How to rock your conversion rate and increase sales 200%


Imagine you’re trying to fill a pool. You’ve got the best pump money can buy, plenty of clean water and a long hose. There is just one problem. The hose is leaky and full of holes. Most of the water from the pump ends up on the lawn, in the ground or on your clothes; very little actually makes it to the pool. Can you picture how much time and energy you would spend trying to fill this pool up?

Do you agree that the method is ineffective and wasteful without a better hose?

You probably do.

The crazy thing is, many businesses have to work sales processes that are just as leaky and ineffective…and why is that you might ask yourself. Because they don’t realise there is a simple framework for plugging those holes and maximising the inflow of paying clients.

That framework is called CRO: Conversion Rate Optimization. CRO is the process of maximising the number of people you convert, i.e. get to take a desired action. It’s what fixes a leaky sales process, or takes it to the next level. In this post, we will introduce you to the history of modern CRO; a history that predates the internet by many decades. We will also explain the core ideas that help maximise conversions and eliminate the “leaks” in your sales process.


Did your brush your teeth in the last 24 hours? If the answer is “yes”, you have got one man to thank: Claude Hopkins, one of the earliest advocates of CRO. In the beginning of the 20th century, most North Americans and many Europeans did not brush their teeth. This changed when Pepsodent, then a young company, hired Claude Hopkins to do their marketing.

His ads helped popularise the daily habit of brushing teeth. They were published in 52 countries and 17 languages, making Hopkins a millionaire back when $500 bought you a high-end car. The ads also made the entire United States brush their teeth; a habit that has persisted to this day. The key to all this success was Claude Hopkins’ obsession with optimising conversions.

For example, he would add unique “codes” to the coupons, free trials and mail order forms used to order Pepsodent toothbrushes. Then he would see which code got mailed back most often and run the ad with that code again.

In the end, the campaign – and Hopkins himself – became so successful that David Ogilvy wrote:

Every time I see a bad advertisement, I say to myself, “The man who wrote this copy has never read Claude Hopkins.”

That is the history of CRO. The ironic thing is that one of the key elements of improving conversions has not changed since then, because the surest way to fix any leaks in your sales process is to do A/B split testing. This leads us to one of the core concepts of CRO.


Testing is the only surefire way to know whether one marketing effort is better or worse than the other. Planned, monitored experiments are the best way to get specific and measurable feedback on whatever you are doing, and that is where A/B split testing comes in.

Your A sample is usually the control: a previous marketing effort for which you already have analytics (like cost per click, opt-in rate or sales conversion rate). The B sample is the test: a new ad, sales page or piece of content that you are testing against the control. Once you see which sample is doing best, it becomes the new control against which you test on an ongoing basis. That, in a nutshell, is A/B split testing: the engine that drives CRO.

Sometimes, you manage to beat the old control overnight. Other times, a control beats all your B samples for years or decades. That is what happened when the Wall Street Journal found a sales letter that made it over 1 billion dollars, beating everything else they tried for 10+ years.

However, to A/B split test successfully, you need to…


Testing everything is a great way to lose time and money, so it’s imperative you have a clear vision for what you are going to measure. The first rule is that you never test two completely different ads, videos or articles against one another. Sure, you may know which one works better, but you will not know why.

Instead, it’s best to make small changes from the control and test them. One example would be to compare two identical sales pages with different headlines. A test like that will give you immediate, measurable information on which headline performs better, which is what you need for effective CRO.

Of course, you will have times when you need to run completely different marketing efforts against one another. However, generally speaking, it’s best to keep A and B samples mostly the same with 1-2 planned difference(s).

One example of a difference to test is the colour of your on-site buttons. Most people would never think that red buttons convert better than green ones, but numerous studies have shown that red buttons get more clicks overall. Repeating this experiment to see if it holds true for your website and business is one savvy way to use CRO. This brings us to core concept #3.


Returning to the swimming pool analogy, the goal of CRO isn’t to fix the leaks in your sales hose once. It’s to show you can have an airtight sales process in the long run. This means building up a database of knowledge through experimentation…and through making a lot of mistakes.

Remember: CRO is a process. Not a quick fix. Many of your tests will fail to beat the control. Sometimes, you will think you found a new control, but after realise that it doesn’t do well over time. This happens quite often, and you will find that whatever you tried didn’t change conversions at all.

Don’t panic! This is all part of the learning process, and converting as many people as possible means making those mistakes again and again. So before you start your first CRO efforts, commit to making a few mistakes. It’s the best way to learn quickly – and make sure that you’re really experimenting and not just treading water.





Related Posts

Comments (3)

Good one! Thank you for sharing such an wonderful content.

[…] Remember, growth hacking is about long-term learning, not immediate results. As such, constant split-testing is critical and should be an inherent element in any experiment. […]

[…] on the blog clearly illustrate this, and today we wish to zoom in on one of these yet again: Conversion Rate Optimization (CRO). However, this time we will do it differently. Instead of writing another post on the topic, we […]

Leave a comment