Keep updated with our newsletter

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form. Please, try again
Project Details

Learn how to use the latest CRO tools and techniques to evaluate and optimise your copy.

blog author

Jon Evans

Jon Evans is a tech copywriter and content specialist. Based in Barcelona, Spain, he helps startups and agencies to drive traffic and sales with research-driven marketing materials.


Federico Jorge: How to write comparison pages for SaaS products

Read blog

8 ways to evaluate your copy like a conversion optimisation pro

Learn how to use the latest CRO tools and techniques to evaluate and optimise your copy.

One of the greatest myths in marketing is that great copy comes from great writers.

That if the writer has enough talent and ‘knows how to sell’, they can just take a brief and come up with world-beating copy.

But here’s the conversion copywriter’s best kept secret…

Copy that converts is usually the result of intense researching, analysing, refining and optimising.

…sounds complicated, right? Well, maybe not.

Getting the insights you need to optimise your copy is now easier than ever.

Conversion experts know that to maximise your copy’s results, you need to understand how your prospects think, feel, and behave – and what their objections are.

Fortunately, clever people have come up with easy ways to get these insights. With the tools and techniques we’re about to discuss, you can:

  • Find out if anything is wrong with your copy before your prospects see it
  • Test copy variations with small audiences before pushing it live
  • Figure out why your copy isn’t converting as expected
  • Understand what messaging your audience responds to best

So let’s dive in– here are eight tools and techniques to evaluate and optimise your copy.

1. Split testing 

A/B Testing

Split testing is a method for determining well how a copy change works on real-life prospects. It involves taking two or more variations of copy and showing them to segments of your audience to see which generates the most conversions (i.e clicks, sales, email opens, etc).

This is considered by many SaaS copywriters to be the holy grail of copy testing, as it’s the only way to find out if your copy truly works in the real world. 

When to use split testing

Split testing is ideal when you want to test copy that has a specific conversion goal. It can be used on almost any digital-format copy, and even some physical marketing materials, as long as there’s a success metric you can measure.

 For example, you might:

  • Find the best email headlines. Send 100 people from your mailing list an email with subject line A and send another 100 people an email with subject line B. Whichever email gets opened more wins, and you send that version to the rest of your list.
  • Test improvements to a landing page. Direct 50% of your traffic to a specific web page, then 50% to a variation with different copy. The page that gets more conversions wins.
  • Find the most effective version of direct mail ads. Send out different versions of physical mailouts out with separate coupon codes (or URLS) so you can measure which version performs better. Claude Hopkins was doing this back in the 1920s!

You can even use split testing to figure out what messaging an audience responds to best. Entrepreneur Tim Ferris chose the title of his book The Four Hour Work Week by running 6 ads with different titles to see which one attracted more clicks.  

Drawbacks of this testing method

Split testing won’t work if your site is low-traffic or low-conversion. If your page doesn’t have significant traffic or significant conversions already, it will be difficult to get meaningful results.

Split testing can’t always tell you why one copy variant outperformed another (this is where user testing and can help). It’s also complicated, and you’ll need a decent understanding of statistics to analyse results properly.

Where to find it?

2. 5-second test

Five second test pre-question

The 5-second test is one of the simplest yet most effective tests there is. It shows users an image for 5 seconds only, then asks the user pre-determined questions about the image they saw.

Using this test, you can find out how people experience your copy in the first moment that they see it. 

When to use this test?

Conversion-focused marketers know that people decide whether to continue reading just a few seconds after landing on a site. The 5-second test is ideal for testing out your headline and whatever comes directly below it. 

Use it to show testers your headline, and then ask questions like ‘what is the page offering?’ or ‘what does this company do?’. This allows you to get an objective answer as to whether or not people quickly understand your headline.

5 second test answers

What is this not good for?

Anything other than headings or sub-headings and concise body copy.

Where to find it?

3. On-site surveys

On-site Survey

Using apps like Qualaroo and Hotjar, you can create surveys that pop up on your site at strategic moments, e.g. when a visitor is about to close a page.

On-site surveys aren’t really a test, but they still help you identify problems with your copy by gathering qualitative data from website visitors.

When would you use this?

When it comes to copy, surveys can help you:

  • Find out if a page has all the info a visitor needs. Set a survey to pop up when a visitor is about to close the page. And ask them ‘did you find what you needed on this page?’. If they clicked ‘no’, a second question asks ‘what information were you looking for?’
  • Find out what hurdles your copy needs to overcome. Set a survey to pop up when a visitor visits a payment page, but then navigates to another page instead of buying. Ask the person ‘what stopped you from buying today?’ The resulting feedback may reveal objections you can overcome in your copy.

What are the drawbacks of this?

