Understanding Conversion Rate Optimization

What is Conversion Rate Optimization?

In Internet marketing, conversion rate optimization refers to the process and techniques of continuously improving a website's effectiveness at turning visitors into customers. This is done by making strategic changes to the website's design, copy or flow and evaluating the results.

Marketers try to optimize for conversion to increase the return-on-investment (ROI) of traffic acquisition efforts such as banner advertising, affiliate marketing, search engine marketing and search engine optimization. By increasing the percentage of website visitors that convert to customers or leads (whatever the desired outcome is), a company is able to derive more revenue from the same amount of website traffic. The process thus lowers the cost-per-acquisition (CPA) of customers and allows a business to, in turn, invest more in traffic acquisition or other marketing efforts, leading to greater profits.

Typically a website operates like a funnel, wherein users take a series of steps or behaviors before buying. For example, individuals might use a website to learn more about a product, browse merchandise, view an offer, register for a free trial and ultimately make a purchase. The conversion funnel can involve several visits to the website as well as other forms of connection such as email. At each step some visitors stall or exit the funnel, such that only a percentage of the total website traffic actually converts to customers. This percentage (# final conversions/total # of visitors) is called the conversion rate. Conversion rates can vary over time (day, season, etc.), by source of traffic (organic/paid search, social media, word of mouth, etc.), and other factors.

For websites with significant traffic volume, even very small, incremental increases in conversion rates can lead to substantial revenue gains. Conversion rate is also not always the best or only metric to consider; for example, a price increase might lead to fewer conversions but ultimately greater revenue.

Image of the Conversion Rate Optimization funnel

The Conversion Rate Optimization Workflow

Increasing conversion rates usually involves making the path through the conversion funnel as clear, simple and effective as possible by removing distractions and barriers, as well as aligning the website experience with what visitors are looking for. Conversion rate optimization typically follows a cyclical workflow of measuring visitor behavior, gathering insight, testing hypotheses and evaluating results.

Image of Conversion Rate Optimization flow chart

Analyzing & Measuring Visitor Behavior

Optimizers start by measuring visitor behavior to understand the current conversion process and rate, including which areas are the least effective at getting users to the next step and which webpages present the highest potential for improvement. Analysis using web analytics software helps optimizers determine baseline metrics, set goals, and prioritize areas for improvement. Some key metrics are unique visitors, conversion rate, exit rate, bounce rate, order value, retention rate and lifetime value.

Web analytics software can also provide visitor data such as demographics, previous visits and behaviors, and device/browser information. This data can be used to optimize content for specific segments of web traffic through personalization and targeting tools, as well as to evaluate the quality of different sources of traffic.

Web analytics is often called digital analytics, as it includes the analysis of customer or user behavior in a web, mobile or native app environment. Digital analytics tools are useful beyond the context of conversion rate optimization as well. For example, companies might use analytics to understand different sources of web traffic, make paid search marketing decisions, or evaluate the effectiveness of search engine optimization efforts.

Gathering Insights

Another key step in the CRO workflow is to gather qualitative insights to understand why visitors are behaving in a certain way. Getting feedback from visitors or observing how they interact with a website can help answer questions like: What are visitors looking for? What are their driving motivations? What are their pain points? What areas of the website are confusing or distracting? What's keeping them from moving forward in the conversion funnel?

There are many different categories of tools that can help a business gather insights from its website visitors or customers, including mouse-, click- and eye-tracking, session replay, user or usability testing, live chat, survey and voice of customer tools. In the context of a CRO workflow, the goal of the insight-gathering phase is to understand why visitors are behaving in a certain way, and to generate testable ideas for strategic site changes that might improve the user experience and increase conversion rates.

Photo of Dr. Karl Blanks - Chairman of Conversion Rate Experts

"Analytics software knows when someone arrived and when they left. But it can't tell you why they came, what their preconceptions were, what their objectives were, and--most importantly--why they did or didn't buy. So qualitative tools are also important."


Photo of Jason Burby - President, Americas at POSSIBLE

"We look at behavioral and attitudinal information--what are people doing and how are they feeling. Next what matters is how you turn around and use that information."

Insights gleaned from website metrics, visitor observation and feedback help generate ideas for strategic changes that could potentially increase conversion rates. The next step is to test these ideas, either by simply updating the website and comparing metrics before and after the change, or by creating two or more versions of the webpage and dividing web traffic randomly among the different versions. The latter process, called split testing or A/B testing, produces more accurate results because it is a randomized experiment and therefore less susceptible to being influenced by normal fluctuations in conversion rate on a daily or weekly basis. A related type of experiment is multivariate testing, in which multiple elements on one page are tested at the same time, thus creating and dividing traffic among many different versions of the same page.

Whichever version of the webpage "wins," or consistently leads to a higher conversion rate, becomes a permanent part of the website. Another idea can then be generated and tested. The process helps digital marketers make iterative improvements to a website by ensuring that a change is actually an improvement before launching it to all users.

Marketers can also run more targeted website tests by analyzing the behavior of specific visitor segments in response to variations being tested. Segments can be as simple as New vs. Returning Visitors, or more complex combinations of geolocation, device or browser used, referring source (search, social, email, etc.), and more specific visitor data, if available. This type of targeting enables websites to show the most effective content to different visitor segments.

Website testing and targeting software can be used to set up A/B tests, split traffic randomly, analyze results and target specific visitor segments. More sophisticated conversion optimization tools allow optimizers to run tests that span multiple pages of a website, and offer multivariate testing, in which the marketer creates several different values for multiple elements on one page to generate countless combinations or versions of that page to test at once. Some tools can perform native app or mobile website testing as well.

The most advanced website testing and targeting tools help websites serve up a more personalized experience for each visitor, in a near one-to-one fashion. Personalization is an attempt to show the most relevant content based on what a company already knows about an individual.