More conversions generally mean more money for your business.
And most companies have natural limitations on their upside—budget for advertising, addressable market, R&D resources, etc.
But some companies limit their own upside further by going at conversion rate optimization (CRO) blindly, without studying visitor behavior. This leads to poorly-informed (and likely unsuccessful) optimization efforts.
In this guide, we share our step-by-step conversion rate optimization (CRO) process, point out what parts of your website can be optimized and help you start to apply a visitor-first approach to your website.
The traditional definition of CRO
Conversion rate optimization is the process of (form completions, add to cart, content download, link click) on your website. At its core, CRO is about optimizing every element on your website to align with visitor needs and expectations. It’s about optimizing the website experience.
With this in mind, we want to offer a modified approach: Visitor-first CRO.
Visitor-first CRO requires teams to invest time in studying visitor behavior first, before optimizing or even coming up with an overall strategy.
By doing this, you’re optimizing for human behavior and likely increasing the percentage of first-time and repeat website visitors who convert along the way.
Let’s take a look at a CRO process that prioritizes your visitors, from start to finish.
The 5-step Conversion Rate Optimization process
The steps in this conversion rate optimization tutorial can be applied to any business, no matter if you run an e-commerce site, operate a through HubSpot or manage some other sort of website traffic.
Let’s walk through how each of these takes another step towards increasing conversions.
We're committed to your privacy. Lucky Orange uses the information you provide to us to contact you about our relevant content, products and services. You may unsubscribe from these communications at any time. For more information, check out our .
GOAL: Identify and select optimization ideas.
First, start by collecting information about visitor behavior
You can do this with a CRO tool like Lucky Orange. This part of the process is the art of optimization and requires the most practice. Start by looking for outliers and things that go against what you’d expect (or like) your visitors to do on the web page.
At this point, you’re looking to spot trends like:
An under or overperforming page or set of pages
Unique behavior by an audience segment (UTM parameter, geographic location, )
Visiting a certain page before leaving your site
Performing certain checkout actions before abandoning
Visiting certain pages prior to adding to cart or checking out
Once you’ve collected data, it’s important to organize the information to make it useful across the organization.
Even if you’re a solopreneur, this step allows you to keep track of ideas for prioritization and later reference. Using a can help you codify your insights.
Before coming up with a hypothesis, you need to prioritize ideas. There are endless ways to do this and your organization will have a unique set of requirements for work to get done.
No matter how you approach CRO prioritization, we recommend applying a rubric that determines potential impact.
This should include things like how much traffic the page gets, how long it’ll take to implement the new idea and other factors you can find in this .
GOAL: Formalize what you’re testing and what you expect to happen.
Now that we’ve chosen an optimization idea to take forward, it’s time to develop a proper hypothesis. This can take many shapes, but here’s what we recommend.
We saw [visitor behavior] on [URL or section of site].
We believe that changing [thing you’re testing] will increase/decrease [what you’re measuring].
We saw X and believe that changing Y will impact Z.
Visitor behaviors can be anything from visitors abandoning forms, avoiding CTA buttons or not reading vital content blocks. And what you’re measuring doesn’t always have to be a full conversion. It can be something as simple as increasing time on page.
More on how to measure success later.
Documenting your hypothesis is important (even if you’re a team of one). Over time, you may run hundreds of A/B tests and at some point you’ll want to look back to see if you’ve tested something before.
Come back to tests from time to time as your products and services evolve. Just because a hypothesis was incorrect the first time around doesn’t mean it won’t work a year or two later.
GOAL: Verify testing elements and audience segmentation is accurate, then launch.
Finally, it’s time to test.
Everything to this point has focused on understanding where visitors are struggling on your site.
This is our visitor-first approach and leads to testing that’s well-informed and customized to a real problem your visitors are having.
Building your test can happen in a multitude of ways including using great tools like . No matter which software powers your A/B tests, be sure to run through this .
Is the test being served to the right audience segment?
Is the test being served to the right percentage of that audience?
Do you have an end date set or will you test until reaching statistical significance?
Does your test variation render properly on mobile browsers?
