How to Use GA4 Funnel Reports to Find Where You're Losing Customers

By Emily Redmond, Data Analyst at Emilytics · April 2026

TL;DR: A funnel report shows you the % of users who complete each step. If 100 land on homepage, 80 hit product page, 20 reach checkout—your leak is at product page, not checkout. Fix the biggest leak first.

I looked at a SaaS company's funnel and watched 1,000 people land on their site, but only 20 convert. My first instinct: the checkout sucks. Turns out, the checkout conversion was 80%. The real leak: only 25 people were making it to checkout.

The problem was at step 2, not step 5. They were fixing the wrong thing.

That's why you need a funnel report. It shows you exactly where the leak is.

What Is a Funnel Report and Why It Matters

A funnel report visualizes your customer journey as a series of steps. It shows:

• How many people start at step 1
• How many make it to step 2
• How many make it to step 3
• And so on

The % drop between each step is where your optimization opportunity is.

Example funnel for a SaaS free trial:

Key insight: The biggest leak is between "pricing page" and "trial signup" (50% drop). That's where to focus. A 10% improvement there = 34 more trials per 1,000 visitors.

How to Build a Funnel Report in GA4

Step 1: Open Explore Go to GA4 → Reports → Explore → Funnel Exploration

Step 2: Define your first step This is usually "landing page" or "session started." Click Add Step and choose:

• If it's a page: select "Page title" or "Page path"
• If it's an event: select the event (e.g., "session_start")

For a SaaS landing page funnel, start with Page path contains "/free-trial"

Step 3: Add subsequent steps Click Add Step for each stage of your funnel:

• Step 2: Product page view
• Step 3: Pricing page view
• Step 4: Trial signup event
• Step 5: Confirmation email event

Step 4: Add dimensions (optional but useful) Add "Device category" to see if mobile and desktop have different drop-off rates.

Add "Session source" to see if organic converts differently than paid.

Step 5: Run the report GA4 will show you:

• Users who completed each step
• Dropoff % between each step
• Dropoff % from step 1

Real Funnel Example: E-Commerce Checkout

Here's a typical e-commerce checkout funnel:

Analysis:

• Biggest leak: "Add to cart" (82% don't add to cart) — but this is normal. Not everyone who visits will buy.
• Second biggest leak: "Purchase" (50% start payment but don't finish) — this is your problem. Fix payment friction.
• Opportunity: If you reduced the "payment" drop from 50% to 40%, you'd convert 7 more people per 1,000. At $100 AOV, that's $700 per 1,000 visitors.

How to Identify Your Biggest Opportunity

Look for the step with the steepest drop. That's your priority.

But be smart about it: a 30% drop at step 1 (to 1,000 users) is a bigger opportunity than a 50% drop at step 5 (from 50 users).

Formula for opportunity size: Users entering step X × % who drop = opportunity size

E-commerce example:

• Step 2 drop: 1,000 × 82% = 820 opportunity
• Step 6 drop: 70 × 30% = 21 opportunity

Step 2 looks big, but it's expected (not everyone buys). Step 6 is your real opportunity.

Debugging a Funnel

Sometimes your funnel data looks wrong. Here are common issues:

Issue 1: Step 2 has more users than Step 1 This means your funnel is wrong. A user can't get to step 2 without hitting step 1. Fix the condition on step 1.

Issue 2: Drop-off is 100% between two steps Nobody is making it to step 2. Either:

• Your condition is too strict (e.g., page path = "/checkout" when some traffic goes to "/checkout-new")
• There's a technical issue (redirect, error page)
• The step doesn't exist in your funnel

Check your pages and conditions.

Issue 3: Drop-off seems too small Remember: GA4 counts users who complete step 1, then later complete step 2—not necessarily in the same session.

If someone visits your site Monday, then returns Wednesday to buy, that's still "completed funnel." Adjust your funnel to same-session only if needed.

Funnel Variations: Segment by Traffic Source

Your funnel isn't one-size-fits-all. Organic visitors might have a different journey than PPC visitors.

To compare:

1. Build your base funnel
2. Add "Session source" as a dimension
3. The report will break down by organic, paid, direct, etc.

Common pattern:

Organic traffic converts better because they're further along the buying journey (they searched for your product intentionally).

If your PPC rate is 1% and organic is 2%, that's normal. Don't panic.

Building a Testing Roadmap from Your Funnel

Once you've identified your biggest leak, build a testing roadmap:

Month 1:

• Test to improve "step 1 to 2" conversion
• Example: Product page isn't compelling. Test new copy or images.

Month 2:

• If month 1 test won, move to step 2 to 3
• If month 1 test lost, try a different hypothesis

Keep going: Each successful test moves your needle. A 10% improvement at step 1 compounds.

How to Prioritize Your CRO Roadmap Using Analytics Data walks through building a full testing roadmap.

Micro vs. Macro Funnel

A micro funnel tracks small engagement milestones:

• Click CTA button → view form → submit form → confirmation

A macro funnel tracks business outcomes:

• Trial signup → activate trial → convert to paid

Build both.

Macro funnel tells you the big picture (3% conversion). Micro funnel tells you where the friction is (20% don't see the form because they don't scroll).

Frequently Asked Questions

Q: Should my funnel include micro-conversions or just the final purchase? A: Both. Track both. The macro funnel (end-to-end) shows the outcome. The micro funnel shows where friction exists.

Q: What's a "bad" drop-off rate? A: It depends on the step. A 80% drop from "viewed product" to "add to cart" is normal. A 50% drop from "started checkout" to "entered shipping" is bad.

Q: Can I include people who drop off and come back? A: Yes, by default GA4 includes them. If you want "same-session only," you can filter by session-level data. This is more restrictive but sometimes useful for understanding immediate friction.

Q: How long should I look at funnel data before optimizing? A: At least 2 weeks (to account for day-of-week variation), better 4 weeks. See How Long Should Your Analytics Observation Period Be?

Q: What if my funnel has 10 steps? Should I include all? A: Focus on 4–5 key steps. Too many steps and the report gets hard to read. If step 3 to 4 has no drop, combine them.

Q: How do I track offline conversions in a funnel? A: You need to import offline conversions into GA4 using Data Import. Then you can include them as a final step. This is more advanced, but doable.

The Bottom Line

Your funnel report answers the most important CRO question: "Where are we losing people?"

Once you know, you can test. Once you test, you improve. It's that simple.

Build your funnel this week. Find your biggest leak. Fix it. Repeat.

Emily Redmond is a data analyst at Emilytics — AI analytics agent watching your GA4, Search Console, and Bing data around the clock. 8 years experience. Say hi →