What is a Funnel?

A funnel represents the staged journey from a prospect's first brand contact to a purchase. At each stage some users drop out; the surviving group shrinks, hence the inverted-funnel shape. Each stage's conversion rate is tracked separately to find where the bottleneck lives.

classic funnel: AIDA

• Awareness · user learns the brand exists.
• Interest · engages with what the brand does.
• Desire · wants the product, compares alternatives.
• Action · purchase, signup, download; the commitment step.

modern funnel: TOFU / MOFU / BOFU

• TOFU (Top of Funnel) · broad audience, educational/awareness content. Metrics: reach, CPM, video views.
• MOFU (Middle of Funnel) · interested users, comparison content, lead magnets. Metrics: CPL, ebook downloads, demo signups.
• BOFU (Bottom of Funnel) · purchase-ready users, pricing, demos, free trial. Metrics: CPA, conversion rate, revenue.

how to do funnel analysis

1. Define stages clearly, e.g. homepage → product page → cart → checkout → order complete.
2. Measure user count at each stage (GA4, Mixpanel, Amplitude, d-lens).
3. Compute the stage-to-stage conversion rate.
4. Find the stage with the biggest drop, the largest "bottleneck".
5. Build a hypothesis focused on that stage, A/B test it, measure.

e-commerce conversion rate benchmarks

• Homepage → product: 30-50%.
• Product → cart: 8-15%.
• Cart → checkout: 50-65%.
• Checkout → order: 70-85%.
• Total homepage → order: 1.5-3.5%.

common funnel pitfalls

• Assuming it's linear · real customer journeys zigzag; the same user comes back 3-4 times, from different devices.
• Investing only in TOFU · traffic arrives but conversion doesn't.
• Investing only in BOFU · short-term sales happen but no new audience arrives, growth stalls.
• Last-click attribution · TOFU's value stays invisible in reporting; data-driven attribution fixes this.