How To Find and Plug Your Sales Funnel Leaks

Visualize yourself carrying a bucketful of water back from a well. Now, imagine tiny holes emerging in your bucket. Slowly, water starts flowing out of your bucket as you walk home. There’s still water left when you arrive, but it’s plain to see you didn’t carry nearly as much as you could have.

The leaky bucket is a metaphor for your sales funnel, a path your visitors take from arriving at your store to making their purchase. Sales funnel leaks can quite literally make or break your business.

This guide will cover how to audit a leaky funnel, why sales funnel leaks can be detrimental to your pipeline (and bottom line), and the best tips for stopping sales funnel leakage.

How to find sales funnel leaks

There are three funnel-tracking-related reports in Google Analytics, which you can use to find leaks in your sales funnels:

• Funnel visualization report (Conversions > Goals > Funnel Visualization).This is the most basic report, which shows you a visual overview of your funnel, depending on the goal selected.
• Goal flow report (Conversions > Goals > Goal Flow).This shows the most accurate path to conversion. Plus, it’s a bit more flexible than the funnel visualization, because it allows you to use advanced segments and date comparison.
• Reverse goal path report (Conversions > Goals > Reverse Goal Path).This shows you your actual funnels. Here, you’ll discover funnels you didn’t even know existed. Essentially, you’ll see the three pages visited prior to conversion.

当回顾这些报告,问自己,visitors are exiting the funnels most often. These are the “leaks” you need to fix. To improve your conversion rate, you’ll need to figure out how to plug those leaks and keep more visitors in your funnels.

Between the three reports above, you should be able to quickly spot any issues with a leaky sales funnel. So, how do you plug the proverbial leaks?

"In God we trust—all others bring data."

In the words of engineer and authorW. Edwards Deming, “In God we trust—all others bring data.” The first step is conducting quantitative and qualitative research to find outwhythe holes exist and how best to plug them.

The lower you go in a funnel, the more impactful plugged leaks will be. A small increase in conversion rate goes a lot further at the bottom of a funnel than the top. Often, it’s smart to work your way from the bottom to the top for that very reason.

How to conduct quantitative research

Quantitative research is numeric and objective. It aims to uncover the “what” behind the behavior of your visitors and customers. Inconversion rate optimization, quantitative research usually refers to one of the following methods:

1. Technical analysis
2. Analytics deep dive
3. Form analytics
4. Heatmaps

1. Technical analysis

If your store doesn’t work well, it won’t convert well. That’s an absolute rule.

While it’s easy to think everyone uses the latest version of our favorite browser or OS, reality is more complex. You might have a shiny new iPhone 14, but someone somewhere is still rocking a Nokia 3310 from way back in 2017.

Technical analysis accommodates these different types of visitors and buyers, and largely covers three core concepts:

Cross-browser and cross-device testing.This is the process of ensuring your store works correctly in as many browsers and on as many devices as possible, which is no small task. The trick is that each browser and device has many versions, and it’svery easyto hit snooze on those update reminders. So, you can’t assume everyone is using the latest version.

You can use a tool likeBrowserStackand your preferred analytics tool to expedite the process. With Google Analytics, for example, you can navigate to two key reports:Audience > Technology > Browser & OSandAudience > Mobile > Devices. Switch from the Data view to the Comparison view to see how the browsers and devices compare to one another. Just be sure to compare within the same family (e.g., Android to Android, Chrome to Chrome).

Here’s an example:

You can see the Comparison view is active in the top right-hand corner and Purchase Completed has been selected as the comparison metric. What you’re looking at is a list of the most popular Chrome browser versions for your store and how well they convert.

These types of reports will help you prioritize your cross-browser and cross-device testing. You can start with the most popular and most troublesome browsers and devices (for your specific store) first.