Will this test impact any other team members or departments? Does this require communication before and after launch?
Have you documented your hypothesis somewhere?
After launching a new split test, it’s time to breathe easy.
No matter what happens, you’re and taking steps towards an amazing customer experience.
GOAL: Monitor performance and communicate your findings.
Once you , evaluation begins. Depending on your traffic levels, you may start to notice an impact within a matter of hours or days.
Our evaluation process is simple—look to see if what you’re measuring is changing.
Regardless of the change you see, of people experiencing your test variation. Note any differences in behavior.
This can be something as small as scrolling further down the page or filling out your form fields in a different order.
If you spot an opportunity to refine your test, adjusting mid-flight is okay. While we recommend against doing this regularly before reaching statistical significance, from time to time you’ll have an aha moment and alter the test while it’s running.
Once the test completes, it’s time to report and communicate. Again this is pretty simple. We’re looking to identify any change in the key measurement along with a few notes of what you saw.
You may see a meaningful difference in how people navigate your web page without an impact on the key measurement.
Oftentimes, studying A/B test results will help you come up with ideas for future testing.
After writing down your findings, it’s time to share with any relevant team members.
This includes anyone who may benefit from knowing what you learned and can change from test to test.
What to optimize on your website
The way your website performs is a combination of every page, word, button and block on the site.
This also means all of those items can be optimized over time. This list isn’t exhaustive but is a great place to start when considering how you’re moving people down the conversion funnel on your site.
Note: There are other optimization categories not listed here. Things like page speed and technical performance or security are also areas where your site needs to perform at a high level. This list focuses on areas where you’ll primarily .
Every word on your site matters, and some matter more than others.
Start by ensuring your are prioritized in important spots like your home page.
Doing this ties everything together and helps people understand how your offer is different/better than competitors.
Be sure to balance primary and secondary messaging throughout the site with your secondary always working to prove or support the main point of your brand’s value to customers.
On our home page, we’ve decided to highlight our primary and secondary messages just before our initial CTA offer.
By doing this, we greet new visitors with a clear explanation of the value they’ll get from us and start mentioning specific tools so they know what to expect should they read further.
Your unique selling points should apply to the brand as a whole and extend throughout your site. If you’re in e-commerce or another industry using online products, dialing in your is vital.
Study to see which content blocks people read and which they pass right over.
This will help you determine what’s important and what’s not—and show you where your optimization focus needs to be. If you’re in e-commerce, use this opportunity to study which and throughout your checkout funnel get the most hover time in recordings.
What about search engine optimization?
If you rely on organic traffic as a conversion source and have the resources to commit to improving your Search Engine Results Page (SERP) positions, then you’ll of course want to spend time working with .
While not something you’ll necessarily apply a CRO tool to, that it’s necessary to include here. We use to monitor keyword positions and plan content topics.
Websites are over time. There are many reasons for this, one of which is optimizing human behavior and a general understanding of what works and what doesn’t.
Does this mean there’s a perfect web page design we should all build?
If everyone had the same traffic sources, products and goals, then the answer would likely be yes.
Since all businesses are not the same, CRO starts by adopting best practices for your specific situation, many of which relate to the layout.
How you organize elements around your site can have a huge impact on how easy your website is to use—and .
Here are some layout elements to focus on when studying and Session Recordings.
Navigation: The goal here is to make all parts of your site easy to find. Surfacing key pages and removing unnecessary clicks to find something will make your site much more usable. And having a great search feature helps users as well.
Call-to-action: Focus first on . Make sure it’s the primary focus of that spot on the page and that it’s above the effective fold (how far down the page most people scroll). After you dial in CTA location, testing button color, shape and size are also worthwhile efforts.
Images: Beyond making sure you only use hi-res, relevant images, be sure they’re placed in spots that don’t get in the way of a conversion. Hero images can be great, but make sure your messaging is visible and that they help guide people down the page or further into your site’s conversion funnel.
Forms: This part of your page is important enough to devote an entire section to below. For now, just make sure your forms are above the effective fold on your page, so that the average visitor sees them and has a chance to convert. Find this metric by looking at a .
