Conversion rate optimisation: more enquiries from the traffic you already have.

Your website gets visitors. The numbers in Google Analytics look respectable. Yet the enquiries lag behind what that traffic should produce. The usual answer is more traffic: a bigger ad budget, another SEO push. That fills a bucket with a hole in it.

Conversion rate optimisation starts at the other end. It looks at the people already on your site and asks why they do not take the next step. At Viralistic we design and build websites in Figma and Astro, with conversion as the starting point rather than an afterthought. This guide covers what CRO means, where visitors drop off, a process you can repeat and when A/B testing is the right tool. The honest answer to that last one: less often than most guides suggest.

What is conversion rate optimisation?

Conversion rate optimisation is the systematic process of increasing the percentage of website visitors who complete an action that matters to your business. American English spells it conversion rate optimization, both are shortened to CRO, and many people simply say conversion optimisation. A conversion can be a quote request, a booking, a purchase, a phone call or a newsletter sign-up.

CRO is research before it is design. You combine analytics, user research and experiments to find out what stops people, then change the page based on evidence rather than opinion.

It helps to separate two kinds of conversion. A macro conversion is the end goal, such as an enquiry or an order. A micro conversion is a step towards it: a click to your services page, a case study read, a contact form opened. Micro conversions show you where in the journey people give up.

How do you calculate your conversion rate?

Your conversion rate is the number of conversions divided by the number of visitors, multiplied by 100. A site with 1,000 visitors and 20 enquiries in a month converts at 2 percent.

Pick one definition of “visitor” and keep it. Sessions and users are different numbers in Google Analytics 4, and switching between them breaks every comparison. For an online shop, also track revenue per visitor. A higher conversion rate with smaller orders can leave you with less.

What is a good conversion rate?

There is no universal good conversion rate. It depends on your sector, your price point, where your traffic comes from and what you count as a conversion. A firm that sells five-figure engagements will see fewer enquiries per hundred visitors than a shop selling impulse buys, and that is normal.

The most useful benchmark is your own baseline. Measure your current rate per page and per traffic source, then check whether it rises after each change. An average from another industry tells you very little about your site.

Why conversion optimisation matters more now

CRO makes every visitor worth more, wherever they come from. Take a site with 10,000 visitors a month and a 1 percent conversion rate: 100 enquiries. Lift that to 1.5 percent and the same traffic brings 150. That is 50 percent more enquiries without spending more on ads.

Clicks are also getting harder to earn. In a Pew Research Center analysis of the browsing data of 900 US adults in March 2025, people who saw a Google AI summary clicked a traditional search result on 8 percent of visits, against 15 percent when no summary appeared. When fewer people arrive, the ones who do carry more weight.

That is why search visibility and conversion belong in one plan. SEO and advertising bring the right people to your site, and CRO makes sure that investment pays off. Our article on SEO vs GEO explains how classic search and AI answers now share that first step.

Why visitors leave without converting

Most conversion problems fall into four groups: an unclear message, too little trust, too much effort and no obvious next step. Run your most important pages past these four questions before you test anything.

Do visitors understand what you offer?

Someone landing on your homepage should see within seconds what you do, who it is for and why it matters. A vague headline like “We make the difference” leaves the visitor to do the work. A clear promise, in your client’s own words, does it for them. This is where conversion really starts: with your brand script, long before anything is tested.

Have you given them a reason to trust you?

Trust comes from evidence: case studies with real names, client logos, reviews, the faces of the people behind the business and clear contact details. The higher the price, the more this weighs. A premium service that looks anonymous loses exactly the visitors who are worth the most.

Does the next step take too much effort?

Every extra form field, every slow page and every button just out of thumb’s reach on mobile costs you people. Speed counts. In a study commissioned by Google, Deloitte and the agency 55 tracked 37 European and American brand sites and more than 30 million sessions. When mobile speed improved by 0.1 seconds across four metrics, progression to the contact page rose by 20.6 percent on luxury sites. Lead generation sites saw 21.6 percent more users reach the form submission page (Milliseconds Make Millions). It is one study, measured over 30 days at the end of 2019, but the direction is clear.

