15 Steps to the Ultimate Lead Capture Landing Page

If you’re looking to build a lead capture landing page, it’s not just a matter of slapping a form on a page and driving traffic to it. Well, you could, but it wouldn’t perform well.

Instead, follow these 15 steps and create a high performance landing page designed specifically for gathering leads.

Read on and learn how to build the ultimate lead capture form, an example of which will be uncovered right at the end.

Step 1. Define your target market & campaign goal

Who are you marketing to?

Before you start building anything you need to know who you are marketing to. The chances are you’re creating a campaign for one of the following targets:

1. Existing customers: You might think that having a customer means you don’t need to continue marketing to them. Sometimes a customer is inactive so your goal is reengagement, through exposure to new content or product demos that may turn them from a fence sitter back into a real customer.
2. Leads in your email or CRM lists: Once you’ve established a list of leads, your goal should be to nurture them with more content and events to turn them into customers. Perhaps you get them to register for an e-course using autoresponders to drip content into their inboxes, slowly building a rapport that eventually moves from a soft sell content marketing strategy, to a request to sign up (potentially with a special offer).
3. Prospects: This is the purest form of lead capture, where you are chasing entirely new prospects through the use of content marketing.

Lead form tip: If you are marketing to customers or leads, pre-populate as many of your lead capture landing page form fields as you can, to reduce the friction involved in filling out the form, and to create a sense of personalization.

Step 2. Define & create a relevant reward

Are you giving something appropriate away in exchange for personal data?

Having established your target audience, you need to figure out what to offer them in order to get their important personal or business data.

Everyone likes to be rewarded, getting free stuff is awesome. But with lead gen there’s a little work to do first, and it gets pretty personal. Typically the very minimum a prospect, lead or customer will have to part with is an email address (covered in step #3).

Before thinking about what you’re asking for in exchange for your content, you need to define the content, and make it relevant to your target’s needs. Here are some of the things you can create:

Digital content to give away

There is a ton of digital content you can give away, first we’ll list them, then explain how they should be used differently for prospects, leads and customers.

• An ecourse delivered over a period of time
• Report/whitepaper with important industry stats
• Ebooks – comprehensive guides to different aspects of your business
• A newsletter with tips related to your area of subject matter expertise
• A podcast – useful for people who like to listen during a commute or workout
• Checklists/scorecards – good for people that need to see how well they are doing and produce a to-do list of things to improve
• Webinars – live sessions, often with Q&A with experts and special guest presenters
• Presentations – slides you can incorporate into your own presentations
• Consultation – this is when someone requests your time to explore if they want to use your product or service

Prospects

Prospects are people you haven’t come into contact with yet, so they need extra reasons to believe in what you’re giving away. They need a softer sell, usually educational material that doesn’t push your product, like an ebook.

Leads

Leads are prospects who have converted by already consuming some of your content. These are the people you really want to convert into customers. As such they should be receiving content with signup CTAs and links to register for product demos.

Customers

Customers come in two forms: active and inactive. Your goals here are to keep active customers happy with instructional material that makes their life easier, and then re-engage your inactive customers with special offers, and updates on new features that could reactivate them. Customers are prime candidates for newsletters and ecourses.

---

Webinar tips: Have a webinar calendar to allow people to sign up in advance. List each session as beginner, intermediate or advance – much like you would with a yoga calendar. And remember to inform people on your landing pages that they will receive a recording of the webinar the next day if they are unable to attend – this will increase registration significantly.
Crucial to your success with lead gen is to balance the reward (the size of the prize) with the friction involved in getting it. This is where your choice of form fields come into play…

Step 3. Define the required form fields

Do you really need to ask for a fax number?

What’s an appropriate number of form fields to ask for in exchange for your content? Ideally, you would test this to find out what number of fields converts the best, while still getting the data you need to successfully segment and re-market to them in the future.
Here’s a breakdown of the content listed in the last section with an appropriate level of information you should ask for (note these are simple guidelines – every situation will be slightly different).

Email only

• Presentation slides
• Checklist/scorecard
• Podcast

Name and email

For a more personalized experience over a period of time.

• Ebook
• Newsletter
• Ecourse
• Report/whitepaper

Name, email and company information

Company information might include: size of company and phone number. Often used for longer sales cycle items.

• Webinar
• Consultation
• Product demo

Note, you could ask for a lot more if you want to, but realistically, an email is all you really need for most things.
Form tip: Scale how much information you ask from prospect, lead, customer (prospect being the least, and customer the most).

