Conversion Funnel Analysis: How to Find and Fix Drop-Offs

Conversion funnel analysis is the practice of measuring how many users move from one step of a journey to the next (for example, landing page, product page, cart, checkout, purchase) and finding the step where the most people drop off. It tells you which part of the journey to fix first, so you stop trying to fix everything at once.

Picture a busy grocery store on a Saturday afternoon. You are hungry, you wander into the bread aisle, a fresh artisan loaf catches your eye, you pick it up, you imagine the toast, and you head to the checkout. You join the queue, you wait, you buy. That is a complete conversion funnel, from the first ping of hunger to the checkout lane. Online, the same journey happens in clicks, and every step loses some people.

What is conversion funnel analysis?

The funnel maps each phase of your potential customer's mindset as you try to engage and persuade them. At every stage the audience becomes smaller and more focused as it approaches conversion. The analysis puts a number on each narrowing, and the numbers tell you where to look.

Why conversion funnel analysis matters

Funnel analysis does three jobs:

• It shows where people leave. Funnel reports show which steps and channels bring a lot of visitors but also lose a lot of them without action. Finding where users leave is like patching leaks in a funnel: you keep more traffic flowing through.
• It shows where your best visitors come from. Funnels do not only find problems. Broken down by source, they reveal which channels send the traffic that completes the journey, so you can put more effort into those channels.
• It helps your team and stakeholders decide. A funnel report is a simple way to show where the online business is thriving and where it could improve. A big drop-off highlighted in red is a powerful motivator for action.

What are the stages of a conversion funnel?

To make the stages concrete, suppose you are launching a new line of organic skincare products.

Awareness stage

This is where you introduce people to your brand and capture their attention. Getting the word out about your products, website and company is the first step in building a lasting relationship with customers. For the skincare brand, that could mean social media advertising, influencer partnerships or content marketing, such as beauty creators on Instagram or Facebook showing the benefits of natural ingredients.

Interest stage

Once potential customers are on your site, you need to spark their interest. Relevant content, appealing visuals and special offers draw them in. The skincare brand could publish articles about the ingredients to avoid and the benefits of organic alternatives, show clear product images, and offer an introductory discount.

Desire stage

Here you build trust and create the desire to purchase. A personalized journey moves visitors closer to buying: recommend products based on the visitor's skin concerns, offer free samples, and show testimonials and reviews from people who have already used the products. Social proof builds credibility and confidence in the brand.

Action stage

The visitor is ready to buy. Do not lose them after all the work in the earlier stages: streamline the checkout, offer secure payment options, and give incentives such as free shipping or a satisfaction guarantee. For the checkout step itself, see our guide to shopping cart optimization.

How B2B and B2C conversion funnels differ

In B2B you sell to companies, and companies are made of people who decide together. A B2B funnel has to convince a team, address the concerns of multiple stakeholders, and survive layers of approvals, evaluations and negotiations. The sales cycle often takes weeks or months, so the analysis focuses on how leads are nurtured until the deal closes.

B2C is more straightforward. The decision is often more impulsive or emotional: consumers see the product, like it, and click "buy". The analysis focuses on making the buying process smooth, appealing to emotion, and standing out from the competition.

What does a conversion funnel look like in different businesses?

In the coffee app example, a large drop between viewing menu details and completing orders gives the team a clear hypothesis: promote the most popular items at the top of the menu and test whether more orders get completed.

How do you calculate funnel conversion rates?

Worked example

The skincare store tracks 10,000 users who land on the site in a month (illustrative numbers):

Overall funnel conversion rate: 240 ÷ 10,000 × 100 = 2.4%. The step rates multiply to the same result: 40% × 20% × 50% × 60% = 2.4%.

The biggest leak here is product page to cart, where 80% of users drop off. That is where the analysis should dig first. A small improvement there has a large effect: if 25% of product-page visitors added to cart instead of 20%, the same funnel would produce 300 purchases instead of 240.

For the click-level view of the same journey, see how to use click-through rate to measure funnel conversion.

Which metrics should you track at each funnel stage?

Note that Google Analytics 4 reports bounce rate as the inverse of engagement rate: a session counts as engaged if it lasts longer than 10 seconds, has a key event, or has two or more page or screen views. For formulas and the full list, see our guide to CRO metrics.

How do you run a conversion funnel analysis?

