How to Set Up Conversion Funnels: A Practical Guide

What is Funnel Analysis?

Funnel analysis is the process of mapping a multi-step user journey and measuring how many users complete each step versus how many drop off. The “funnel” metaphor comes from the shape of the data: many users enter at the top (first step), and progressively fewer make it through each subsequent step to the bottom (goal completion).

Consider a typical ecommerce checkout. A thousand users view a product page. Of those, 400 add the item to their cart. Of the 400, 200 begin checkout. Of those 200, only 120 complete the purchase. That’s a 12% overall conversion rate—but the funnel reveals where the losses happen. The biggest drop is between product view and add-to-cart (60% lost), not between checkout and purchase (40% lost). Without the funnel, you might spend weeks optimizing the checkout form when the real problem is the product page.

This is the core principle of funnel analysis: you can’t fix what you can’t see. Aggregate metrics like bounce rate and conversion rate tell you something is wrong. Funnels tell you exactly where.

Why Funnel Analysis Matters

Think of your website as a bucket with holes. Water (traffic) pours in at the top, but only a fraction makes it to the bottom (conversions). You could pour in more water—spend more on ads, write more content, run more campaigns—but that’s expensive and the water still leaks out. The smarter approach is to find the holes and patch them.

Funnel analysis is how you find the holes.

Without funnels, teams make decisions based on intuition. A product manager might assume the pricing page is the problem because it “feels” complicated. A designer might redesign the homepage because it looks dated. But when you build a funnel, the data often tells a different story. Maybe 85% of users breeze through the pricing page, but 60% abandon during account creation because the password requirements are confusing. That’s not a hypothesis—it’s a fact, measured step by step.

Funnels also give you a framework for prioritizing improvements. If step 2 loses 50% of users and step 4 loses 10%, fixing step 2 has five times the potential impact. This matters when you have limited engineering and design resources—which is always.

Types of Funnels to Track

The right funnels to build depend on your business model, but most websites benefit from tracking at least two or three of these:

Signup and Registration Funnels

For SaaS products and membership sites, the signup funnel is the most important funnel you’ll build. A typical sequence: landing page → pricing page → signup form → email verification → first login. Each transition is a place where potential users give up. Common problems include forms that ask for too much information upfront, confusing plan selection, and email verification flows that lose users in their inbox.

Ecommerce Purchase Funnels

The classic ecommerce funnel: product page → add to cart → cart page → checkout → payment → confirmation. Cart abandonment rates average around 70% across the industry, but funnels reveal whether your users are leaving at the cart stage (price shock, shipping costs) or the payment stage (trust issues, limited payment options).

Onboarding Funnels

Getting a user to sign up is only half the battle. Onboarding funnels track whether new users complete the steps that make them successful: completing their profile, connecting an integration, creating their first project, inviting a team member. Users who don’t complete onboarding are far more likely to churn.

Feature Adoption Funnels

When you launch a new feature, build a funnel that tracks discovery → first use → repeated use. This tells you whether the feature has a discovery problem (users don’t know it exists), a usability problem (they try it once and give up), or a value problem (they try it but don’t come back).

Content Engagement Funnels

For content-driven sites, track the journey from article view → second page view → email signup or resource download. Content funnels help you understand which articles drive meaningful engagement versus which ones attract one-time visitors who bounce.

Setting Up Your First Funnel in Inspectlet

Inspectlet’s funnel builder lives inside the Events page under the Funnels tab. Here’s how to set up a funnel from scratch.

Step 1: Navigate to the Funnel Builder

Open your Inspectlet dashboard and go to the Events page. You’ll see a Funnels tab at the top—click it to open the funnel builder. This is where you’ll define the sequence of steps that make up your funnel.

Step 2: Define Your Funnel Steps

A funnel needs at least two steps. Each step represents an action the user must take, and Inspectlet supports three types of actions:

• Visited URL — the visitor loaded a specific page. Use this for tracking page-to-page progression like landing page → pricing → signup. This also works with virtual page URLs, so single-page applications can use URL-based funnel steps.
• Clicked on button — the visitor clicked a specific element. Use this for tracking interactions that don’t necessarily change the URL, like clicking “Add to Cart” or expanding an accordion.
• User Typed — the visitor typed in a specific field. Use this for tracking form engagement, like whether users actually start filling in the email field on your signup form.