Step 4. Craft a headline to match your promotional ad source

Does your headline match the ad that was clicked on?

Message match is the term used to describe the connection between the call-to-action (CTA) of an advertisement and the headline of the destination page.
There are two reasons why this is important:

1. The human factor: When someone clicks a CTA they have a level of intent that must be matched when they arrive at their destination. For example if they clicked on an ad for used pickup trucks, and they arrived at the homepage of a used car business where there was no direct mention of pickup trucks. Immediately the message match has been broken and they will feel like they have made a bad click. The end result of this is that they either get lost trying to find what they came for, or ,ore likely they just hit the back button and leave.
2. The machine factor: It’s not just people that get affected by poor message match. If you are doing paid advertising such as Google AdWords, there is a bot that reads your target page to see if there is a connection to the keywords in the ad. In the previous example, you’d get a lower Google Quality Score for your ad, and consequently a higher price for your ad.

An example of good message match

Here’s an example from an Unbounce banner CTA that sits at the end of our blog posts. Read the text on the button, as this will be the last thing in the mind of a visitor when they click it.

Now consider the headline in the destination landing page, and how closely it matches that of the CTA:

Always pay close attention to your message and you’ll increase your conversion rates by virtue of more people staying on a page they asked to visit.

Step 5. Establish relevant trust elements

Is your lead capture page trustworthy?

Trust is key to making people believe you won’t be spamming them after getting your email. Social proof is a key element here as we’ll see in the list of trust element you can include on your page:

Trust elements

• Registrant count: If you are running a webinar, list the number of registrants to encourage a herd mentality (if that many people are going, it must be good). Note: only start listing the number when it gets to an impressive count.
• Download count: Similarly, show how many people have downloaded your ebook.
• Share count: Showing the social share numbers shows how many people find your content valuable enough to tell others about it. The best place to put this is on your confirmation page when they’ve already expressed interest.
• Anti-spam statement: Put a simple statement next to the email form field that says you won’t spam them ever.
• Testimonials: Customer testimonials show that others have made the effort to report back on the quality of your content. The best kind are video testimonials, followed by written ones with a photo and official title. Never use a fake testimonial – they’re usually easier to spot that a wig.
• Customer logos: If you have an impressive client list, take advantage of this by prominently showing them on your page.
• Media mentions: You’ve seen these before. It’s the list of logos from CNN, NBC, FOX News, TechCrunch etc. If you’ve been lucky enough to have been mentioned in big publications or on TV, make sure to show off.
• Security badges: If you’re asking for sensitive data, include these to show that you are using security best practices, such as a secure server. Examples would be McAfee or Symantec.
• Content previews: Leverage people’s desire for “Try before you buy” by including a preview of your content such as a chapter of your ebook. Amazon has perfected this with their “Look Inside” concept.
• Past content examples: If you are asking someone to sign up for a newsletter, show them an archived example so they know what they are signing up for.

Step 6. Write a short persuasive and descriptive intro paragraph

Make it clear what people should do to start understanding your service.

Your opening paragraph should be a short and succinct extension of your headline, which serves as an introduction to some bullet points (coming up next) describing the product or service you’re promoting.
Let’s look at an example from a webinar registration:

---

Join us at 2pm EST for a special webinar about lead gen landing pages, featuring [insert star name], hosted by [insert another star name]. We’ll be talking about best practices for generating leads using webinars (how meta), and we’ll be covering the following:
Insert bullet points…

• Point 1
• Point 2
• Point 3

The star power and time limited component give it a persuasive edge, while the details and bullet points give it the descriptive clarity it needs to answer your questions and convince you of the benefits of attending.

Step 7. Bullet point the key benefits of your solution

Bullet points are easier to scan than a paragraph of text

Following directly from the last point (no pun intended), given how little time most people spend on most pages – you have maybe 5 seconds to garner their interest. Most of this is done via your headline grabbing their attention, but assuming this has already happened, you need to dig into some simple benefit statements rather than ramble on in a long paragraph like this one. If you can turn your text into about 3-5 bullet points each explaining why they should fill in your form, then your lead gen page will be more effective. As an exercise, let’s break down whaty I’ve just said into bullets:

• You have 5 seconds to convince someone to stay
• Don’t use a rambling paragraph
• Use 3-5 bullets points
• Focus on the benefits
• Explain why someone should complete your lead gen form

See how much easier that was to read?
Point made.

Step 8. Decide on image/video requirements

Which medium best communicates your product or service?

