The Definitive Guide To Conversion Optimization

The Definitive Guide To Conversion Optimization

Written by Neil Patel & Joseph Putman

Chapter One

What Is Conversion Rate Optimization?

When it comes to internet marketing, you can generate more revenue in one of two ways. You can drive additional traffic to your site in order to increase sales, or you can improve the effectiveness of your site to boost sales with the same amount of traffic you’re currently receiving. Conversion rate optimization (CRO) focuses on the latter.

With conversion rate optimization, you evaluate your website’s sales funnel to identify ways you can improve your site in order to get a higher percentage of people to sign up for your product or service. Next, you come up with a hypothesis to test, and then you create a new version of a web page or landing page to test against your current version to see which one is more effective at getting visitors to sign up or buy. In the end, you implement the variation that convinces the highest percentage of people to buy what you’re selling.

What may come as a surprise is that both small and big changes to copy, layout, and design can have a big impact on the number of people who sign up for your service or buy your product. Switch Video, for example, found that changing a single word in a call-to-action button copy increased qualified leads generated from their homepage by 221%. In another test, Performable, a company acquired by Hubspot in 2011, was able to increase click-throughs 21% by using a red button instead of a green one.

Performable was able to increase click-throughs 21% by using a red call-to-action button instead of a green one

Results like this make it obvious why conversion rate optimization optimization is so valuable to companies. Even small changes can have a significant impact on the bottom line.

However, testing is the key factor when it comes to CRO. Without testing, you’re left relying on your gut instinct to determine what will be more effective, but once you test, you know right away whether a change leads to an increase or decrease in conversions. It takes the guesswork out of knowing what you should say and how you should design a new web page.

The number of tests you can eventually run are endless. You can test different headlines, new value propositions, varying button colors, different call-to-action copy, and much more. Each change has the opportunity to impact conversions, and small wins add up over time.

Why Is It So Valuable?

As you can probably imagine, conversion rate optimization is important because it generates more sales for the same amount of traffic you’re currently receiving. Instead of sinking additional money into PPC ads or other digital marketing methods to drive more traffic, you’re more efficiently convert your current traffic into leads or sales. And if you do decide to drive more traffic, your CRO improvements mean you’ll get more out of your increased marketing efforts.

Consider a hypothetical scenario where you’re the owner of a software-as-a-service (SaaS) business. Before optimizing your site, you convince 5% of people to sign up for your service which costs $50 per month. That works out to 50 new customers per every 1,000 visitors and $2,500 in monthly revenue. Now, suppose you optimize your site and raise your conversion rate from 5% to 7.5%. At this rate, 1,000 visitors turns into 75 customers and $3,750 in monthly sales.

Do you see what happened? You didn’t change anything else about your business, but after making your conversion funnel more effective, sales increased by $1250 for the same number of visitors and with the same pricing as before. Your sales funnel is more effective, and you make more money as a result.

One key formula to remember is this:

If you double conversion rates, then you cut your cost per acquisition in half.

This means if you currently spend $5.00 to acquire each new customer, after you optimize your site and double conversions, your cost per acquisition goes down to $2.50. At that point, you can afford to invest in more advertising or simply benefit from the increased profit.

As you can see, if you want to increase your sales and boost your bottom line, CRO is the way to go.

Case Study: The 2008 Obama Presidential Campaign

The 2008 Obama Presidential Campaign has become a famous example of how conversion rate optimization can help businesses, organizations, and even political candidates accomplish their goals.

During the campaign, Dan Siroker, the founder and CEO of Optimizely, was serving as the Director of Analytics for the Obama Campaign. He proposed that different headline and image/video combinations should be tested to see which one was the most effective at getting people to sign up for the President’s e-mail newsletter. The campaign team knew that a large percentage of people who sign up would end up supporting the President financially, but the trick was getting the most people possible to sign up for the campaign newsletter.

Dan’s team ended up putting together 24 different variations with four button options, three images, and three videos. Interestingly, most people on the team assumed the videos would be the most effective at getting people to sign up.

This is a screenshot of the winning variation from Barack Obama’s 2008 successful presidential campaign.

So what was the result? After testing with 310,382 visitors with each variation seen by around 13,000 people, the version with a picture of Barack Obama’s family and a “Learn More” call-to-action button led to a conversion rate of 11.6% for the winning variation, a 40.6% increase over the original conversion rate which as 8.26%.

Even more impressive than the increase in conversions from the winning variation is the fact that the new version let to approximately 2,880,000 additional sign ups, 288,000 more volunteers, and an additional $60 million in donations over the course of the campaign.

What Exactly Is A/B Testing?

A/B testing is when you set up two different variations of a web page or landing page and send an equal amount of traffic to each. You then measure the number of conversions for each variation, and declare a winner based on which one generates the most conversions. Once a winner is identified, you switch to that version so you can begin benefitting from the improved conversion rate provided by the winning variation.

A/B testing is different from before and after testing. With before and after testing, you measure conversion rates on your site to set a benchmark. You then create a new version to see if it improves or decreases conversion rates. The problem with this kind of testing is that it doesn’t take a scientific approach to measure the results. Conversion rates fluctuate from week to week so it’s impossible to know for sure whether the change in conversions is from the new version or from a higher or lower quality of traffic for that particular week.