Add steps one at a time, choosing the appropriate type for each. You can mix and match types within the same funnel—for example, step 1 could be a URL visit, step 2 a button click, and step 3 another URL visit.

Step 3: Run the Funnel and Read Results

Once your steps are defined, click Run Funnel. Inspectlet processes your data and returns the results in seconds. You’ll see:

• Overall conversion rate — the percentage of users who entered step 1 and completed the final step
• Per-step visitor counts — how many users reached each step
• Step-to-step conversion percentages — the conversion rate between each consecutive pair of steps
• A visual bar chart — showing the funnel progression at a glance, making it immediately obvious where the biggest drops occur

Reading Funnel Data

Raw numbers are useful, but the real insight comes from knowing how to interpret them. Here’s a framework for reading funnel data effectively.

Overall Conversion Rate

This is your headline metric—the percentage of users who complete the entire funnel. It’s useful for benchmarking over time (is our funnel getting better or worse?) and for comparing against industry averages. But don’t optimize for this number directly. Improving a single step often has more impact than trying to improve everything at once.

Per-Step Drop-Off

Look at each step transition individually. A healthy funnel has relatively even conversion rates between steps. If most steps convert at 75–85% but one step converts at 40%, that step is your problem. Focus your investigation there.

Common patterns to watch for:

• Steep first drop-off — If you lose most users between step 1 and step 2, your traffic may be poorly qualified, or your landing page isn’t compelling enough to move users forward.
• Late-stage drop-off — Losing users at the final step (payment, confirmation) often indicates trust issues, unexpected costs, or technical problems.
• Gradual even decline — If every step loses a similar percentage, the issue may be overall friction rather than a single broken step. Simplifying the entire flow (fewer steps, less required input) may help more than optimizing individual steps.

Identifying the Biggest Leak

Calculate the absolute number of users lost at each step, not just the percentage. A step that converts at 50% but only has 100 users entering it loses 50 users. A step that converts at 80% but has 10,000 users entering it loses 2,000 users. The second step is a much bigger opportunity, even though its conversion rate looks better.

Investigating Drop-Off with Session Replay

Funnel data tells you where users drop off. Session replay tells you why. This is where Inspectlet’s funnel integration becomes powerful.

Watching Abandoner Sessions

Inspectlet’s funnel results are tied directly to the session list. When you see a drop-off at a specific step, you can view the sessions of users who abandoned at that step. You’re not guessing why users left—you’re watching it happen.

For example, if your checkout funnel shows a 45% drop between “entered shipping address” and “submitted payment,” you can watch the sessions of the 55% who didn’t make it. You might discover that users are abandoning because shipping costs appear for the first time at this step, or because the payment form throws a validation error on certain credit card formats, or because international users see a form that only accepts US addresses.

Looking for Patterns

Watch at least 10–15 abandoner sessions per step before drawing conclusions. A single session might show an edge case. Patterns emerge across multiple sessions. Look for:

• Repeated hesitation — Users hovering over a field or button for several seconds before clicking away. This signals confusion or uncertainty.
• Back-and-forth navigation — Users going forward in the funnel, then back, then forward again. They’re looking for information they didn’t find.
• Form field struggles — Users repeatedly typing and deleting in a specific field. The field’s validation, placeholder text, or format expectations may be unclear.
• External navigation — Users leaving the funnel to visit your FAQ, pricing, or terms pages. They have unanswered questions that should be addressed within the funnel itself.

Form Analytics Funnels

Inspectlet’s form analytics provides a complementary type of funnel that operates at the field level rather than the page level. While your Events funnel tracks page-to-page or action-to-action progression, a form analytics funnel shows how users move through individual fields within a single form.

Field-Level Funnels vs. Page-Level Funnels

Page-level funnels answer: “Where in the overall journey do users drop off?” Field-level funnels answer: “Where within a specific form do users drop off?” These are different questions, and you often need both.

