CRO: Top of Funnel vs. Bottom of Funnel Changes

When it comes to conversion rate optimization, you already know that there are an infinite number of tests you can run. The only constraint should be the amount of time, money, and other resources you can devote to improving the overall experience.

Because most companies don’t have an unlimited budget to throw at CRO, you will be forced to prioritize. And those marketing teams who are best able to prioritize their efforts will win on two fronts:

  1. Early successes will have greater impact on the business

  2. Because those early successes will add value, your efforts will get noticed by those that can pour more resources toward future improvements

So how do you start to prioritize the work? You have a couple of different options. You will want to find a way to estimate the potential impact of your changes. And to do that, you should start to categorize those changes into Top of Funnel or Bottom of Funnel changes.

Top of Funnel Changes

Top of funnel changes are going to be made very early in the conversion path. These are things like the ads that someone would click on, or the landing pages that they are taken to.

Changes at the top of the funnel have a simple goal – get more people into the funnel. The theory goes that if you can get more people into the funnel, assuming a constant conversion rate, you will get more sales.

Marketers who choose to focus on the top of the funnel see that as the greatest opportunity because it represents the largest possible audience. The nature of the funnel is that the widest audience exists at the top, with more people falling off with each step in the journey toward that ultimate sale.

The more people you get into the top of the funnel, the more people will be left over at the end.

Bottom of Funnel Changes

Bottom of the funnel changes are going to be made nearest to the end of the conversion path. Often, this would involve changes to the product pages or the actual shopping cart experience of an ecommerce website.

Changes at the bottom of the funnel are aimed at getting more people who start the checkout process to finish. The theory goes that these people are the most qualified, and if you lose them at this point in the journey, it’s like throwing money away.

Marketers who choose to focus on the bottom of the funnel know that it can difficult to reach a wider audience. And so it is better to focus on those prospective customers that have already expressed a high level of interest in your products or services.

The more of them you can keep from leaving the process altogether, the more sales you will generate.

Which One is Better?

Neither. Sorry to disappoint. Both are important, and I know people who can make a strong argument for one or the other.

The argument for starting with the top of the funnel is that there are often easy ways to add more people to the funnel. And if we assume that the overall conversion rate once they are in the funnel remains constant, every new person you add will grow your revenue.

The argument for starting with the bottom of the funnel is that there is often low-hanging fruit opportunities to improve the checkout process, essentially plugging the holes in your funnel so that you lose fewer people in the crucial final steps.

If I am forced to pick one or the other, I pick the bottom of the funnel. The reason is a simple one. Once you plug the holes, any improvement you can then make at the top of the funnel will have greater overall impact.

But the choice, for your company, is yours.

There Are an Infinite Number of A/B Tests You Can Run

almost perfect.jpg

So much of your efforts in conversion rate optimization should be centered around running controlled experiments. And the most popular of those is the A/B test, wherein you show one half of your website visitors the current version of a page or element, and the other half a new version of that same page or element, in order to see which version performs best.

And in the lifecycle of website testing, there is no end. Each and every test that you run will open up the possibility to run several more. Each and every result that you produce will lead you to new learnings about your website, your audience, and the conversion process. And with those new learnings come new ideas and new solutions.

There are an infinite number of A/B tests that you can run. And so your job, in this realm, is never over.

You must recognize that small changes can have large impacts. You don’t need to overhaul an entire page to dramatically raise conversion rates. Sometimes something as small as the color of a button, or the phrasing of a headline can mean the difference between a conversion and a bounce.

Did adding a call to action to the page lift the conversion rate? Great. But is that the best call to action? The only way to know is to continue to test.

Did showing the price on your product page lift the conversion rate? Great. But is there a better way to highlight it that would drive even more sales? The only way to know is to continue to test.

Punctuation, font size, image choice, color, words, actions, navigation options, social proof, and more. They can all be tested. They can all be added or removed. And they can all be given different design treatments.

A website or webpage is never perfect. The only way to know whether or not you can improve its’ performance is to keep testing.

Top Conversion Rate Optimization Tips


Conversion rate optimization (CRO) is a system for increasing the percentage of visitors to a website that conversion into customers, or more generally, take any desired action. CRO has gained prevalence in recent years with the growth of popularity of “growth hacking”, as companies look to take advantage of new technologies that make online testing simple and intuitive.