Any survey is potentially biased, because you are effectively only getting feedback from the kind of people who respond to surveys. 

In addition, surveys on websites get pretty low response rates – you might find that only 2% of people actually respond to them. If your traffic is low, it could take you a long time to gather a useful amount of data.

Examples of use

Using on-site surveys, Teespring found out that customers were concerned about giving their credit card information to the site. The company added reassuring copy under their ‘buy now’ buttons and saw a 12.7% conversion increase.

Where to find this:

4. Session replays

Session Replays

Session replays let you watch a recording of the user’s screen as they use your website. You’ll be able to see how they scroll, where they move their mouse, where they click, and which parts of the page they pause to read.

Like on-site surveys, session replays aren’t a test, but are effective for gathering data around how visitors experience website copy.

When to use session replays to test copy

Session replays are primarily used to identify where users encounter navigation issues and bugs on a website. However, they can be used to identify problems with copy too—so the method is ideal when you need to figure out why a page isn’t converting well.

For example, you might find that your visitors are repeatedly scrolling back up and re-reading a chunk of copy. This could mean that visitors don’t understand the text and you need to make it clearer. 

What are the drawbacks of this method?

Session replays won’t give you any great insights on pages that don’t have a specific conversion goal, or pages that don’t have a lot of content to scroll through. In addition, watching session replays can be extremely time-consuming.

Examples of use

Using session replays, Vocabla learned that visitors were quickly skimming through their sign-up page. After reworking their page copy and design to make the benefits of their offer clearer, they saw a 218% conversion lift

Crazy Egg’s on-site surveys revealed that visitors had doubts about how their platform’s subscription model worked. After adding new copy to their checkout page, the company improved conversions by 116%.

Where to find this:

5. Heatmaps

Heatmaps are a visual representation of how visitors behave on your website. When a visitor lands on a page, a tool records exactly where they scroll, click, or move their mouse to. The tool then aggregates data from thousands of visits and presents it in the heatmap forma.

In the scroll heatmap below, the red areas are ‘hottest’, meaning the areas that people see most. The blue areas are the ‘coldest’, meaning visitors see them the least. 


The picture below is a click heatmap. Notice that users are clicking on things other than buttons and links, showing that they are being distracted by elements that aren’t clickable.

How would you use heatmaps to test copy?

Heatmaps are very helpful for long-form sales pages and landing pages. They can give you specific insights that can help you determine how you can improve your copy. For example:

  • Discover which sections of copy you need to improve, move or delete. A scroll heatmap could show you that visitors aren’t reading a key section of your copy. 
  • Discover what people are curious about on a page. For example, heatmaps can show non-clickable elements that people click on (which is a good indication that they want more information).
  • Identify weak CTAs. If people are clicking on text other than your CTAs, that can indicate that the CTA copy isn’t clear or enticing.

What are heatmaps not good for?

Heatmaps aren’t great when used alone. They might show that visitors are looking or clicking in certain places, but they won’t tell you why people aren’t reading or responding to the copy on your page. 

However, scroll heatmaps are great when used in combination with other testing methods. Use them to inform what copy elements you change in split tests, or use them in combination with surveys to get deeper insights.

Examples of use 

Etsy store owner Jenni Waldrop used heatmaps to conversion-optimise her copy by identifying which copy sections were enticing people to opt-in to her email list.

Game developer Runescape saw on its heatmaps that many visitors were clicking on a non-clickable sub-heading. They concluded that the visitors wanted more information, and after adding more details saw a 10% increase in purchases. 

Where to find this?

6. was created by the people behind CXL. It’s a service that lets you upload a single piece of copy, like a web page or email, and get data around how real people react to it. 

The copy you submit is viewed by 20 testers from specific demographics groups (you can decide their location, age range, employment sector and much more). Each tester scores your copy according to its clarity and relevance, and leaves comments explaining their thoughts on different sections.

When the test is over, all the votes are compiled to give a score out of 100 for each block of copy. If the score is low, you can look at the comments from the individual testers to determine how you could improve the copy.

What is this good for?

The copytesting website states that the service is ideal for direct response copy like landing page and emails, and that right now it’s best for B2C.

It can help you optimise your copy by revealing:

  • Where copy falls flat
  • Which information is missing
  • What doubts users have when reading it

Therefore, it’s probably best used to supplement A/B testing by testing copy before it goes live, or determining why one copy variant performs better than another. 

Companies who rely on copy to generate sales and signups, like SaaS and ecommerce businesses, could use it to check for weaknesses before pushing copy live. The service could also be useful to agencies, copywriters and consultants who want to get unbiased feedback on copy (or impress their clients with legit data).

In my opinion, this testing method presents some advantages over simply showing your work to other copywriters. The testing audiences won’t hold back on their criticisms, nor will they tear your copy to shreds out of pure competitiveness!