This is a classic A/B test – whether to use images or video on your page, or none at all. The decision often comes down to the purpose of the page. Let’s run through some of the rewards you are giving away (from step #2) and see what would be most appropriate:

• Ebook: An image showing the cover, and the authors is the most appropriate choice here.
• Ecourse: A personal video describing what the course is about is typical, and most likely the most persuasive method.
• Webinar: Photos of the participants are the most powerful choice here.
• Podcast: Here you’d want a short audio clip.
• Slide presentation: An embedded Slideshare or lightbox gallery presentation with a sample of the full version is a great idea here.
• Online product demo: A video showing the context of use (the product in action) is the best way to showcase a product.

Let’s look at a couple of examples:

Ebook download

Webinar registration

Click image for full size view

Notice how each of the examples uses only the required information in the form, uses a short intro paragraph, bullet points and relevant images.

Step 9. Write an actionable CTA

Make your CTA desirable so people want to click it

This one’s conversion 101. Your CTA should describe exactly what will happen when you click on it. We just talked about message match, so you should understand the importance of a descriptive CTA.
To make it more actionable, use persuasive copy, that has some urgency or benefit statements in it. Even simple words like “Get”, “Try”, “Now” or “Today” can work.
Here’s an example of good and bad CTAs:

A good CTA

Look how clear the description on this button is. Both lines are very actionable and descriptive.

A bad CTA

This is more of a rule to follow. The absolute worst thing you can ever write on your buttons is the dreaded “Submit” word as it describes nothing about what’s going to happen when you click on it. If you’re using this on your buttons, go and change them right now!

Step 10. Create urgency factors

Before the candle burns out…

The key to creating a sense of urgency is to create the illusion of a pain point in the mind of the visitor to your page. To understand how this works, let’s first look at a couple of examples from the world of ecommerce.

Amazon delivery deadline

Amazon preys on the fear of not getting Christmas presents delivered on time (who wants crying kids on Christmas morning?) They do this by helpfully telling you the last day that you should order to guarantee delivery by the date in question. They’ve taken this even further (as shown in the example below) where they are applying urgency to any date – no doubt to capture last minute present buying for birthdays which can occur on any given date.

Expedia seats limitations

Expedia uses a live seat count for their flights, to encourage immediate bookings. Looking at the best flight for your journey and seeing that there are only 2 seats left is a great way to add urgency to the booking process. Note that they only do this when there are a couple of seats left – not when there are 200 left.

I have no idea if there is actually one seat left, or if this is just a ply to encourage bookings. It works extremely well though – from personal experience.

So how do we go about applying these principles to our lead gen landing pages?
We can apply both of these principles to our lead gen pages as follows:

Ebook/report – free for a limited time

For an ebook download, you can say that it’s free for a limited time only, and that after a certain date it will cost $14.95 for example. This will increase the number of downloads before the date you choose – which of course can be continuously extended.

Webinar – limited seats

If you are running a webinar, you can increase both the urgency and exclusivity of the session by stating how many virtual seats are still available. For example: register now – only 20 of 100 spots left.
Essentially, anything you can do to create a scenario of urgency through utilizing a pain point, you’ll increase your lead gen form conversions.

Step 11. Apply the principles of Conversion Centered Design (CCD) to your form

Apply the principles of conversion centered design to perfect your form. In this instance, encapsulation is demonstrated.

CCD is a discipline targeted specifically at achieving the goals of a business, from the standpoint that the primary purpose of any page is to guide the visitor toward one specific action. It’s an approach that uses persuasion and psychological devices to convince a visitor to convert. Landing pages sit at the heart of CCD, used to focus the attention of a prospect toward a single objective or conversion goal.
In the examples following, we take a basic lead capture landing page, and transform it using 3 of the core principles of CCD: encapsulation, color contrast, and directional cues. The end result being a stronger lead gen form designed for conversion.

Using encapsulation: Notice how the form stands out more in the version on the right, due to the use of an encapsulation container. This is most often done simply by placing the form in a containing box to provide a contrasting background.

Using color & contrast: Now the form is really starting to pop. Notice how there are two primary areas of the form that are brought forward by the use of color & contrast: the form header and the CTA. the reason for using the same contrasting color for both is to provide a sense of correlation. The header should contain pertinent information that describes what you are getting by submitting the form, and the benefit of doing so. For example: “Download our free ebook to master the art of conversion.” Using the same color as the CTA will naturally allow your eye to follow the trail down to the CTA after reading the contents of the header.

Using directional cues: Notice the use of two arrows in the example on the right. By taking the previous version and adding some extra visual persuasion to the form, the first arrow brings your attention from the introductory copy to the form header (which as described above, should contain the description of the purpose of your form) and a second arrow from the form header down to the CTA.

