What You Need to Know About Conversion Rate Optimization

A little over a decade ago, Theresa Farr, a marketing consultant, was asked by a luxury vacation home rental company to help grow its website traffic. After a little SEO work and paid advertising, the company received a huge boost in website visitors. Its sales, however, didn’t rise with it.

Perplexed, Farr popped the hood. She discovered that the checkout page on the company’s website offered a poor user experience. So while people were showing up in droves, when it came time to pull out their wallets, they abandoned the site.

“Why are we putting all this effort and resources to drive traffic to a site that’s not converting?” Farr remembered wondering. “It was a huge aha! moment.”

She had stumbled upon the importance of conversion rate optimization (CRO).

Those were the early days, “when everybody was focused on SEO,” Farr, who runs a CRO consultancy, told Built In. “Not as many people were thinking about conversion.”

Now, conversion rate optimization is considered to be an essential marketing activity.

Rand Fishkin, co-founder and CEO of SparkToro, told Conversion Rate Experts that CRO is the most important thing a marketer can do — “because it makes every visitor exponentially more valuable.”

With enough paid ad spend, anyone can drive more traffic to their website. But once people get there, persuading a greater percentage of them to actually buy something is a challenge that can’t be solved by money alone.

Which is where CRO comes in.

ReadAudience Development Turns Visitors Into Regulars

How to Calculate Conversion Rate

First things first: To calculate conversion rate, take the number of conversions and divide it by the number of website visitors. Multiply the answer by 100 to get a percentage.

For example, suppose each month an e-commerce site gets 1,000 visitors, and 30 of them end up buying something. The conversion rate is 3 percent.

30 / 1000 = .03

.03 x 100 = 3

What’s a Conversion?

Technically, a conversion could be any action you want a website visitor to take: Fill out a form, sign up for a newsletter, follow a social media account.

The conversion that matters most, though, is the one where people pay you.

“Everything is bullshit until somebody writes a check or gives you their credit card,” Richard Bitz, a CRO consultant, told Built In.

Examples of Conversion Rate Optimization

While there’s no one-one-size-fits-all solution for optimizing a website’s conversion rate, common approaches include doing one or more of the following:

• Changing the headline copy
• Changing the CTA copy
• Changing the design
• Changing the site structure/navigation
• Showcasing a sign-up form or “buy now” button more prominently
• Adding a video
• Adding user reviews or testimonials
• Presenting the pricing options differently
• Explaining the product benefits differently

To give an example: According to Bitz, one time a client who needed help with conversion rate optimization told him, “We need to change the homepage.” Bitz said, “We’ll get to that later, let’s first just change the copy on the checkout flow.” By his account, after the copy change was made, sales increased 80 percent.

While this sounds both exciting and easy, CRO experts suggest people resist the temptation to immediately jump in and make gut-driven changes. The solution is almost never as simple as changing a CTA button’s color from red to blue. The first thing to do is zoom out.

Read4 Perspectives on Net Promoter Score

How to Optimize Conversion Rate

Investigate

Start by digging into the quantitative data to pinpoint where any disconnect occurs. Upon analyzing Google Analytics, is there a noticeable point along the funnel where users drop off, such as the checkout page? Signup page? Product page? Homepage?

After you’ve gathered quantitative data, it’s time to collect qualitative data. This means paying attention to the audience.

There are a number of ways to do this.

One is through conducting customer surveys, using questionnaires that either pop up in real time on site or are emailed to customers offsite.

Surveys should ask visitors who convert (and those who don’t) who they are, what their pain points are, what they’re looking for and what their experience with the site is like. If they don’t convert, ask them what about the product or price turns them off. Their answers may reveal one — or several — reasons why they aren’t converting.

Another option is user testing. Several services offer the ability to recruit real people to observe as they complete tasks on the site. This allows observers to see where in the checkout or sign-up flow users are getting confused or frustrated. This is especially useful if users are too polite to tell the honest truth during written surveys — or if they can’t quite pinpoint what, exactly, threw them off.

You can use mouse-tracking and heat-mapping tools too. They help uncover where on the page users are scrolling. The culprit of the bottleneck could be as simple as a confusingly placed button.

When the results come in, a clearer picture should emerge of what’s potentially keeping website visitors from converting into customers. At that point, it’s time to run some tests.

It’s important not to skip steps or take shortcuts during the investigation and research, since doing so might require you to re-run your tests. Tests take time, and time costs money.

Test

The qualitative and quantitative data produce one or more hypotheses. The tests ... test them.

A common way to do this is to run an A/B test (sometimes called a split test).

Using an A/B testing tool, any website page can be set up in such a way where some users experience the original, and others experience the new version.

Keep tests running until they reach statistical significance. That means enough conversions need to occur on both versions for the test to actually prove or disprove the hypothesis. That number will vary and can be calculated by using a statistical significance calculator, which is free and easy to find online.

The test should also run at least as long as the sales cycle. If the typical sales cycle for the product being sold is three months, and the test is only held for three weeks, the results will not reflect the reality.

At the end of the test, it should be clear whether or not the changes helped solve the underlying problem, by improving the conversion rate. If the test doesn’t move the needle, it’s time for a new hypothesis, and a new test.

What’s a Good Conversion Rate?

Conversion rate benchmarks vary from industry to industry, action to action.

One company that analyzed more than 4,000 landing pages across 16 industries discovered that the median conversion rate was just above 3 percent. Elsewhere it was reported that the average conversion rate for e-commerce sites hovers around 2 or 3 percent.

That said, there’s no magic number to hit. The biggest companies in the world continually run tests to optimize their conversions. The goal is always more. CRO is about moving in a direction, not reaching a specific number.

Farr doesn’t think industry benchmarks are particularly helpful to measure against.

“My clients want to make more money,” she said, adding that many of them come to her having tried CRO tactics before. “But they know they can do better.”

Read NextWhat’s the Difference Between UX Writing and Copywriting?