What is a good conversion rate for a website?

The short answer: about 5% of visits, for a site that wants enquiries and counts phone calls. About 2% of visits, for a UK shop. But you can only judge your own rate if your tracking counts every enquiry, including calls and emails.

What is a conversion rate?

It’s the share of visits to your site that end in the thing you want. That might be a sale, an enquiry, a call or a booking.

The sum is simple: conversions ÷ visits × 100. Say 1,000 visits bring in 30 enquiries. Then your rate is 3%.

The hard part isn’t the sum. It’s what you count as a conversion, and what you count as a visit. This is why tracking matters most.

What is a normal conversion rate for a website?

For an enquiry site, about 5% of visits, if phone calls are counted. For a UK shop, about 2% of visits end in a sale.

Across the internet, you’ll see “2–5% is good”. I can’t find a study behind it, so I don’t use it. Use the table above instead.

Is my conversion rate good?

Put in last month’s numbers. You’ll get your rate, how far chance alone could move it, and how it compares with sites like yours.

Struggling to find your numbers? Get in touch with us.

The five things I check before I tell a client their rate is fine

Yes, more than anything else. Paying is a bigger step than asking a question. So shops convert at well under half the rate of enquiry sites.

Shopify shops sit lower still. Line up 438 stores by rate, and the one in the middle converts 1.3%. Only the top 10% reach 3.6%. So a shop at 3% is doing well. An enquiry site at 3% may not be.

These are only averages. Wherever you start, the goal is the same: push your own rate as high as it will go.

This is the check that matters most. What do you count? For one of our clients, the same October reads 1.7% or 8.5%.

A click on “Book Now” shows someone is keen. But it isn’t a booking. So I report both numbers, not just the one that looks best.

Counting only forms can fool you the other way, too. Take a B2B manufacturer we work with. In the same month, 60 visits had a click on an email address. And 25 had a click on a phone number. Count forms only, and it converts 2.5%. Count every way of getting in touch, and it’s 4.6%. Calls matter too. In Ruler’s data, they were over half of all leads in legal (56%) and professional services (53%).

My rule is simple. Count every way someone can get in touch. Divide by visits. Then make sure you track the same number, the same way, every month. And the more you track, the more you learn about how your visitors behave.

Get the detail of your tracking right. Track call clicks, form sends and add-to-baskets as separate events, so you can tell them apart.

Yes. On enquiry sites, visits from paid search convert at 5.4%. Visits from paid social convert at 2.1%. People who searched for what you do are further along. They’re ahead of people who scrolled past an ad.

Why the gap? Someone who searches is asking for what you sell, right now. A social ad interrupts someone who was doing something else. The first person arrives warm. The second arrives cold, and needs more convincing before they get in touch.

Device matters too. Across 99 billion visits, desktop converted 74% higher than mobile. So a big social campaign can pull your rate down. So can a month with more visits on phones. Yet the site itself may be fine.

Split your rate by source and device before you judge it.

A landing page has one job and one button. The typical landing page converts 6.6%. So compare a landing page with other landing pages, not with your whole site.

How you write matters too. Landing pages written for a reading age of 10 to 12 converted at 11.1%. Pages written at a professional level converted at 5.3%. That’s why I look at the rate page by page. A home page, a service page and a blog post do different jobs.

Judge each page against pages that do the same job.

Yes. The bigger the decision, the lower the rate. It’s the main reason some businesses convert lower than others. Legal and software enquiry sites average nearly 8%. Travel sites average under 2%.

Big decisions convert lower. That’s normal.

How we put the five checks together

We work with a wide range of businesses across our portfolio: shops, enquiry sites, B2B firms and event companies. Here’s how we put the five checks together for any of them. Under each check are the real numbers from one of them: a B2B manufacturer that sells to architects, in October 2026.

1. What it asks for: a sale or an enquiry. We compare shops with shops (about 2%) and enquiry sites with enquiry sites (about 5%).The manufacturer: an enquiry site, so we compare it with about 5%.
2. How we count: every form, call and email click, divided by visits. Counting forms alone can make a healthy site look half as good as it is.The manufacturer: forms only, 2.5%. Every form, call and email click, 4.6%.
3. Where they come from: we split the rate by channel and device. Search usually converts higher than social, and desktop higher than phones. So a busy social month can pull the rate down on its own.The manufacturer: Google search 5.4%, direct visits 3.9%. Desktop 5.1%, phones 2.8%.
4. The page they land on: we compare pages that do the same job. Home pages with home pages. Product or project pages with each other.The manufacturer: visits that start on the home page 5.5%; on a project page 1.5%. People browse project pages, so we judge them against each other.
5. Size of the decision: we use the industry’s own average, not “2–5%”.The manufacturer: construction and engineering sites average 4.9%.

