Quick answer: Conversion rate optimization (CRO) is the process of getting more of your existing visitors to take the action you want, such as buying, booking or sending an enquiry. It works in a cycle: measure what visitors do, find where and why they drop off, form a clear hypothesis, test a change, and keep what works. Done well, it makes every other marketing channel more profitable because you get more from the traffic you already pay for.
Most businesses put their effort into getting more visitors: more SEO, more ads, more social posts. That matters, but it ignores a simple point. If your website turns a small share of visitors into customers, doubling traffic doubles your costs as well as your results. Improving how well the site converts can grow sales without spending more on traffic at all.
This guide explains the basics in plain terms.
What counts as a conversion?
A conversion is any action that moves a visitor closer to becoming a customer. Your main conversion depends on your business:
- Online shop: a completed purchase
- Service business: a quote request, phone call or booked consultation
- Software or subscription: a free trial sign up or paid plan
- Publisher: an email newsletter sign up
There are also smaller steps along the way, often called micro conversions: adding to basket, downloading a brochure, watching a product video or starting a form. Tracking these helps you see where people lose interest.
How to calculate conversion rate
Conversion rate is the number of conversions divided by the number of visitors (or sessions), multiplied by 100. If 2,000 people visit in a month and 40 of them send an enquiry, your conversion rate is 2 percent.
Two cautions:
- Compare like with like. A conversion rate for all traffic hides big differences between, say, visitors from paid search and visitors from social media.
- Do not chase industry averages. Published “average conversion rates” mix very different businesses, offers and prices. Your own past performance is the useful benchmark.
Step 1: Set up proper measurement
You cannot improve what you do not measure. Before changing anything:
- Set up goals in Google Analytics. A goal can be a thank you page view, a completed checkout, or an event such as a button click. Universal Analytics makes event tracking easier, especially with Google Tag Manager.
- Turn on ecommerce tracking if you sell online, so you can see revenue and average order value as well as transactions.
- Build funnels. A goal funnel shows how many people move through each step, such as basket, delivery details, payment and confirmation, and where they leave.
- Track phone calls if calls matter to your business. Call tracking services can show which pages and channels lead to calls.
- Filter out spam. Fake referral traffic can distort your numbers. Set up filters so it does not skew your conversion rate.
Check the tracking works by completing a test conversion yourself.
Step 2: Find where visitors drop off
With measurement in place, look for the biggest leaks. Good places to start:
- Pages with lots of traffic and high exit rates, especially product, pricing and service pages
- Landing pages for ads with high bounce rates
- Funnel steps where a large share of people leave
- Device differences. If mobile visitors convert far less than desktop visitors on the same page, there is often a usability problem
- Forms with many fields or confusing questions
Focus on pages close to the conversion and pages with the most traffic. Small improvements there have the biggest effect.
Step 3: Understand why
Analytics shows what happens. It rarely shows why. For that you need qualitative research.
Heatmaps and scroll maps
Tools such as Crazy Egg and Hotjar show where people click and how far they scroll. You might find that visitors click an image that is not a link, or never scroll far enough to see your call to action.
Session recordings
Watching recordings of real visits, with personal details hidden, can be eye opening. You see people hesitating, going back and forth, or struggling with a form field.
On site surveys and polls
A one question poll such as “What is stopping you from booking today?” can reveal objections you never thought of: unclear pricing, worries about delivery, or a missing payment option.
Talk to customers and staff
Your sales and support teams hear objections every day. Ask them what customers ask before buying, and what confuses them.
User testing
Ask a few people who match your audience to complete a task on your site while thinking out loud. Even a handful of sessions usually uncovers obvious problems.
For a structured way of reviewing a site, our CRO audit guide walks through a full checklist.
Step 4: Form a hypothesis
A good hypothesis connects a problem to a change and an expected result. For example:
“Visitors on mobile are abandoning the quote form at the phone number field. Making the phone number optional will increase completed quote requests from mobile visitors.”
Writing it down this way keeps tests focused and makes it easier to learn from the results, even when a test fails.
Prioritise your ideas. A simple approach is to score each one on potential impact, confidence that it will work, and how easy it is to do. Start with high scoring ideas.
Step 5: Test the change
A/B testing
In an A/B test, half your visitors see the current page and half see the new version. You compare conversion rates. Tools include Optimizely, Visual Website Optimizer (VWO) and Google Analytics Content Experiments, which is free.
Key rules:
- Test one main idea at a time, so you know what caused the difference
- Run tests for full weeks, to cover differences between weekdays and weekends
- Wait for enough data. Stopping a test early because one version is ahead is one of the most common mistakes
- Decide your success metric before you start
Our A/B testing guide goes into more detail on sample sizes and reading results.
When you do not have enough traffic to test
Smaller sites often cannot get reliable test results in a sensible time. That is fine. Make changes based on clear evidence from research, such as fixing a broken form or answering a common objection, then compare performance before and after over a longer period. Be honest that the result is less certain.
Step 6: Learn and repeat
CRO is not a one off project. Record every test, what you expected, what happened and what you learned. Over time, this record shows what your audience responds to and makes future ideas better.
Common quick wins
Every site is different, but these issues come up again and again:
- Unclear value proposition. Visitors should understand what you offer, who it is for and why choose you within a few seconds.
- Weak or hidden calls to action. Make the next step obvious, and use specific wording such as “Get my free quote” rather than “Submit”.
- Long forms. Ask only for what you need now. You can collect more later.
- Missing trust signals. Reviews, guarantees, clear contact details, recognised payment logos and security information all reduce hesitation.
- Surprise costs. Show delivery charges early. Unexpected costs at checkout are a common reason for abandoned baskets.
- Slow pages. Visitors leave slow pages, particularly on mobile connections.
- Poor mobile experience. Small buttons, tiny text and forms that are painful to fill in on a phone.
For pages built to convert visitors from ads or email, see our guide to landing page optimization.
Mistakes to avoid
- Copying what worked for someone else. A change that helped another site may hurt yours. Test it.
- Testing trivial things first. Button colour tests rarely matter as much as the offer, the copy or the form.
- Ignoring revenue. A change that increases sign ups but brings lower quality leads may not be a win.
- Calling tests too early. Short tests often produce results that disappear later.
- Treating CRO as tricks. Lasting improvements come from making the site clearer and more helpful, not from pressure tactics.
FAQ
What is a good conversion rate?
It depends on your industry, price, traffic sources and what you count as a conversion. The best benchmark is your own site’s past performance.
How much traffic do I need for CRO?
You need a fair amount of traffic for reliable A/B tests. Without it, you can still improve conversions through research led changes and careful before and after comparison.
Is CRO the same as UX design?
They overlap. UX design is about making a site easy and pleasant to use. CRO focuses that work on measurable business goals and tests the results.
Get more from the traffic you have
If your site gets visitors but not enough enquiries or sales, a structured CRO programme is usually the most cost effective fix. See how we help businesses turn visitors into customers.