1. Define the customer journey and map its touchpoints

   Create a customer journey map that shows how customers move through the funnel stages, which touchpoints they meet, and how often they reach the "aha" moment. It tells you which steps to track and which ones are likely to need work. Our guide to user flows shows how to diagram the paths inside your site.
2. Set up conversion events

   Track the macro conversion (purchase, subscription) and the micro conversions that lead to it (product view, add to cart, sign-up). Different funnel stages may need different events, based on your audience and how engaged they are at that point.
3. Build the funnel in your analytics tool

   In Google Analytics 4, go to Explore and create a funnel exploration with one step per event. Choose a closed funnel if users must start at step one, or an open funnel if they can enter at any step. Use a path exploration to see the routes users actually take, forward from a starting point or backward from a purchase.
4. Segment users by source, device and conversion point

   A blended funnel hides the problem. Break it down by channel, device and customer type, and group customers by the conversion points they reached (trial sign-up, first purchase, subscription). Look closely at the funnel of your best customers: what worked for them is a model to improve the funnel for everyone else.
5. Investigate the biggest drop-off

   The numbers show where users leave, not why. Use heatmaps, session recordings and on-site surveys on the leaking step to find the friction points: what users expected, what confused them, and what they needed to continue.
6. Turn the friction into a hypothesis and test it

   Write a hypothesis for the leak ("Showing delivery costs on the product page will increase add-to-cart rate"), run an A/B test, and measure the result on both the step rate and revenue. Then move to the next-biggest leak.

A note on Google Analytics

Older funnel guides refer to Universal Analytics goal funnels. Universal Analytics stopped processing data on July 1, 2023 (July 1, 2024 for Analytics 360 properties). In GA4, use funnel exploration, which supports up to 10 steps, open and closed funnels, standard or trended views, elapsed time between steps, and a "next action" breakdown, and path exploration for the routes between steps.

Which tools do you need for conversion funnel analysis?

• Web analytics
  Google Analytics 4

  Tracks acquisition, engagement and key events, and builds funnel and path explorations. Probably the most common starting point for funnel analysis.

  Funnel exploration, path exploration
• Behavior analytics
  Heatmap tools (e.g. Hotjar, Crazy Egg)

  Heatmaps, click maps and scroll maps show where users engage on a page and where friction sits, so you can improve layouts, CTAs and content placement.

  Best for: the "why" behind a drop-off
• A/B testing
  Omniconvert Explore

  Test variations of headlines, CTAs and page layouts on the leaking step, personalize by segment, and run on-site surveys to ask users what stopped them.

  70,000+ experiments, 23.2% average uplift
• CRM
  CRM software (e.g. Salesforce, HubSpot, Zoho CRM)

  Centralizes customer data, tracks interactions, segments audiences and nurtures leads at every stage, giving a more complete view of customer behavior. Essential for long B2B funnels.

  Best for: lead nurturing and B2B
• Product analytics
  Advanced analytics platforms (e.g. Mixpanel, Amplitude)

  Event-based tracking with cohort analysis, funnel visualization and predictive features for deeper analysis of user behavior, especially in apps and SaaS products.

  Best for: apps and SaaS funnels

How do you optimize a leaking conversion funnel?

Suppose you have run the analysis and the numbers are not what you expected. These are the practices that fix the most common leaks:

• Keep the process simple. Reduce the number of steps to buy and the effort each step takes.
• Align the landing page with the promise. If your ads promise free tomatoes, do not sell expensive ketchup on the landing page.
• Reduce cart abandonment. Be transparent about prices and delivery costs, keep the checkout simple and intuitive, and consider an exit-intent popup with a relevant incentive.
• Optimize for mobile. Make sure every step works, and is accessible, on every device.
• Personalize. Use zero- and first-party data for dynamic content and segment-specific experiences that match individual preferences.
• Re-engage users who left. Retargeting ads and email campaigns bring users back into the funnel with personalized touchpoints.

What are the challenges in conversion funnel analysis?

Data silos and integration issues

When data is scattered across platforms, combining it into one dataset is difficult, and without a unified view of customer interactions accurate insight is hard to get. Invest in data integration tools that move data between systems. That is how you break down silos and centralize the data behind your funnel decisions.

Keeping up with changing consumer behavior

Consumer behavior changes with technology, culture and the market. Factors such as sustainability, eco-friendly delivery and fair pay now play a part in consumer choices; a 2023 study in Sustainability (Šostar and Ristanović) assesses the factors that influence consumer behavior. Stay agile: monitor behavior continuously, use analytics to track emerging trends, and adapt your marketing as they change.

Privacy and data protection regulations

You always need a balance between data-driven insight and user privacy. Comply with regulations such as GDPR and CCPA: collect data transparently, get explicit consent, secure the data, and anonymize sensitive information. Transparent privacy practices also build consumer trust.

How are AI and predictive analytics changing funnel analysis?

Machine learning models analyze historical data to anticipate customer actions, from purchase intent to churn risk, and AI-powered tools apply that to lead scoring and dynamic content, so you can adapt your strategy before users leave.

The same technologies are changing the experience inside the funnel. Conversational AI, voice search and augmented reality give brands new ways to engage people at each stage. Sephora's Virtual Artist, for example, scans the shopper's face and lets them try on makeup virtually. Add omnichannel marketing and integrated data, and you get a more complete story of customer interactions across touchpoints.

Remember the bread from the beginning? Nothing stops you from creating the same desire in your own audience. Use conversion funnel analysis to find where that desire gets lost, fix it, and build an experience that brings people back again and again.

Frequently Asked Questions