Website conversion optimization

Out of a hundred visitors to a website, usually only a few get in touch. Each of those inquiries is called a conversion. Conversion optimization is about getting more of them, without spending a cent more on ads. Its longer name is conversion rate optimization, or CRO for short.

The levers

Message

Does the offer match the question?

The first lever comes before any button. Does a visitor understand within seconds what’s on offer and who it’s for? And why they should get it here?

Trust

Proof where it counts

Case studies, testimonials and real numbers belong right where people make up their minds, and so does a real person to contact. Online stores show how much trust matters at checkout. In one US survey, 19% of shoppers who abandoned a checkout said they didn’t trust the site with their card details.3

Action

One clear next step

Every page leads to one main thing, for example an inquiry or a call. When several equally loud buttons sit side by side, attention gets split between all of them.

Forms

Ask only what’s needed

Every field in a form is a small hurdle. The average online checkout had 11.3 fields in 2024, although most stores could manage with 8.3 For an inquiry, three things are often enough, namely a name, a way to reach the person and what they need.

Speed

Tenths of a second

In a study commissioned by Google, a mobile site that loaded 0.1 seconds faster brought 8.4% more sales for retailers. For travel sites it was 10.1% more bookings.2 The data comes from 2019 and covers 37 brands and over 30 million visits.

Do you have enough visitors for an A/B test?

Visitors per variant

21,000

Test duration

about 14 months

That’s too long for a clean test. You’re better off checking the page for weak spots and making the most important changes right away.

Find the weak spots ↓

In an A/B test, half of your visitors see the current page and the other half see a changed version. For the result not to be a fluke, each half needs enough visitors. The calculator uses the usual statistical standards for that.4, 6 The hoped-for lift is relative, so 20% means going from 2.0% to 2.4%, for example.

The process

  1. Research

    Where do visitors come from, what do they search for, and where do they drop off? On top of that, someone from outside looks at the message and at how the site guides people.

  2. Hypothesis

    Each idea gets written down in one sentence, following the pattern “If we change X, Y goes up, because Z.” That way you can check it later.

  3. Make the change

    With enough traffic, the change runs as an A/B test. Otherwise it goes live directly, and you compare cleanly before and after.

  4. Measure

    You compare periods of the same length and keep an eye on seasons and ad campaigns. Only then comes the next change.

What to measure and which tools to use

  • Define the conversion

    First you decide what counts as success. If a page has several goals, each one gets counted separately.

  • Events in analytics

    Your analytics needs to count it when someone submits the form, taps the phone number or reaches the thank-you page. Without that, every rate is just an estimate.

  • Traffic by source

    Visitors from search, ads, email and AI assistants like ChatGPT behave differently. That’s why you look at each group separately.

  • AI search

    Visits from AI answers deserve a look of their own. Do they land on the right page, and does that page lead to an inquiry? There’s more in our GEO guide.

  • Analytics tools

    Visitor stats usually run on Google Analytics, Google’s own analytics tool, or on Matomo. Matomo is free and can run on your own server, so the data stays in-house.8 Both count inquiries as goals once you set that up.

  • Heatmaps and testing tools

    Heatmaps show where visitors click and how far they scroll, and Microsoft Clarity offers them free for good.7 A/B tests need a dedicated tool like Optimizely or VWO, because Google shut down its free testing tool, Optimize, in September 2023.9

What conversion optimization costs

RouteScopeCost
In-houseYour own team does the research, comes up with ideas and makes the changes, with tools for click maps and testsstaff time, tools $10 to $500 a month5
AgencyOngoing research, testing and reporting$800 to $10,000 a month5
AI Quick CheckAn automated report on the homepage and some subpages, with a plan that starts with the most important change. Visitor data and testing aren’t part of it.$9.99 one time

The full picture of optimization costs is on our page on website optimization services.

Know now what your website needs.

Our AI Quick Check reviews your website and tells you what to fix first. Built by web and SEO experts, in your inbox within minutes.

