Generate Leakage Hypotheses

For each major drop-off stage, generate 3-5 hypotheses for *why* users are leaving:

**Hypothesis format:** "Users are dropping at [stage] because [reason]. Evidence: [what would confirm this]. Effort: [Low/Medium/High]."

Draw from common funnel failure patterns:
- **Friction**: Too many steps, confusing UI, slow load time
- **Value not clear**: Users don't understand what they get if they continue
- **Trust gap**: Missing social proof, unclear privacy, unfamiliar brand
- **Wrong audience**: Users who arrived don't match the product's target persona
- **Technical failure**: Errors, broken flows, mobile incompatibility
- **Timing mismatch**: Asking for commitment (payment, sign-up) too early

### Step 4: Prioritize Experiments

For the top 2-3 drop-off stages, recommend experiments using an ICE score (Impact × Confidence × Ease):

| Experiment | Stage | Hypothesis Tested | Expected Impact | Confidence | Ease | ICE Score | Recommended? |
|-----------|-------|-------------------|----------------|-----------|------|-----------|-------------|
| [Experiment name] | [Stage] | [Hypothesis #] | [H/M/L] | [H/M/L] | [H/M/L] | [1-10] | [Y/N] |

### Step 5: Generate Output Report

```
## Funnel Analysis: [Product / Flow Name]

**Period**: [date range]
**Overall conversion**: [X]%

### Stage-by-Stage Breakdown
[Table from Step 1]

### Biggest Leaks (ranked by volume)
1. [Stage]: [N] users lost ([%] drop) — [brief reason]
2. [Stage]: [N] users lost ([%] drop) — [brief reason]

### Leakage Hypotheses
[Step 3 hypotheses for top stages]

### Recommended Experiments
[Step 4 table]

### Quick Wins (low effort, moderate+ impact)
- [Experiment]: [Expected lift] — [1-sentence rationale]

### Benchmarks
[Industry benchmarks for this funnel type if available]

### Next Steps
- Run [Experiment 1] first — highest ICE score
- Instrument [stage] to gather qualitative data on drop-off reason
- Review with [team] to validate hypotheses before building
```

## Notes
- Funnel data can be misleading without segmentation — the same funnel often has very different conversion rates by acquisition channel, device type, or user persona
- Always validate hypotheses with qualitative data (session recordings, user interviews) before committing to a fix
- Benchmark against industry norms, but your product's context matters more — a 2% free-to-paid conversion in B2B SaaS is excellent; in consumer apps it may be poor

Wersje

RepozytoriumWynikLicencja
Ta strona · tarunccet/pm-skills100MIT
fabioannovazzi/app_files100AGPL-3.0
SkillMedev/skills100MIT
event4u-app/agent-config90MIT
nimrodfisher/data-analytics-skills90MIT
Amey-Thakur/AI-SKILLS90MIT

Migawka bazy z 13.11.2026 · wersja paczki ce3ea2132764 (skrót SHA-256 całej paczki)

Pobrania w ostatnich 30 dniach: 0 · ranking

Zgłoś problem z tym skillem: szkodliwa treść, błędny opis albo prośba autora o usunięcie.

Podobne skille

  • 100bi-funnel-analysis
    Buduje lejek konwersji ścieżki produktowej, liczy konwersję w każdym kroku, wskazuje miejsca utraty użytkowników i porównuje wyniki z benchmarkami branżowymi.
  • 100funnel-doctor
    Diagnozuje lejek konwersji: znajduje miejsca, w których użytkownicy odpadają, wylicza skalę strat i ustala przyczyny, np. w checkoucie lub rejestracji.
  • 100funnel-analyzer
    Analizuje lejki konwersji i wskazuje miejsca, w których ludzie odpadają, np. w koszyku, rejestracji lub ścieżce użytkownika.
  • 100pm-funnel-analyzer
    Diagnozuje spadki w lejku konwersji, formułuje hipotezy i rekomenduje interwencje z miernikami sukcesu, z wskazówkami o istotności statystycznej i etapami lejka z kontekstu domeny.
  • 95cja-funnel-health-check
    Analizuje lejek konwersji wieloetapowej, aby znaleźć, gdzie użytkownicy odpadają i które kroki tracą najwięcej, wraz ze wskaźnikami konwersji i fallout.
  • 100cro-optimizer
    Pobiera dane analityczne na żywo przez Humblytics MCP, analizuje lejki konwersji, wskazuje miejsca odpadania i generuje priorytetowe hipotezy testów A/B oraz mapy testów.

Uwagi i zgłoszenia

Wiadomość trafi do Bartosza Sokolińskiego. Adres e-mail jest potrzebny tylko do odpowiedzi. Szczegóły w polityce prywatności.