One week you might get a large amount of traffic after getting mentioned on Mashable, but the traffic ends up not being a good fit for your product, so even though traffic and possibly total sign ups increase, the conversion rate could go down. This can lead you to think the new version led to the decrease, but actually lower quality traffic should be held responsible.

What you need to remember is that a scientific approach to testing is the only way to know with a high level of certainty whether or not a new version is impacting conversions positively or negatively. Multivariate tests are another more advanced option, but we recommend sticking with A/B tests for now until you get more familiar with conversion rate optimization and how to conduct tests on your site.

Is It Really Necessary to Test, or
Can You Just Learn from What Other People Have Done?

When it comes to conversion rate optimization, a lot of people talk about testing, but not as many end up actually doing it. For the ones who don’t, they usually copy tests that produced a winner on another site. There’s a problem with that approach, however.

First, if you just copy what someone else has done, you don’t know for sure whether or not it will have the same effect on your site. Just because Performable increased conversions 21% by changing their call-to-action button from green to red doesn’t mean you’ll experience the same lift. Red may complement their design better than it does yours, or it might resonate with their customers more than it will with yours. Unless you test yourself, you’ll never know for sure.

Second, you can’t always trust wins reported by other sites. Lots of people talk about the wins they get from conversion rate optimization, but you don’t know whether the test has been carried out properly. How long was the test run? What’s the statistical probability that the test found the winning version? Answers to these questions will give you a better idea about the accuracy of the results, but often you can’t find them out unless you test for yourself.

How Often Should You Test?

When it comes to testing, you should do it every time you make a change that will potentially affect conversions. If you change the copy buried on a customer service page on your website, you don’t need to test to see what effect it has, but if you change the copy or design on a prominent page, particularly one that’s part of your conversion funnel, then you should run a test.

The reason for this is that you never know how a change will impact sales. You might be convinced that a new testimonial will boost conversions, but after testing you find out it doesn’t. Seasoned CRO consultants and professionals learn never to trust their gut. Quite often a version you’re convinced will win doesn’t, and one you aren’t so sure about actually does.

One of the biggest benefits with taking this approach to testing every significant change you make on your site is that you can always resolve arguments with co-workers based on data. If you don’t have test results, then you’ll go back and forth debating which version is better based on the preference of each member on your team. The problem with this is you never know who’s right. But once you test, you know exactly which version you should use. In this way, conversion rate optimization is a great way to limit inter-office debates over which design or copy version is better than another.

How Much Improvement Should You Expect?

You shouldn’t always expect a drastic improvement. Many blog posts brag about conversion boosts of 100% and more. While you may see those kinds of results, and obviously that’s the goal, you won’t always experience huge wins.

But you also shouldn’t be disappointed. By increasing conversion rates an average of 19% over four tests, you’ll end up doubling your conversion rates. When you look at it that way, you realize that even small wins can add up and lead to big improvements in the end.

What Should You Be Testing For?

Ultimately, you should be testing for your final conversion goal for your website. If your goal is to increase free trial sign ups, then you need to measure for that. If it’s to increase sales in an eCommerce store, then that’s the conversion goal you should measure.

If the goal of your website is to increase free trials and eventual paid sign ups, then you should measure that as your conversion goal, not click-throughs from the homepage to the sign-up page.

If you don’t, you’ll end up measuring micro-conversions which doesn’t always lead to an improvement for your final conversion goal which is what really matters. For example, you might measure the number of people who move from your homepage to the sign-up form for your free trial. After running a test, you find out that 10% more people move on to that page with your new version so you decide to make the change.

What you didn’t know is that even though it increased click-throughs to the sign-up page by 10%, it decreased paid sign ups by 5%. And that’s why you need to measure your final conversion goal. Micro-conversion improvements are great, but that’s ultimately not the goal of your test. Your goal is to increase sales.

This kind of scenario is not uncommon. One version may increase click-throughs to the sign-up page by 2%, but the other one increases actual signups by 8%. Or in another scenario, both versions might lead to the same number of people clicking through, but the new version leads to more people eventually ordering. Unless you measure for your final conversion, you’ll never know, so it’s much better to measure final conversions and not over emphasize micro conversions.

Chapter One Summary

  • Conversion rate optimization is the process of A/B testing different versions of web pages & landing pages to find out which one converts more visitors into customers.
  • CRO is valuable because it increases revenue and lowers your cost per acquisition. In fact, doubling conversion rates means cutting your cost per acquisition in half.
  • A/B testing is the best method for finding statistically relevant results that aren’t tainted by changes in traffic quality.
  • You always need to run your own tests to confirm someone else’s results will work for your business. Different industries can have different results, and you can’t always trust that someone else’s test was carried out with the highest level of accuracy. Thus, the only way to know for sure is to run your own tests.
  • You also should test any changes that may affect conversions on your site. Just because you think something is guaranteed to improve conversions doesn’t mean you shouldn’t run a test first to make sure.
  • Even small wins add up, so don’t be disappointed if your test only leads to a small increase in conversions. Big wins are exciting, but everything adds up.
  • Remember to always measure your final conversion goal. Micro-conversion improvements are great, but what really matters is how the change affects the conversion that impacts your bottom line.

Four ways to share