  • Audience analysis
  • Market analysis
  • Content analysis
  • UX analysis
  • SEO
  • AI search
  • Action plan

$9.99 one-time payment · no subscription, no account

PDF by email, usually in under 5 minutes

First page of a sample report with the headline what the website needs now and three steps in order

Common questions about conversion rate optimization

What is website conversion optimization?

Conversion optimization is everything that gets more visitors to do what a site is built for. That means getting in touch, calling, booking or buying. It doesn’t take more visitors, just a better page.

What is conversion rate optimization (CRO)?

It is the longer name for the same work. The name puts the focus on the share of visitors who get in touch or buy, which is the conversion rate.

Is conversion rate optimization worth it?

Usually, because every improvement keeps working for all future visitors. Take a page with 3,000 visitors a month and a 2% conversion rate, which makes 60 leads. If the rate rises to 2.4%, that is 72 leads, so twelve more each month without a dollar more spent on ads.

How is CRO different from SEO?

SEO gets more people to find the website on Google. CRO deals with what those visitors do once they arrive. The two are linked, because a page that answers exactly the question from the search usually brings in more leads too.

What is a good conversion rate?

Across more than 41,000 landing pages the median is 6.6%, and depending on the industry it runs from 3.8% to 12.3%.1 For a whole website, that figure only compares roughly, though. Your own rate before and after a change tells you far more than any benchmark.

How much traffic do you need for an A/B test?

That depends on how well the page converts today and how much of a lift you’re hoping for. For example, at a 2% conversion rate and a hoped-for lift of 20%, each version needs about 21,000 visitors. The calculator above shows how long that would take on your site.

Do you need a CRO agency?

Not necessarily. An agency pays off mainly when there is enough traffic for ongoing A/B tests, and that typically costs $800 to $10,000 a month.5 For smaller sites, a focused review for weak spots is often enough, with your own team or web agency making the most important fixes.

Does conversion optimization work the same for ecommerce?

The levers are the same, but in an online shop the checkout carries the most weight. The average checkout in 2024 had 11.3 form fields, while most shops would only need 8.3 Every field cut there makes buying a little easier.

What if there isn’t enough traffic to test?

Then you check the page for weak spots. The changes that promise the most go live right away. Afterwards you compare the rate before and after, over periods of the same length.

How much does page speed matter?

A lot. In a study commissioned by Google, a mobile site that was 0.1 seconds faster brought retailers 8.4% more sales. The data comes from 2019.2

Sources

  1. Unbounce, Conversion Benchmark Report (“The median conversion rate across all industries is 6.6%”, 41,000+ landing pages), accessed Nov 5, 2026: unbounce.com
  2. web.dev, Milliseconds make millions (“A 0.1 second improvement of mobile site speed increases conversion rates by 8.4% for retail sites and 10.1% for travel sites”; 37 brands, 30+ million sessions; study by 55 and Deloitte, commissioned by Google), accessed Nov 5, 2026: web.dev
  3. Baymard Institute, checkout form fields (“the average in 2024 was 11.3 form fields”, “most sites need only 8 form fields”) and reasons for cart abandonment (“19%: I didn’t trust the site with my credit card information”), accessed Nov 5, 2026: baymard.com, baymard.com
  4. Optimizely, Sample Size Calculator (“95% is an accepted standard for statistical significance”), accessed Nov 5, 2026: optimizely.com
  5. WebFX, Conversion Rate Optimization Pricing (“anywhere from $800 to $10,000 monthly on CRO services from an agency”, tools “$10 to $500 per month”), accessed Nov 5, 2026: webfx.com
  6. Wikipedia, Power (statistics) (“many researchers and funding bodies assess power using 0.80 (or 80%) as a standard for adequacy”), accessed Nov 5, 2026: en.wikipedia.org
  7. Microsoft Clarity, FAQ (“Clarity is a free service forever.”), accessed Nov 6, 2026: learn.microsoft.com
  8. Matomo, On-Premise (“Self-host Matomo Analytics on your own servers to ensure data remains in your hands.”, “free to download”), accessed Nov 6, 2026: matomo.org
  9. Google, Optimize Help (“Google Optimize and Optimize 360 are no longer available as of September 30, 2023.”), accessed Nov 6, 2026: support.google.com