Google’s yardstick for a good experience is the Core Web Vitals. The largest content should appear within 2.5 seconds (LCP), the page should respond to a tap or click within 200 milliseconds (INP) and the layout should barely shift (a CLS of 0.1 or less). Those targets should hold for at least 75 percent of page views.

Is there one clear next step?

A page with five buttons competing for attention asks the visitor to choose. Often they choose nothing. Give each page one main action, make it visible without scrolling and repeat it where it makes sense. “Plan a call” says more than “Learn more”.

The conversion rate optimisation process, step by step

Conversion rate optimisation works as a repeating cycle: measure, research, form a hypothesis, improve or test, and learn. Each round sharpens the next. The first three steps together are what most agencies sell as a CRO audit.

1. Check that your measurement is right. Define in Google Analytics 4 which actions count as conversions, then confirm that form submissions, clicks on your phone number and bookings are actually recorded. If the data is wrong, every later step is too.
2. Find where people drop off. Follow the route from landing page to enquiry. Which page loses the most visitors? Which pages draw plenty of traffic but produce little? Heatmaps and session recordings show what people actually do there.
3. Ask real people. Numbers tell you what happens, conversations tell you why. Ask a few people from your audience to complete a task on your site and watch. Jakob Nielsen found that testing with five users uncovers about 85 percent of usability problems, and he recommends several small rounds over one large study.
4. Write a hypothesis. For example: “Because mobile visitors often abandon the form halfway, we expect three fields instead of eight to produce more enquiries.” A good hypothesis names an observation, a change and an expected effect.
5. Decide what to tackle first. Score your ideas with a simple model such as PIE (potential, importance, ease) or ICE (impact, confidence, ease). Start with what promises a lot and costs little.
6. Fix or test. Obvious faults you simply fix. A broken button does not need an experiment. When you are choosing between two good options and have enough traffic, an A/B test is the right tool.
7. Record what you learn. Note the hypothesis, the result and the lesson for every change, including the ones that made no difference. After a year you own a body of knowledge about your clients that nobody else has.

A/B testing without fooling yourself

In an A/B test you split visitors at random between the current page (A) and a variant (B), and measure which one performs better on a single goal chosen in advance. It replaces opinion with evidence.

The payoff can be large. Ron Kohavi and Stefan Thomke describe in Harvard Business Review how a Microsoft employee working on Bing proposed in 2012 to change how ad headlines were displayed. The idea was rated low priority and sat untouched for more than six months. When an engineer finally tested it, revenue rose by 12 percent, worth more than $100 million a year in the United States alone. It became the best revenue-generating idea in Bing’s history, and nobody saw its value until the test.

How much traffic do you need?

More than most people expect. Evan Miller gives a rule of thumb in How Not To Run an A/B Test: the number of visitors per variant is about 16 × p × (1 - p), divided by the square of the difference you want to detect. Here p is your current conversion rate. The rule fits the usual settings of 5 percent significance and 80 percent power.

Worked out for a page that converts at 2 percent today:

If that page gets 3,000 visitors a month, a test built to detect a 20 percent lift runs for more than a year. Small differences need enormous traffic. Miller’s site also has a free sample size calculator for your exact numbers.

Do not stop a test the moment it looks good

The most common A/B testing mistake is checking results along the way and stopping as soon as the variant seems to win. The maths behind “95 percent confidence” assumes you fixed the sample size in advance. In Miller’s worst-case example, with a significance check after every visitor, 26.1 percent of tests declare a winner when there is no real difference. You think you are working with a 5 percent chance of a fluke. It is more than five times that.

Decide the sample size before you start and wait until you reach it. Run tests over whole weeks, so a busy Monday or a quiet weekend does not skew the result, and test one main goal at a time.

What if you do not have enough traffic?

