Funnels and Drop-off Analysis
A funnel maps the steps users take from first interaction to conversion. The value isn't in knowing how many people convert — it's in seeing exactly where they stop.
When you know that 76% of users abandon between "product page" and "add to cart," you know exactly where to focus. Without funnel data, you're guessing.
This guide covers how to build funnels that reveal actionable insights, diagnose why drop-offs happen, and prioritize fixes by revenue impact. For the broader context of conversion tracking, see the pillar page.
How funnels work
A funnel is a sequence of steps that users should complete in order. Each step filters users — some continue, some leave.
| Step | Sessions | Drop-off |
|---|---|---|
| Landing page | 10,000 | — |
| Product page | 4,200 | 58% |
| Add to cart | 1,100 | 74% |
| Checkout start | 780 | 29% |
| Payment entered | 620 | 21% |
| Order confirmed | 540 | 13% |
The funnel shows where users leave. High drop-off at any step means something is wrong there — confusing UI, technical problems, pricing objections, or missing trust signals.
Funnel design principles
Keep steps sequential. Each step should require the previous step. "Viewed product → Added to cart → Started checkout" is valid. "Viewed product → Read blog → Added to cart" isn't — the blog visit isn't part of the purchase journey.
Don't over-segment. Too many steps make data noisy. Group related actions: "Filled shipping → Filled billing → Submitted" can become "Completed checkout form."
Include micro-conversions. Steps before the final conversion reveal intent: pricing page views, add to cart, account creation. These help you understand where interest turns into friction.
Diagnosing drop-offs
High drop-off at a step means users wanted to continue but didn't. The question is why.
Technical problems
Errors that prevent users from completing actions. A JavaScript exception on the checkout page. A form that won't submit. A button that doesn't respond.
Connect funnel data to error data. If sessions with errors have dramatically lower conversion rates, those errors are directly costing you revenue. The error prioritization framework helps quantify this.
UX friction
Confusing interfaces, unclear copy, too many required fields, unexpected costs, slow loading. These problems don't throw errors — they create friction.
Rage clicks and other behavioral signals indicate UX problems. Users clicking repeatedly on something that doesn't respond suggests confusion or frustration at that step.
Traffic quality mismatch
Not all traffic converts equally. If one traffic source shows 90% drop-off at step one while others show 40%, the problem might be targeting, not your funnel.
Segment funnels by traffic source to identify quality issues before optimizing the wrong thing.
Pricing and value perception
Sometimes users reach checkout and decide not to proceed. They saw the price. They compared to competitors. They weren't convinced.
This isn't a funnel problem you can fix with UI changes — but funnel data tells you where the decision happens. If drop-off spikes after users see the total (including shipping), that's a pricing signal. If it happens before they see pricing, it's something else.
Load time and performance
Slow pages cause abandonment. Each second of delay reduces conversion rates measurably. If a funnel step loads slowly, users leave before the step even renders.
Check Core Web Vitals for each step in your funnel. A slow checkout page loses more revenue than a slow blog post.
Common funnel types
E-commerce checkout
Product page → Add to cart → Cart review → Checkout start → Payment → Confirmation
Key metrics: Cart abandonment rate (typically 70%), checkout completion rate, payment failure rate.
SaaS signup
Landing page → Pricing view → Signup start → Account created → First action completed
Key metrics: Pricing page conversion, signup form completion, activation rate.
Lead generation
Landing page → Form view → Form start → Form complete → Confirmation
Key metrics: Form view to start rate (measures intent), form completion rate (measures friction).
Multi-step forms
Step 1 → Step 2 → Step 3 → Submit → Confirmation
Key metrics: Drop-off per step, time per step, back-button usage.
Mobile vs desktop funnels
Mobile traffic often converts at lower rates — but that doesn't mean mobile is "broken." Mobile users behave differently.
| Metric | Desktop | Mobile |
|---|---|---|
| Sessions | 4,200 | 5,800 |
| Conversion rate | 4.8% | 1.9% |
| Avg. steps before exit | 3.2 | 1.8 |
Lower mobile conversion might indicate:
- Checkout UI issues on small screens
- Payment methods that don't work well on mobile
- Research behavior (users browse mobile, buy desktop)
Segment your funnels by device to see where mobile-specific problems exist versus where the pattern is expected.
Connecting funnels to revenue
The ultimate question: how much is this drop-off costing?
Weekly cost of drop-off = dropped sessions × expected conversion rate × average order value
If 500 users per week abandon at checkout start, your baseline completion rate from that point is 60%, and average order value is $85:
500 × 0.60 × $85 = $25,500/week in potential revenue
Even improving that step by 10% recovers significant revenue. This math justifies engineering time and design resources.
The key insight: funnel analysis turns vague problems ("we're not converting") into specific, quantifiable issues ("checkout step 2 costs us $8,000/week"). That specificity makes prioritization clear and progress measurable.
Related pages
Parent: Conversion Tracking — Track the moments that matter and understand where users drop off
Deep dives:
- Checkout Drop-off — Why customers abandon at checkout
- Form Drop-off — Why forms get abandoned
- Mobile vs Desktop — Device-specific conversion differences
Pillar topics:
- Error Monitoring — Catch technical problems that break funnels
- Performance & Uptime — Slow pages cause drop-offs
Related:
- Prioritizing Errors by Revenue Impact — Framework for quantifying error cost