We now have a lead capture form based on the principles of conversion centered design.

Step 12. Architect for the 6-point punch

Architect your page for the 6-point punch

 

You’re probably wondering what the 6-point punch is. It’s a concept based on the primary elements a successful lead gen page requires to be effective. Essentially, it acts like a checklist of the important elements already discussed, as a reminder that to be really conversion focused, you need to create a page with all of these aspects working in concert together.

1. Headline with strong message match
   Here the ad that was clicked should be a very close match to the headline as discussed in step #5.
2. Intro with bullets
   Remember to keep your intro short, and simplified by the use of bullet points.
3. An appropriate media choice
   In the example below an ebook image is shown.
4. Conversion Centered Design influenced lead gen form
   The form area shows all of the conversion centered design principles describe in the last step.
5. Urgency factors
   Remember to add some kind of time, price or space limited statement to your page.
6. Trust factors
   In the example below, testimonials are added to the bottom of the page. A good addition to this would be a preview of the ebook to let the visitor know what they are going to download.

Here’s a good example of an ebook download landing page with some of these principles in action. Later on you’ll see how it could be improved via a collaborative A/B testing exercise.

Click image for full size view

Step 13. Decide what you want to do post conversion

What do you want people to do after they convert?

Despite having a well targeted and relevant reward on your page, you could be generating more paying customers from your prospects – and I’m going to show you how.

It’s important to think about what will happen after you generate a new lead, and what you want them to do now that you have their attention. You want to take advantage of that magic moment directly after the customer says “Yes, I like you” to keep them in your sphere of influence while they’re “in the mood”. You can do this by focussing their attention on the first thing they see after submitting your lead gen form – the confirmation page.
The lead generation confirmation page could be considered the digital equivalent of a supermarket candy shelf. Admit it, you’ve fallen victim to those shelves of sweet goodness stacked by the checkout as you wait in line – aimlessly throwing crap you don’t need into your basket.
So what should you ask them to do on your confirmation page? Here are 5 quick tips that you can implement now:

1. Social: Ask them to follow you on a social media network (Twitter, Facebook) – or share your page.
2. Invite them to a webinar: Make sure the subject matter is closely related to the goals of the leads you’re capturing – and of the landing page they were just on.
3. Give them a discount coupon (or code): If you give people a discount when they’re not asking for it and they don’t have to give you anything in return (after all, they just completed your lead gen form) – they’ll get a great buzz from being treated so well.
4. Send them to an important “What To Do Next” page: Many people enjoy a guided experience. Tell them what to do, and they’ll often do it. Use only a single link for this and send them to a “Greatest Hits” or “Top 10” list page that shows off the very best content and information you have to offer.
5. Ask them to subscribe to a newsletter: Newsletters (or a drip campaign if you want to push them through a progressive story arc) are a powerful permission marketing tool that keeps you in your leads stream of consciousness for an extended period of time. The confirmation page is a good time to ask if someone would like to participate if you can automatically transfer their details to the new newsletter email list without asking for more data.

Step 14. Create a wireframe for your page

Create a rough wireframe to figure out your page element placement.

Before you start designing your page, you should grab a marker and use a whiteboard to lay out a wireframe for your page. There are a couple of reasons for this: it saves you design time, by getting the layout right before you start building a page you might need to deconstruct and re-build, it also allows you to run a few initial tests using your sketch.
Before you run the tests I’ll mention below, make sure your wireframe is quite detailed and includes some real copy and some CCD elements such as color and contrast etc.

The 5-second test

This is a fun one to do with a whiteboard (or piece of paper hanging on the wall if you decided to sketch it that way or printed out a page from a wireframing tool like Balsamiq.
Here’s how it works. Get a lineup of people who’ve not seen the page before, and one at a time, position them right in front of the page with their eyes closed. Then ask them to open their eyes and look at the page. After 5 seconds, tell them to close their eyes again and tell you what the page is about. If they can’t tell you after 5 seconds of exposure (the typical amount of time a visitor will stay if your page isn’t clear), then your core value proposition (headline) isn’t clear enough. If this happens repeatedly, rewrite and start again. Do this until you have a very clear page.

The 6-foot test

For this one, you need to stand 6ft away from the sketch and see what the most dominant areas on the page are. If the CTA doesn’t stand out enough, you should consider adding more contrast, size and some whitespace around it. Similarly, if the headline is buried, apply the same concepts to make it more easily readable.