Suppose your page-level funnel shows a 40% drop-off on the signup page. You know users are leaving during signup, but you don’t know which part of the form causes it. The form analytics funnel might reveal that 90% of users complete the email and name fields, but only 55% complete the phone number field. Now you know: the phone number field is the problem. Maybe it’s required when it shouldn’t be, or users don’t understand the expected format.

The two funnel types work together. Use page-level funnels to identify which step has the problem, then use form analytics to diagnose what within that step is causing friction.

Optimizing Funnel Performance

Once you’ve identified where users drop off and watched sessions to understand why, it’s time to make changes. Here’s a structured approach.

Prioritizing Fixes by Impact

Not all funnel improvements are equal. Prioritize by:

1. Volume of users affected — A step that 10,000 users hit per month is a bigger opportunity than one that 200 users hit.
2. Severity of drop-off — A step with 30% conversion is more urgent than one with 70% conversion.
3. Ease of fix — If watching sessions reveals that users abandon because a required field is confusing, making that field optional or adding helper text is a quick win. Redesigning an entire flow is a bigger investment.

Start with the highest-impact, lowest-effort fixes. These quick wins build momentum and demonstrate the value of funnel analysis to your team.

A/B Testing Funnel Changes

When possible, A/B test your funnel optimizations rather than deploying them directly. This gives you clean causal evidence that the change improved conversion, not just a before-and-after comparison that could be confounded by seasonality or traffic changes.

Focus your A/B tests on the specific step you’re trying to improve. If you change the shipping cost display on the checkout page, measure the conversion rate from checkout to payment completion—don’t just look at overall funnel conversion, which might fluctuate for unrelated reasons.

Re-Measuring After Changes

After deploying a fix (or concluding an A/B test), re-run your funnel to measure the impact. Compare the new per-step conversion rates to your baseline. Did the target step improve? Did any other steps change unexpectedly? Sometimes fixing one step reveals a bottleneck that was hidden behind a bigger problem upstream.

Make this a recurring process. The best teams run their core funnels weekly or monthly and treat step-over-step conversion as a key metric in their dashboards, right alongside revenue and active users.

Advanced Funnel Strategies

Segmenting Funnels

Aggregate funnels show you the average experience, but averages can be misleading. Segmenting your funnels reveals how different user groups behave:

• New vs. returning users — New users often have much lower funnel conversion rates because they’re less familiar with your site. Returning users may skip steps entirely.
• Device type — Mobile funnels frequently have steeper drop-offs than desktop, especially on forms and checkout flows. If your mobile conversion rate is half your desktop rate, you have a mobile UX problem.
• Traffic source — Users from organic search may convert differently than users from paid ads. Paid traffic is often warmer (higher intent) but may have different expectations based on the ad creative they saw.
• Geography — International users may drop off at steps that assume US-centric formatting (zip codes, phone numbers, payment methods).

By segmenting, you avoid optimizing for the “average user” and instead address the specific problems each group faces.

Multi-Path Funnels

Not every user takes the same path to conversion. Some users visit the pricing page before the product page. Some skip the product demo entirely. When your funnel assumes a single linear path, you miss the users who take alternative routes.

Build multiple funnels for different entry points and paths. For example, create one funnel for users who enter through your homepage and another for users who land directly on a product page from search. Comparing these funnels reveals which entry paths produce the highest-converting users—and where each path loses people.

Combining Funnels with Broader Behavior Analysis

Funnels are one piece of the puzzle. Combine funnel data with other user behavior analysis techniques for a complete picture:

• Heatmaps on high-drop-off pages show where users are clicking (and where they aren’t). If the “Continue” button gets few clicks but the “Back” button gets many, users are actively retreating.
• Session recordings of abandoner sessions (as discussed above) reveal the qualitative story behind the quantitative data.
• Error tracking can correlate JavaScript errors with funnel drop-off points. If a step has a high drop-off rate and frequent errors, the errors may be the cause.