If you and your company are using conversion rate optimization to improve your results, or just thinking about getting started in this area, here are my top 5 conversion rate optimization tips:

  1. Start with those tests that have the potential to have the most impact. Prioritize in terms of potential value. Much testing is about finding incremental lift, but there are likely big improvements to be found early on.
  2. Make sure you start by defining the metric, and make sure everyone agrees. What is a conversion? That’s what you are trying to get more of. People must agree on the goal in order to agree on the results.
  3. Be open to surprises. One of the biggest mistakes people make when they’re first getting started is assuming they know the results of a test before the test. Keep an open mind and let the data do the talking for you.
  4. Research and study what other companies/professionals have experienced in their own tests. You can find a wealth of historical data and best practices simply by spending a few hours on Google. These will give you testing ideas and open you up to new ways of creating a better user experience.
  5. Understand statistical significance. In simplest terms, statistical significance is the point at which we can confidently conclude that the results of a test we are running are real, and not just a coincidence.

Simple Website Fixes – Part 11

Welcome to the latest edition of our newest weekly blog series, Simple Website Fixes. Each week we will identify and explain one easy change that you can make to your company’s website in order to improve performance. Last week’s fix was – Add Speed.

This week’s fix = Use Optimizely

It is not possible for anyone, me included, to know exactly what changes you should make to your website. We might have some idea, based on past experience and industry knowledge. But we can never be sure.

That’s why the answer to so many questions is, it depends. Every company is different. Every customer is different.

But one surefire change you can make to your website, one that is guaranteed to improve its performance over time if used correctly, is to start testing. Running simple A/B tests and measuring the impact those changes have on performance will always be a winner. And Optimizely is a tool that makes that strategy easier than ever.

Optimizely is a site testing platform designed for marketers. It is an easy ‘plug and play’ system that requires a developer to simply add a small snippet of code to a company’s website to start testing. From there, you don’t need any technical knowledge to operate. Make changes, define goals, and measure results all in one place, for a low monthly cost based on the number of people interacting with your site.

Any self-proclaimed expert can tell you what you should do to improve your website. But only you can actively test changes on your site with real users and find out what works best.

Implement a testing strategy and you are sure to improve your site’s performance.

Have an idea for a simple website fix? Submit it here and maybe we will include it in an upcoming post.

Price Testing Explained

Price testing is a very simple concept that can be carried out in a thousand different ways, from least complex to most.

The basic idea is this: rather than using the tools at our disposal to choose the right price, let’s allow the customers to tell us what the right price is.

We all know price is a major lever that marketers can use to improve performance. What most people don’t know is the extent to which price matters, and all of the different ways you can impact the price.

How do we set a price to begin with?

Even in price testing, we need a starting point. There are two ways to set a price that make any sense at all. The first is the most common. Start with costs and add a margin on top of it. If it costs us $10 to make something, and we want a 20% profit margin, we charge $12.

The second one is less common, but many would argue more appropriate. The reason it’s less common is because it’s more difficult. Here your starting point is value. What value does this product or service offer relative to the alternatives? For example, if we’re launching a new state of the art pen that lasts twice as long as the longest currently on the market, perhaps we’d set the price at double whatever they charge for theirs. Since ours lasts twice as long, all other things equal, we can assume the value to consumers is double.

There are many different ways to test.

Companies can and do test prices all the time. One of the best examples of this is Amazon. Amazon is constantly using algorithms to try to find the “right” price for all the items it sells.

Five different people might log on at exactly the same time and view the same products but see slightly different prices. The volume of merchandise that Amazon sells and the amount of traffic their site gets allows them to test and find meaningful results in a short amount of time.

We aren’t all so lucky. Here are a few of the common ways of testing prices that you can use:

  1. Use discounts and offers on your website to measure the impact of lower prices
  2. Show different prices to different users and measure the relative performance
  3. Revise the price of your products over time, adjusting them weekly or monthly to see if that impacts sales
  4. Create different versions of your products and offer them at different prices to see which performs better
  5. Use an email or call list to sell at different prices and measure conversion rates

The key to successful price testing

We use price testing to determine the “best” price, often the price that creates the greatest total revenue (sum of revenue from all purchases). A lower price might lead to more sales vs. a higher price. But a higher price means more revenue per sale. So it’s about finding the right mix.

Like other forms of testing, this process can continually repeat itself. We won’t all be as intelligent in our testing and pricing as Amazon, but there is no reason why we ever have to be done testing. Once we find a winning price, we make that the new baseline and then continue to try and “beat” it.