Mobile optimization.Mobile is a whole other beast. It’s important to keep that in mind when you’re optimizing a mobile experience. What people want and need on mobile is very different from what they want and need on desktop. Intentions, motivations and contexts all change. A good mobile experience is not just a desktop experience on a smaller screen; a good mobile experience is a good mobile experience, full stop. Don't let a poor mobile experience ruin your entire sales funnel.

Page speed optimization.If your site is too slow, you’ll lose visitors before you even have a chance to market to them. If you’re using Google Analytics, you can use the Behavior > Site Speed > Page Timings report to identify slow pages. Then, run those pages throughPageSpeed Insightsfor tips on how to improve the page speed.

2.Analytics deep dive

If you’re usingShopify reports and analytics, you can rest assured your setup has been configured correctly. But what about tools like Google Analytics that require setup on your part? You’d be surprised how easy it is to configure an analytics tool incorrectly. Before you dive deep into your analytics, ask yourself:

• Am I collecting all of the data I need?
• Can I trust the data I’m collecting?
• Is anything broken or tracking incorrectly?

If your analytics are inaccurate or incomplete, decisions you make based on that data are misguided and, ultimately, useless.

Once you’re confident in your data, you can dive in to better understand how your visitors and customers behave. Here’s what to keep in mind as you swim through the data:

Start with a question or problem.British economistRonald Coaseonce said, “If you torture the data long enough, it will confess to anything.” In other words, if you go into data analysis with preconceived notions and assumptions, you’ll find something to back it up if you look hard enough. To avoid this trap, it’s important to start with a question to answer or a problem to solve. To be sure the question or problem is worth your time, ask yourself what you’ll do with the answer or solution. If you can’t clearly define the next steps, it’s probably not the right question or problem.

"If you torture the data long enough, it will confess to anything."

Start where the value is.You’ll have no shortage of data at your fingertips, regardless of what analytics tool you turn to. So, how do you get the most amount of value as quickly as possible? Start with high volume, low conversion pages (e.g., an old blog post) or low volume, high conversion pages (e.g., the checkout page). Conversion improvements will have a bigger impact here.

Fix broken links.Broken links mean 404 errors, which are bad for both theuser experienceandsearch engine optimization. The faster you can identify and fix broken links using your analytics tool, the better. In Google Analytics, for example, you can find broken links with the Behavior > Site Content > All Pages report. Search for the page title of your 404 page (i.e., “404 Page Not Found”) and select Page Title as your Primary Dimension in the top left-hand corner.

When you click through to “404 Page Not Found” (this will be slightly different for every site), you’ll find a list of URLs that are returning a 404. You can then use the Secondary Dimension drop-down in the top left-hand corner to select Full Referrer, which will show you URLs referring traffic to your 404s. This is helpful if you want to ask them to link somewhere else.

Internal search is a gold mine.If you haveinternal searchon your site, you’re sitting on a proverbial gold mine. Every time visitors search, they’re telling you what they want and how well you help them get what they want. In Google Analytics, for example, you can use the Behavior > Site Search > Search Terms report to uncover insights. The report will show you a list of the most popular search terms. All you have to do is look at the Time After Search metric to see how well you serve each search term. If searchers find what they’re looking for, time after search will be high. If not, it will be low. Whenever time after search is low for a popular term, that’s a product opportunity.

Segmentation is crucial.Segmenting your analytics data is invaluable. The more ways you can slice and dice your data, the more ways you can uncover insights. If you haven’t already, consider reading up onGoogle Analytics segments. This same concept should be applied to whichever analytics tool you use.

3. Form analytics

如果你有一个表格在你的网站或网页,whether it’s your checkout form or a simple lead generation form, consider it an important conversion point. A form is an interaction—an exchange between you and the visitor or customer. The more you know about that interaction and the friction associated with it, the better.

A form analytics tool likeZukocan help answer important questions, like:

• Which form fields cause the most error messages?
• Which fields do people hesitate to fill?
• Which fields do people leave empty, even if required?

You can use the data to reduce friction and improve the conversion rate.