---

Once you’re satisfied with your page, it’s time to do the final design and get ready for launch and A/B testing.

Step 15. A/B testing your lead capture landing page

A/B test your landing page for maximum conversions.

Here’s where the real fun starts, running an A/B test on your page. But first you need to publish it and start driving traffic to it (from PPC, social, email, display etc.). Quick plug: Unbounce makes it super simple to build, publish and A/B test your pages without any help from I.T. – okay, pitch over.
When you start seeing conversion results from your landing page, you’ll undoubtedly come to the correct conclusion that it could be better – every page can be. This signals the time to run an A/B test. It’s also when the classic question of what to test comes up.
Before jumping into testing, you need to know a few of the fundamentals involved in the A/B testing process, and then some techniques for figuring out what to test.

---

Gathering insight for a test

Before you know why you are going to run a test, you need to get some actionable insight into what could make your test effective. One of the best ways to do this is by using visitor feedback. This can be done in a number of ways:

• Use services like Qualaroo to add a survey to your pages to ask people questions about their experience.
• Using live chat on your pages so that you can talk to people at the point of conversion and figure out why they are sitting on the fence, or what they need to become a converted customer. Unbounce recommend services like Olark and ClickDesk
• Sending an email to your customers to find out what it was that made them sign up.
• Use online usability tools like UserTesting.com or Loop11 to see where people are dropping off.
• Connect heatmap software like Crazy Egg to your page to see where the most clicked and focused areas of the page are.
• Have a brainstorm with people from different departments in your company

Why are you testing?

A big mistake is trying to test something without a real purpose. This is where a test hypothesis comes in – a statement of what you are going to test and your theory behind why it will be a success. As an example:

“The page does not have a clear call-to-action, and prospects spend too long trying to understand what to do next. Adding a large orange button right under the main benefits will help them identify the CTA and perform our desired action.”

Once you have a hypothesis you’re in a better position to create a test page to compete against your original page in an A/B test.

How to test

Now that you’re ready to run a test, you need to follow a few rules to ensure your experiment is clean:

• Each page in your test should receive at least 100 unique visitors
• The test should last at least a week to account for different daily behaviors
• The statistical significance of the experiment should be over 95% to remove the potential that your results are based on chance. This is a measure of confidence that your experiment is valid and can be ended, knowing that the results are trustworthy.

---

Brainstorming

Educated with how to gather feedback and run an effective test, we’re going to focus on one element of gathering insight – the group brainstorm.

Include people from different departments to get a range of opinion.

There are a lot of feedback mechanisms listed above – all of which are effective, but armed with the data from these techniques, you will be tempted to go it alone and try to figure out how to implement change. This will work most of the time, but having more than one opinion is a great way to establish new ideas and priorities.
Here’s a basic overview of the brainstorm process.

1. Draw a giant version of your landing page on a whiteboard (or just print it out as big as you can).
2. Gather a representative from several departments/disciplines around the company (marketing, design, customer service, management, software development) as your brainstorm participants.
3. Provide everyone with a block of stickies and a sharpie pen.
4. Write down a list of the primary page elements that could be tested to start the process with.
5. Ask everyone to come up with ideas to test as you walk through the list of elements. For each idea, they must write it down on a stickie and verbalize the idea as it’s passed to you (as the facilitator) and stuck on the page in the relevant position.
6. Open it up to big picture ideas with no restrictions.
7. Once you have exhausted all of the ideas, have everyone come and vote (by checking a stickie) on their top 3 ideas.
8. Create a prioritized list of which ideas to test, ordered by the number of votes.
9. START TESTING…
10. A final point. Never stop testing. Test, Rinse & Repeat. Because as I mentioned earlier, every page can be better.

As a result of a session like this, imagine our example from earlier, modified based on the principles we’ve learned along the way and the ideas gathered from a group brainstorm.

As you can see from this version, it now includes all of the elements of a great lead capture landing page:

• Reward: The ebook
• Appropriate form fields: Name and email address only – removing the unnecessary company and phone number
• A short intro
• 3 bullet points
• Relevant media choice: An image of the ebook, along with author photo & bio below it
• Trust factors: A testimonial, ebook preview, and privacy policy
• Urgency: A time-limited free price offer
• CCD elements: Directional cues, encapsulation and color contrast applied to the form area
• A descriptive CTA

http://unbounce.com/landing-pages/ultimate-lead-capture-landing-page/

What is Split-Path Testing (and When Should You Use It)?

 

One of the testing strategies is experimenting with the path people take towards their final conversion/purchase. This is referred to as a split-path test or alternative path test.