Common Funnel Analysis Mistakes

Funnel analysis is straightforward in theory, but teams commonly make these errors in practice:

1. Building Funnels with Too Many Steps

A 15-step funnel is hard to interpret and hard to act on. Each step adds noise. Keep funnels to 3–7 steps that represent meaningful decision points, not every micro-interaction. If you need more detail on a specific step, use form analytics or session replay to zoom in.

2. Tracking the Wrong Steps

The steps in your funnel should represent the user’s decision-making process, not your page structure. If users can add to cart from three different pages, your funnel step should be “clicked Add to Cart” (a button click), not “visited the product page” (a URL). Match the funnel to the user’s intent, not your sitemap.

3. Ignoring Segments

A funnel that averages all users together can hide important differences. If mobile users convert at 2% and desktop users at 8%, the blended rate of 5% doesn’t tell you much. Always segment before you optimize.

4. Optimizing Without Qualitative Context

The funnel tells you what is happening. Session recordings tell you why. If you see a 50% drop at step 3 and immediately start redesigning step 3 without watching abandoner sessions, you might fix the wrong thing. Maybe the problem isn’t step 3 at all—maybe step 2 sets wrong expectations that lead to drop-off at step 3.

5. Setting Up Funnels and Never Revisiting

Funnels lose value if you build them once and forget about them. Your site changes, your traffic mix shifts, new features launch. Set a cadence—weekly or monthly—to re-run your core funnels and compare against your baseline. Treat funnel health as an ongoing metric, not a one-time audit.

6. Making Changes Without a Baseline

Before you change anything, run your funnel and record the current numbers. Without a baseline, you can’t measure whether your optimizations actually worked. A conversion rate optimization process without measurement is just guessing.

Getting Started

Funnel analysis isn’t a one-time project. It’s a continuous practice of measuring, investigating, fixing, and re-measuring. The companies that do it well treat their funnels as living metrics—always running, always informing decisions.

Here’s a practical starting point:

1. Identify your most important conversion goal — signup, purchase, trial activation, whatever drives revenue.
2. Map the 3–5 steps between first visit and that goal — landing page, key decision pages, form submissions, confirmation.
3. Build the funnel in Inspectlet using the Events → Funnels tab. Mix URL visits, button clicks, and typed inputs as needed.
4. Run it and find the biggest drop-off — calculate where you lose the most users in absolute numbers.
5. Watch 10–15 abandoner sessions at that step — look for patterns, not one-offs.
6. Fix the most impactful issue and re-measure — confirm the improvement with data before moving on.

That cycle—measure, investigate, fix, re-measure—is the entire practice of funnel optimization. Every round makes your product a little more effective at converting visitors into users and users into customers.

Frequently Asked Questions

What is a conversion funnel?

A conversion funnel is a model of the steps a user takes to complete a goal on your website—such as signing up, making a purchase, or completing onboarding. It’s called a “funnel” because fewer users complete each successive step. By measuring the drop-off between steps, you can identify exactly where users abandon and focus your optimization efforts where they’ll have the most impact.

How many steps should a funnel have?

Keep funnels to 3–7 steps that represent meaningful decision points in the user journey. Too many steps make funnels hard to interpret and act on. If you need more detail on a specific step, use form analytics for field-level data or session replay to watch exactly what happens at that step.

What’s a good conversion rate between funnel steps?

Healthy funnels typically see 60–85% conversion between each step, though this varies by industry and funnel type. The absolute number matters less than relative performance—if most steps convert at 75% but one converts at 35%, that step is your problem. Focus on the steps with the steepest drop-off relative to adjacent steps.

How do I find why users drop off at a specific step?

Combine funnel data with session replay. When you see a drop-off at a specific step, watch 10–15 recordings of users who abandoned at that point. Look for patterns like hesitation, back-and-forth navigation, form struggles, or error encounters. Heatmaps on the drop-off page can also reveal whether users are scrolling far enough to see the next-step button.

Can I track funnels for single-page apps?

Yes. Inspectlet auto-detects virtual page views in single-page applications, so route changes in React, Vue, Angular, and other SPA frameworks work as funnel steps without extra configuration. You can also define funnel steps based on button clicks and form interactions, not just URL changes.