Funnel Analysis
A method of tracking and visualizing the sequential steps users take toward a conversion goal, identifying where and why users drop off at each stage.
What Is Funnel Analysis?
Funnel analysis tracks users through a sequence of steps toward a conversion goal and measures the drop-off between each step. It mirrors how we intuitively think about processes — as linear sequences with leakage — and makes it easy to see where the largest losses occur. The biggest drop-off is usually the biggest opportunity, assuming you can diagnose its cause.
Also Known As
- Marketing team: "conversion funnel," "marketing funnel"
- Sales team: "pipeline funnel," "sales funnel stages"
- Growth team: "activation funnel," "AARRR funnel"
- Data team: "sequential conversion analysis"
- Finance team: "revenue funnel"
- Product team: "user journey funnel," "onboarding funnel"
How It Works
Your checkout funnel: Homepage (100,000) → Product page (40,000, 60% drop) → Cart (12,000, 70% drop) → Checkout (8,000, 33% drop) → Purchase (3,600, 55% drop). Overall conversion: 3.6%. The biggest absolute loss is homepage-to-product (60,000 users). The biggest relative drop is cart-to-checkout (70%). But the most actionable stage might be checkout-to-purchase (55% drop), where surprise shipping costs often cause the loss. Diagnose the why at each step — different losses have different causes.
Best Practices
- Define each step crisply — "visited page" vs. "engaged with page" produce different funnels.
- Prioritize by expected-lift × leverage (big drop-off × clear hypothesis × testable fix).
- Always check for non-linear paths — real users backtrack, skip, and detour.
- Segment funnels by source, device, and intent to find hidden patterns.
- Pair funnel metrics with qualitative research (session replay, surveys) to explain the why.
Common Mistakes
- Focusing on the biggest absolute drop when the cause is targeting (upstream), not the page itself.
- Assuming every user should follow the linear path when real behavior is branchy.
- Optimizing one step without measuring downstream impact on final conversion.
Industry Context
SaaS and B2B track signup → activation → first value → paid conversion funnels. Ecommerce and DTC track product-view → add-to-cart → checkout → purchase. Lead gen operations track form-view → form-start → form-complete → MQL qualification.
The Behavioral Science Connection
Every funnel step can involve friction, doubt, and distraction. Loss framing may matter near payment, information load may affect complex steps, and prior effort may either increase commitment or create abandonment. Funnel data does not identify those mechanisms on its own; qualitative research and experiments are needed to distinguish them.
Key Takeaway
The biggest drop-off is not always the best opportunity — prioritize where you have both a clear hypothesis and operational control.