This could give you the needed lift when nothing seems to move the needle anymore.

What Is a Split-Path Test?

A split-path test is defined as “a type of A/B/n test where, instead of just altering a single page, you change multiple sequential pages.”

Split-path testing sends users down a different path instead of showing different page variations. It includes things like multi-step checkout paths, multi-page forms, and product recommendation wizards. As Chris Goward explained in You Should Test That!:

 

Chris Goward:

“Alternative path testing allows you to maintain consistency of design elements throughout the path and change the number or sequence of steps. In a simple example, if you were to test a checkout path and wanted to test the button color in the car, you would change the buttons on all pages in the checkout to match. Alternatively, testing individual pages would produce inconsistency throughout the path that could decrease sales and outweigh any improvement from the change.”

While split-path testing isn’t talked about as often as your regular A/B/n tests about button colors, it isn’t a new concept. Bryan Eisenberg and John Quarto von Tivadar wrote about this a while ago in Always Be Testing:

“This is different in that you’re testing the performance of grouped pages against other grouped pages. For example, you could test a checkout process by splitting it into two variations, one with four steps (or pages) and another with only three steps.”

How User Flow Affects Conversions

A major factor affecting your conversions is user flow. It’s the path a user follows through your website interface to complete a task (make a reservation, purchase a product, subscribe to something). It’s also called user journey or conversion path.

Non-linear checkout flow (Image Source)

Here’s an example of a typical conversion path you might have:

Facebook update –> Landing page –> Shopping cart sequence –> Payment funnel –> Pre-bill order confirmation page –> Billing page –> Confirmation page

Split-path testing is one way of optimizing your conversion path. As for why it’s important to do so, here’s how HubSpot explains it:

Just as it’s vital to make sure you’re maximizing the output of each individual page on your website, it’s incredibly important to understand how all of your pages work together. Which pages of your website make you the most money? Which ones generate the most leads? Which ones are preventing your prospects from advancing through your sales funnel? On a relay team, you can put together the greatest four runners in the world. But if they drop the baton at every handoff, you’ll never finish a race. Similarly, if a reader moves from one page to the next and gets confused, the only button they are likely to click is “back”.

Though Andrew Anderson, Head of Optimization at Malwarebytes, explains that assumptions about the drop-off between the user flow can limit the scope of your testing program:

Andrew Anderson:

“Conversely there are times where pages not matching are beneficial. It can be a huge benefit and also a massive risk to jump directly into assuming a full flow. It doesn’t have to at the start of the journey, you can test individual components and then view the impact of the system as a whole by doing a full path test. There are times that the “hand off” can be losing efficiency, but assuming so can be extremely impact limiting. Testing individual components and the whole allows you to see how much those assumptions about consistency really matter.”

Testing alternative conversion paths can open up new opportunities to boost conversions that cannot be obtained by simply changing page elements.

How Split-Path Testing Can Optimize Conversion Paths

Setting up split-path tests is similar to your regular old A/B/n tests in that you’re tracking the same goal. Each group of pages is treated as variation to be tested against the control, and each variation has the same conversion goal page – probably the order confirmation or thank you page.

You need to analyze the data the same way as a regular A/B test, with the same statistical rigor.

In addition, be careful with selection bias in split-path testing. As Andrew Anderson explains, many people think that you should only split-path test with new users, which creates a biased population:

 

Andrew Anderson:

“There is an extremely naive view that you should only test split paths on new users only as the complete change is supposed to be a major problem for existing users. This however assumes that those existing users are not people you care about or hope to make money from in the future, because eliminating them creates an extremely biased population and can invalidate your results.”

Remember also that site navigation isn’t usually linear. That means that gravity doesn’t pull visitors very cleanly down your funnel, no matter how much you wish that were the case. Linear assumptions of data is common, and it’s a bias that can be detrimental to any of your optimization efforts.. As Eisenberg and Quarto-vonTivadar explained:

“Although this can be an easy way to test strongly linear scenarios, you should keep in mind that few visitors navigate your site in a truly linear fashion (even though you might wish they did for analytics-tracking purposes!). Persuasion instead tends to occur along nonlinear paths, although that is a subject far beyond the scope of this book. Further, the longer the “click distance” between testing page and conversion page, the greater the likelihood that some visitors will wander off to other portions of your site, which Website Optimizer may record as a failed conversion for testing purposes, even though such a result doesn’t reveal visitor intent.”

For that reason, split-path tests are most appropriate on sections of your site that do follow a linear path, like your checkout flow and registration forms.