Mobile: Does your site render how you’d expect it to on mobile and tablet size browsers. Are people able to complete your desired action or does an element get in their way? Filter Session Recordings by device type to only see traffic for mobile phones, tablets, etc.
Lucky Orange highlights five key form metrics that will help you fix form issues.
This includes things like which form field causes the most abandonment, the order in which people complete your forms and any fields people complete more than once.
The best forms are simple and intuitive. This means removing fields that aren’t completely necessary and giving people easy-to-understand descriptions of what they need to provide in each field.
Your should start with metrics and include a variety of tests like re-ordering or replacing fields and potentially breaking forms into more than one step.
Always keep in mind any impact form changes may have on later steps in your conversion funnel—especially if sales people or further qualification will be necessary after the initial .
How to find the best conversion rate optimization tools
Your business is unique. And your current focus and growth trajectory is even more specific.
This is why finding a great CRO tool that fits your needs in the near and long-term is vital. But reading overly-technical product descriptions and sitting in valueless demos is the worst.
Here’s what we recommend you consider when it comes time to start using CRO tools:
Can this tool track everything and everyone on my site?
Does this tool offer a feature I’m already using elsewhere (chat, , form analytics) and can I consolidate?
Does the tool sample traffic, leaving me without a complete picture?
Are there customer communication tools to inform my tests and support broader marketing efforts (, announcements)?
What collaboration features does the tool have?
Can the tool scale as my business grows or is it designed for a certain size company?
Common questions about conversion rate optimisation
How do I calculate conversion rate?
Calculate your conversion rate by taking the number of conversions and dividing that by the number of sessions or the number of people who sign up for a free trial of your service and subscription and then upgrade to a paid option.
Depending on what you’re studying, the elements in this equation can be swapped for unique visitors, leads generated, etc.
How do I improve my conversion rate?
In general, improving your conversion rate is about optimizing each little piece of your website over time. Sometimes you’ll spot a huge opportunity and others, you’ll make a tiny change to a piece of copy and see major results.
What is a conversion funnel?
A website conversion funnel is a route people take to learn about your product or service, consider if it’s right for them and then make a purchase. In the case of Lucky Orange’s tool, you choose a page from your site to track and then can see behavior between those pages in a certain order.
Conversion Funnels for CRO highlight points in the journey people are dropping out of the funnel. This is usually a good place to begin with optimization efforts.
What’s a good conversion rate for my business?
There are many places online to find industry-specific conversion rate benchmarks. And while they can be valuable in some cases, most of the time we recommend setting your own benchmarks and working to improve them.
There are so many variables at play that impact your conversion rate, that focusing on your own site experience as the starting point is a better path forward.
What web pages should I test first?
Start with your most visible pages, those that get the most traffic or are the centerpiece of your current marketing campaign.
If you don’t have any current focus areas, start with your home page, product pages, top blog posts or anywhere else getting enough activity to build a solid amount of data inside your CRO tool.
Part of building a solid split testing hypothesis is having ample examples and .
Who should be my CRO expert?
Whether your role centers around digital marketing, sales, product development or something else entirely--you can become a .
Regardless of how you decide to handle optimization, having clear roles and expectations is highly valuable. And giving time for people to and become an expert is often necessary for newer businesses.
Does SEO impact my conversion rate?
A lot of current thoughts on content marketing revolve around aligning topics and channels with a user's stage in the customer journey.
Similar rules apply when considering the relationship between SEO and CRO. The more high-quality answers you provide on your questions, the more organic traffic you’ll get. And if that traffic is satisfied by your answers, they’ll be more likely to convert.
So, yes there is a connection but it should primarily be leveraged in your content strategy as opposed to studied with CRO tools.
Where to start with CRO
Keep things simple. You don’t need to hire an agency for conversion optimizations.
Remember, you’re studying visitor behavior to learn more about the people who want to buy something from you—and that you know them better than anyone else in the world.
If you get stuck in a rut, study a high-performing competitor’s pages for inspiration.
With the right processes and tools and commitment to the long game of optimization, you’ll no doubt increase your conversion rate to levels you may have previously thought impossible.