4. Heat maps

Heat maps are visual representations of data, where values are represented by colors. Most tools will use warm colors (red, orange, yellow) to demonstrate high values and cool colors (blue, green) to demonstrate low values.

There are two main types of heat maps in conversion rate optimization:

• Click maps.The data that goes into click maps often seems more useful than it really is. Truthfully, click maps are best used for identifying places on your site that visitorsthinklink out. So, in other words, what are your visitors clicking on falsely, thinking they’ll be served a link? You can now turn those non-linked elements into links to deliver a better user experience.
• Scroll maps.Scroll maps are a bit more useful. They can help you prioritize your messaging, especially on category, product, and landing pages. If your scroll map suddenly jumps from red to blue, for example, you might need to use visual cues (like an arrow) to keep visitors scrolling. Alternatively, you might want to move your most important messaging above the drop-off point.

How to conduct qualitative research

Qualitative research is exploratory and subjective. It aims to uncover the “why” behind the behavior of your visitors and customers. In conversion rate optimization, qualitative research usually refers to one of the following methods:

1. On-site surveys
2. 成本的mer interviews
3. 成本的mer surveys
4. User testing
5. Session replays

1. On-site surveys

You’re likely already familiar with on-site surveys. They pop up as you browse around different websites, prompting you to answer a quick question. For example, here’s an on-site survey that Asics uses:

There are two primary types of on-site surveys:

1. Exit surveys.These are activated when the visitor shows exit intent, like hovering over the browser’s taskbar. This is your chance to gather feedback and insight from visitors before they leave. Exit surveys can be especially powerful when used on paying customers.
2. On-page surveys.These are activated when a visitor visits the page, either immediately or after a set period of time (e.g., 30 seconds). This is your chance to gather feedback and insight from visitors who are still browsing. With on-site surveys, you want to focus on extracting actionable data. The best way to do that is by identifying specific points of friction on the site.

In the Asics example above, visitors are asked to rate their experience on a 10-point scale. It will be difficult for the Asics marketing department to extract any meaningful data from the survey results. The results will tell the team how their store performs overall, but it won’t help them identify friction, which means it won’t help them improve their store’s experience.

The Asics request is quantitative (numeric) in nature. On-site surveys work best when they’re used qualitatively, which means asking an open-ended question that allows for anecdotes and explanation.

Here are some friction-focused open-ended survey question suggestions:

• What was the purpose of your visit to our site?
• Were you able to complete your task?
• If not, why?
• Is there anything holding you back?
• Do you have unanswered questions?

On-site surveys perform best when you:

• Ask one open-ended question
• Ask a yes/no question and ask for an explanation or elaboration after submission

2.成本的mer interviews

Nothing can replace getting on the phone and talking to your customers. If you can meet with your paying customers in person, that’s even better.

There are a million questions you could ask to get to the heart of who your customers are and why they really buy from you. What’s important is that you go into interviews prepared and well equipped to uncover insights.

That means:

• Recruiting the right participants.Not all customers are created equal. More often than not, you’ll get the best insights from recent customers,repeat customers, and lapsed customers. The question you’re trying to answer or the problem you’re trying to solve can help you decide which of those customer groups you should interview.
• Asking the right questions.There is no shortlist of questions you should always ask in a customer interview. Stick to short, open-ended questions. Be careful to remove yourcognitive biasesand assumptions from your questions. Before asking questions about the solutions you provide, be sure you understand the problem your customers are experiencing. Often the most insightful interview questions are problem-focused, not solution-focused. Finally, it’s not all about questions; you can have participants engage in role play, give demonstrations, etc.
• Documenting the interviews properly.Notes can be helpful, but you want to ensure you’ve recorded the interviews as well. Have them transcribed by a service likeRev. Capture audio, video, and written notes while interviewing. If you’re interviewing in person, it’s best to recruit a temporary assistant who can help. It’s difficult to connect with the participant and really engage while also worrying about documentation.