Image Source

Split-Path Testing is An Innovative Test

Split-path testing is considered an ‘innovative test’ – as opposed to an incremental test, split-path testing offers offers the opportunity for much greater lifts. That said, it’s riskier as well for multiple reasons.

As Andrew Anderson mentioned, “Perception is not always reality however as you may just be exponentially adding bias on top of bias and allowing yourself to take more time and opening yourself up to even more local maxima.”

If you’ve hit a glass ceiling (i.e. a local maximum) and need to do something innovative and creative to bust out of it, split-path testing is a possible but highly cost inefficient solution.

Let’s Talk About The Local Maximum

What is the local maximum?

It’s when, after a progressive period of A/B/n testing, your gains start to slow down. You’re experiencing diminishing returns even though you’re testing a bunch of iterations. It’s frustrating, but it happens to everyone.

What can you do about it?

There are many solutions, but most of them center on shifting your paradigm and getting a bit creative. In other words, swinging for the fences with some innovative tests, trying to change customer behavior/experience to increase conversions.

Though, as Andrew Anderson explains, split-path testing is not a silver bullet solution to the local maximum problem:

 

Andrew Anderson:

“This is almost always a sign of poor test discipline and a lack of creativity, and sometimes thinking about a different problem, the entire path, helps give temporary relief to that problem. That being said it is never a full solution to the problems of bias and storytelling.”

Split-Path Testing Subscription Pathways

Marketing Experiments wrote about a few tests they ran optimizing the subscription funnel of some large publications. They performed split-path testing to simplify the funnels. Here’s what the first test looked like:

Image Source

In this case, reducing the steps from 9 to 3 increased conversion rates by about 300%. As they wrote, “Reducing the number of steps is probably the single most-significant improvement you can make to your subscription path.“

In their second example, they employed a similar strategy: reduce steps to subscription. This is the flow that they started with (resulting in zero registrations in this case):

Image Source

They decided to chop a few of those pages off and pair it with an email capture form on the initial offer page for the next test:

Image Source

Finally, they removed another page from the sequence while retaining the same offer page:

Image Source

Of course, their sample size is too small to draw any conclusions between the four page and three page examples, but it seems they are moving the flow in the right direction.

Add Registration Steps

You can also take the opposite strategy and add in more steps to the checkout funnel. Here’s Andrew Anderson’s take:

Andrew Anderson:

“Keep in mind that adding more pages can also increase performance. A very large customer I worked with had a very strong inverse correlation between the pages a user viewed and the revenue they spent (more pages meant more money). When you swing big, it is even more vital that you challenge your assumptions, meaning that the cost goes up exceptionally high, as does the possible reward.”

So, let’s say you’re running a SaaS startup and want to increase the amount of paid signups you have. Your optimization efforts have leveled out – value proposition tweaking is no longer bringing any results. So you think that perhaps you can increase quality leads by changing the checkout flow itself.

You go from: Home page → Sign up → Confirmation page

To: Home page → Features → Sign up → Confirmation page

You hypothesize that if customers are to view your features page before being asked for their credit card, they’ll be more likely to sign up. Perhaps they’ll be higher quality leads. They may have a lower attrition rate. So, you can split-path test this to be sure.

Change The Order of Steps in a Funnel

Sometimes, changing the order of the flow but keeping the same content can provide a lift (Andrew estimated that he’s seen anywhere from 5-45% lifts and for some reason the scale tends to go up with larger sites.”

So imagine you’re selling a subscription product and your checkout flow looks like this:

Home page → Sign up → Payment Page → Create Account → Thank You

You could try changing the order to switch the payment and the create account pages, like this:

Home page → Sign Up → Create Account → Payment Page → Thank You

Similarly, if you ran an eCommerce site, you could experiment with running the billing address before the shipping, or vice versa. You get the picture.

Andrew Anderson told me about a Malwarebytes example where they ran a test changing their entire shopping cart user flow, going from multiple pages to different orders and page layouts. They tested a large variety of examples and found that moving from a 3 page flow to a one page flow and introducing the purchaser information fields much earlier results in a dramatic uptick in overall user RPV.

Test Low Touch vs High Touch Paths

Another split-path test one could run is sending users down a fully online purchase path or a live representative quote or demo path.

Picture that you’re running a SaaS company that sells form analytics software. You send 50% of traffic down your traditional funnel where they can pay online and get started on their own, and you send 50% down a “get a quote” or “get a demo” path and have a sales rep take it from there.