What are the drawbacks of this method?

The results from Wynter also may not accurately show how your audience will respond. Because the test audience isn’t your real audience, the copy may not speak to their desires, pains and experiences.

For example, imagine your copy is selling private health insurance. A real prospect who is seeking health insurance because they recently lost a relative to a disease may feel differently about the copy compared to a copy tester who isn’t worried about health issues. Naturally, the test's accuracy will vary according to the product and audience.

In addition, the test audience might not fully understand your terminology if you’re selling a very niche product. That said, the Wynter website does explain that it can create custom audiences on request, so this may help with testing copy around niche products.

Where to find this?

7. User testing

user testing

User testing is a method that involves observing how real people navigate through a website or digital product. It usually involves asking a participant to complete a specific task, e.g finding a specific product and adding it to their basket. You then watch what they do, and/or ask the participant to say their thought process out loud.

This method is harnessed by UX professionals to reveal weaknesses in layout, design and copy. However, using websites like, you can optimise the method to focus on copy

For example, you can simply show the participant your home page and ask set questions like ‘how would you describe what’s on offer here?’ The resulting answers allow you to understand how people interpret your copy (this ends up being similar to the service that offers).

How can user testing help you evaluate your copy?

User testing is a great way to validate web copy, especially for complex feature or product pages, and even pricing pages

Use this method when:

  • You’re undertaking a conversion optimization project and want to identify if any of your copy leaves users confused or in doubt 
  • You’ve re-designed your website and want to evaluate it before going live
  • You’ve changed a critical part of your website, like your homepage headline or a key product page, and you want to be sure the new version will work

Drawbacks of this method

User testing is that it can be logistically hard to do. First off, you need to recruit impartial participants to give you feedback, and they need to be at least somewhat representative of your actual audience.  It can also be difficult to conduct testing in a way that doesn’t bias users. 

The other drawback is that what people say and do are two separate things. Just because your test audiences are positive about your copy, it doesn’t mean they would actually click ‘buy’ in real life. 

Examples of use in conversion optimisation projects

Conversion Rate Experts applied user testing methods to discover how they could optimise the website of an internet travel agent. Tests revealed that visitors found the site’s prices confusing, and they were often unsure about what was included in the price. 

Accordingly, the team added copy that clarified what the pricing included, resulting in a 19% conversion boost.

Where to find this:

8. Bonus: Eye tracking

Eye Tracking

I’m including this as a ‘bonus’ because it’s unlikely that any lone copywriter would use eye tracking…but theoretically it’s a method you could get copy insights from.

Eye tracking tools watch a person’s eyes to track where they look when viewing a specific material (like a web page). In the past, companies wanting to do eye tracking studies had to complete them in a lab using sophisticated tools, making the process very costly. 

However, sites like now offer SaaS eye tracking technology that works through your webcam. 

How could eye tracking help you test copy?

Eye tracking isn’t purely for analysing copy, but it could help solve certain copy problems when other research methods aren’t working. Used in conjunction with user testing methods, it can help you to:

  • Find out if vital copy is missing or not giving visitors enough information to navigate through the site (for example, because the data shows that visitors see buttons or menus but don’t use them to complete tasks)
  • Find out if something else on the page is distracting visitors from key copy (such as a CTA button)

Drawbacks of this method

Eye-tracking is not very practical for copywriters. The method can tell you whether your prospects are seeing a key piece of copy or not, but it probably won’t be helpful without additional insights from user testing. 

Eye-tracking can be useful in conversion-optimising web pages, but a lot of the optimisation it leads to is in the realm of design and UX, rather than copywriting.

Examples of use in conversion optimisation projects

British Gas used eye tracking to identify that their page needed to bring more attention to offer description and CTA copy, gaining a 52% conversion lift.

Where to find this?


If your business relies heavily on copy for selling or getting conversions, it’s important that your copy creates the right journey for prospects. This is especially true for online business and e-commerce, where innovative competitors are often only a click away.

And while copywriting is still an important skill to develop, it’s easier than ever to develop an effective process for creating copy that sells.

  • Use tools like heatmaps, session recordings, and on-site surveys to try and understand what needs aren’t being met by your existing copy. 
  • When you’re writing new website copy, use, 5-second tests and user testing to validate your ideas with real people.
  • Use split testing to validate your hypotheses for improving your copy and find out which variants get the best results.

While using these methods take a little time up-front, they eliminate the guesswork from creating and refining copy. Ultimately, that makes it much easier to  get the conversions (and profits) you want!


Main Image by OVAN from Pexels

Cover Image by Pixochris on Vecteezy

Photo by Jason Dent on Unsplash

More blogs

More to read