After the interviews, review the documentation and consider what you’ve learned. It helps tocreate a hypothesisabout what you believe before conducting the interviews so you can prove or disprove it through research. This information can then be used to improve both your marketing funnel and sales funnel where applicable.

3. Customer surveys

With on-site surveys, you’re asking one, maybe two, questions to site visitors. Full-fledged customer surveys allow you to ask your recent customers multiple questions.

When you’re putting together a customer survey, you want to focus on:

• Defining who your customers are
• Defining their problems before and after purchasing from you
• Defining the hesitations they had prior to purchasing
• Identifying words and phrases they use to describe your store and products

The best way to do that is to send the survey to recent, first-time buyers who have had no previous relationship with you.

Try to gather around 200 survey responses before analyzing them. This is a simple rule of thumb, not an absolute rule. It allows you to identify trends and patterns without leaving you to sort through an overwhelming amount of survey data.

Here are some questions to consider asking in your customer survey:

• Who are you?
• What are you using this product for?
• What problem does it solve for you?
• What do you like about the product most?
• Did you consider any alternatives?
• Why did you choose this product over the competition?
• What’s the one thing that nearly stopped you from buying?
• What was your biggest concern or hesitation about buying from this store?

Notice all of these sample questions focus on either thevalue your product deliversor the perceived friction pre-purchase. The more you know about those two things, the better. They’re critical conversion factors for improving elements on your product pages, likeproduct descriptionsand checkout experience.

You can also send surveys to repeat customers and lapsed customers, but recent first-time customers are usually the best place to start.

4. User testing

User testing is the process of watching real people try to perform tasks on your site while they narrate their thoughts and actions out loud.

This is useful because you’re likely too close to your store to recognize its faults and weaknesses. Watching someone completely unfamiliar with your store is often humbling and always insightful.

When conducting user testing, you want to assign the participants at least three tasks:

• A broad task.For example, “Find a video game you like and would consider buying.”
• 一个特定的任务。For example, “Find a Nintendo Switch game between $40 and $50 and add it to your cart.”
• A funnel completion.For example, “Purchase something you would like to buy.”

If you’re using a tool to conduct user testing, chances are you will gain access to session recordings as testers finish the assigned tasks. They will read the tasks and complete the tasks themselves, without your intervention.

If you’re hosting a live user testing session, focus on watching and listening carefully. In advance, check with a colleague to ensure the instructions are crystal clear. Avoid asking for personal opinions or answering questions about the tasks during the session.

5. Session replays

Session replays are similar to user testing, but you’re dealing with real people with real money who really have intent to buy your products. You’ll be able to watch as your actual visitors navigate your site, minus the narration, which can be helpful to audit a potential broken sales funnel.

What do they have trouble finding? Where do they frequently pause? Where do they get frustrated? Where do they seem confused? Where do they give up and leave?

Session replays require excellent note-taking skills. As you watch replay after replay, you’ll want to take notes on any patterns you recognize and obvious flaws uncovered.

Watching your visitors, who are less familiar and comfortable with your store, will be eye-opening.

Pay special attention to pages and points of the conversion funnel that your visitors consistently stumble over or struggle with. Minor improvements here and there can add up significantly for your overall marketing efforts.

Reduce your sales funnel leaks

There will always be funnel leakage. Finding and serving the right customers is hard. At times, you might feel that you plug one hole just to spot two more. But the more effectively you can plug the leaks as you spot them, the more optimized each trip to the well will be.

Don’t tire yourself out by running home from the well (i.e., sending more and more visitors into a leaky funnel). Instead, focus on keeping your bucket in good working condition. It will require constant maintenance, but it will be worth it to keep that water (money) in the bucket, and ultimately lead to more conversions, customers, and sales.

Illustration by Luca D'Urbino

---

准备好建立你的在线商店吗?欧宝体育官网入口首页OB欧宝娱乐APP开始你的自由trial of Shopify—no credit card required.

Email address Start free trial