In this scenario, you need to factor in the increased cost that accommodates the increased workload for your sales team. Here’s how Andrew Anderson put it:

 

Andrew Anderson:

“Keep in mind that at this point you need to measure true net RPV as the cost dramatically goes up at that point. That being said I worked once with a financial lead gen company that tested this and had a 24% net RPV gain by forcing people to talk. Fewer leads but much higher cost. Conversely at Malwarebytes we have found the exact opposite. The more we try to move people away from talking to our B2B reps, the less money we make and it is dramatic (and also a -.81 correlation).”

In general, the sales rep/quote approach tends to work better with more complicated or more expensive products (this is the norm for many enterprise software products).

In this case, as in the others, there’s always a chance it won’t work and you’ll have spent a large opportunity cost. In addition, mind your sales cycles and your churn later on down the line with either path.

Warning: Heavy Development Resources Might Be Required

Split-path testing has promise of a huge lift but also the possibility of a huge waste of time.

For instance, coding these variations is much more costly and time consuming than testing page elements. Developers need to code outside of their testing tool and usually run split url tests to accomplish this (could also run multi-page tests for some checkout flows, but that’s theoretically different from the type of split-path test we’re talking about).

 

 

Dave Gowans, Head of Conversion at Conversion.com, gave the following example:

 

Dave Gowans:

“The biggest challenge for Alternative Path testing is the amount of development work that goes into the tests. Whether you build an alternative version of the funnel in the backend or using a testing tool, the development effort multiplies with the number of pages involved, so you could end up spending many days of time, even for a relatively small design change.

For one client, an online pharmacy, we spent hundreds of hours building a complete re-skin of their funnel to match a new design on a single page. It showed almost no impact on the overall conversion rate. That time could have been much better spent on other, smaller, tests delivering more impact.”

And as Andrew Anderson said, “You almost always need a unique user flow path or a cookie based system to control server side implementation. In many cases you will not even use the testing tool for data tracking.” It’s a minor issue, but one to consider with split-path testing.

Just imagine that you need to have a developer work on your variation full time for two weeks. How much does that cost you in time and opportunity cost (what else could they have built in that time)? Andrews says you can mitigate some of this cost by never wasting the time on 1 alternative experience.

According to him, “2 wasted weeks as a 10% success rate is far worse than 6 weeks at a 75% success rate and a 50-100% increase in expected scale of outcome. If you are going to go this big and take the time, the math always shows that you need to do even more experiences. The cost does not exponentially grow but the expected outcome does.”

Another issue: the only data you have is current funnel performance, meaning that existing analytics (correlative data) is basically useless. It tells you what did happen, no what should happen.

A Possible Split-Path Testing Strategy

Being that there is great risk with split-path testing, here’s a possible strategy to consider:

Start by improving your existing funnel. Use the ResearchXL model to figure out all friction, distraction, and other issues. When you begin to hit diminishing returns, then it may be time to experiment with split-path testing. By testing whole flows you can actually get better data, in that you’re able to find real problems and distinguish them from data anomalies and/or biased perceptions.

Finally, when considering split-path testing, keep in mind the PIE framework. With this type of test, you’re balancing the Potential with the Ease of Implementation, which is not low. Therefore, with the higher risk associated, you always need to increase the expected outcome. Dave Gowans from Conversion.com echoed this sentiment:

 

Dave Gowans:
“Our advice is to always carefully evaluate the impact and ease of every test – putting the effort into spending time with your developers, understanding the most effective way to make the change and how many hours it will take. This small investment of time upfront is easily justified by the time you’ll save by not embarking on a infeasible project. Look closely at the impact of your proposed test (both the upside and the potential downside of not testing a change). Does it really make enough difference to test it?”

Conclusion

Split-path testing runs on the same statistical models as a regular A/B/n test, but is different in that it fundamentally changes a user experience.

It is both a potential huge win but also an opportunity to get even more caught up in your own biases. It is the definition of swinging for the fences, but if you do it with discipline and with an understanding of opportunity cost and efficiency, it can be a huge win for your entire organization.

This could be good, in that you can achieve bigger uplifts, or this could be bad in that you spend a lot of money and time and development resources on a test that doesn’t move the needle at all.

 

http://conversionxl.com/split-path-testing/

25 Simple Ways to Grow Your Email List

                                        

 

I have some bad news: Your email marketing database degrades by about 22.5% every year. Your contacts' email addresses change as they move from one company to another, opt-out of your email communication, or abandon that old AOL address they only use to fill out forms on websites.

As a marketer, it's your job to make sure you're constantly adding fresh contacts to your email marketing campaigns so you can keep your numbers moving up and to the right.