Then A/B testing is usually the wrong tool, and that is true for many premium brands and B2B sites. These approaches still work:

• Watch five people from your audience complete a task and note where they get stuck.
• Review heatmaps and session recordings for patterns.
• Ask clients why they chose you, and what nearly stopped them.
• Test bold changes rather than small ones. A new page structure moves the needle more than a new button colour, and bigger effects need fewer visitors to prove.
• Measure micro conversions, such as clicks to your contact page. They happen more often than enquiries, so you learn faster.

When you ship a change without a test, compare a few weeks before and after, and allow for seasons and campaigns that distort the picture.

CRO tools worth knowing

Most websites can start conversion rate optimisation with a small, largely free set of tools.

Google Optimize, Google’s free testing tool, was switched off on 30 September 2023, so older guides that recommend it are out of date. Whichever testing tool you choose, make sure it loads variants quickly. A script that slows the page or flashes the original version first distorts your result.

Heatmaps and recordings capture the behaviour of real people. Microsoft states that Clarity needs valid consent for its full feature set, session recordings included, in jurisdictions that require consent, which covers visitors from the EU. Align your tools with your cookie banner and privacy statement.

CRO for premium and B2B brands

For premium brands, conversion is about calm and confidence. Countdown timers, pop-ups that beg for an email address and “only 2 places left” banners sometimes work for impulse purchases. For a brand that sells on quality, they cost credibility.

What works instead: a promise that is clear at once, proof that you have done this before, a site that feels fast and flawless, and a next step that sounds personal. Our article on luxury websites that convert shows how that combines with distinctive design. Tailoring the experience to each visitor is the next layer, which we cover in our guide to website personalisation.

How we help

At Viralistic, conversion is built into the design. We start with your brand script and a wireframe, so the route to an enquiry works before anything is styled. Then we design in Figma and build the site in Astro: fast, light and easy to measure. If you already have a website, we look together at where visitors drop off and which improvements come first.

Webdesign projects start around EUR 4,000, and larger CMS-driven platforms can reach EUR 20,000 or more. Our guide to what a website costs compares that with every other route. Read how we work in our webdesign approach, or see how we bring the right visitors to those pages through SEO and AI search.

Frequently asked questions

Is a 2% conversion rate good?

It depends on what you sell and what you count. For an online shop with low prices, 2 percent can be ordinary. For a firm selling high-value services, 2 percent of visitors sending an enquiry can be strong. Compare against your own baseline per page and traffic source, and judge the rate alongside lead quality and revenue, not on its own.

What is the difference between CRO and SEO?

SEO helps the right people find your website through Google and AI answers. CRO makes sure those visitors then get in touch, book or buy. They reinforce each other: more traffic to a page that converts poorly achieves little, and a page that converts well but is never found achieves nothing either. The strongest results come from planning both together.

What does a conversion rate optimisation specialist do?

A CRO specialist finds out why visitors do not convert and fixes it. The work covers checking analytics, analysing the funnel, reviewing heatmaps and recordings, running user tests, writing hypotheses and running A/B tests where traffic allows. A good specialist works closely with design, copy and development, because most improvements are changes to the page itself.

How much does conversion rate optimisation cost?

It depends on the approach. The basic tools, such as Google Analytics 4, Search Console and Microsoft Clarity, are free. A/B testing platforms set their own prices, usually based on your traffic. At Viralistic, conversion is part of webdesign: projects start around EUR 4,000, and larger CMS-driven platforms can reach EUR 20,000 or more.

How long should an A/B test run?

An A/B test runs until it reaches the sample size you calculated in advance, ideally over several whole weeks. Do not stop early because a variant looks like it is winning, since repeated checking sharply raises the chance of a false winner. On a low-traffic page a reliable test can take months, and user testing is often the better choice.

Is CRO only for e-commerce?

No. For service firms, B2B companies and premium brands, a conversion is an enquiry, a call or a booked meeting. That is where CRO matters most, because a single client is worth a lot. In the Deloitte study, lead generation sites also saw clearly more visitors reach the form once the site became faster.