Then one more sum. With too few visits, chance alone can move a rate a long way. If the industry average sits inside that range, the rate is normal. The manufacturer’s 4.6% from 7,468 visits could really be anywhere from 4.2% to 5.1%. The industry’s 4.9% sits inside that, so its rate is normal, not low. Then the question isn’t “is it bad?” It’s “where do the rest of the visitors go?”

How do you find out why your conversion rate is low?

Rule out the three things that aren’t the site first: how you count (check 2), where visitors came from (check 3), and too few visits (the range in the checker). If it’s still low, watch what people do before you change anything.

The rate tells you something’s wrong. Tracking what people do tells you where they leave, and why. This is how I find it.

1. Map the journey

   These are the steps from arriving to converting. The page they land on. The page that sells. The form or checkout. The thank-you page. That’s your funnel.

2. Find the step where most people leave

   That’s the leak. Fix it first.

3. Watch what they do there

   Every click, error and wait at that step. This is where the reason shows up.

4. Fix it, then measure before and after

   Same count, same base, same length of time. That’s the only way to know a fix worked.

Here’s what that looked like for a luxury fashion shop we work with. Shoppers got to checkout, then got stuck at the payment step. We tracked what they did there. Half of them clicked “Continue to Payment” twice. The button gave no sign it was working, so people thought the site had frozen.

Shoppers who clicked “Continue to Payment” twice

Before1 in 2After1 in 559% fewer, in the 17 days after the fix

Time from basket to payment

50% fasterTimed before and after the fix

We made every checkout button show it was loading. Then we fixed the errors behind the drop-offs, the most common first. That’s a common story.

After any fix, the best benchmark is your own last month, counted the same way. We track that for clients every month with Insights. That means every form, call and email click. And what each change did to them.

Find the step where people leave. Watch what they do there. Fix the biggest leak first. Then measure before and after.

Quick answers

Is my conversion rate good?

It depends on what your site asks for, and your industry. For a UK shop, about 2% is average. 3% or more puts you near the top 10% of Shopify shops. For an enquiry site, about 5% is average. But that runs from under 2% in travel to nearly 8% in legal and software. Before you judge yours, check you’re counting calls and email clicks, and that you have enough visits. The checker above does the sums for you.

How many visits do I need to judge my conversion rate?

More than most people think. At 500 visits, a 2% rate could really be anywhere from about 1.1% to 3.6%. At 10,000 visits, it’s 1.7% to 2.3%. The checker works out the range for your own numbers.

What is a bad conversion rate?

One that’s falling while your visits stay the same. One low figure can come from how you count. Or from where your visitors came from, or too few visits.

Should I count phone calls as conversions?

Yes, if people call you from the site. In Ruler Analytics’ data, calls were over half of all leads in legal and professional services.

Should I use sessions or users?

Either, as long as you always use the same one. The user rate is always higher. That’s because people visit more than once.

What is a good lead conversion rate?

That usually means how many leads go on to buy. That’s a sales question. This guide is about how many visits become leads.

Sources

1. Ruler Analytics, Conversion Rate Benchmarks: Based on 5+ Million Conversions Tracked Across 13 Industries, 26 May 2026.
2. IRP Commerce, Ecommerce market data, September 2026.
3. Littledata, Average ecommerce conversion rate (all devices), 4 November 2026.
4. Unbounce, Conversion Benchmark Report, 2024.
5. Contentsquare, Conversion Rates in 2026: Benchmarks, AI, and What’s Driving Growth, 9 March 2026.
6. Google, Analytics Help: key event metrics and the Traffic acquisition report, marking key events and the Landing page report.
7. First Page Sage, B2B Conversion Rates By Industry – 2026, updated 18 September 2025.
8. Baymard Institute, Cart Abandonment Rate: 50 Statistics & Why People Abandon, updated 28 October 2026.
9. Coullweb, Insights data, real visitors only: a training and conference company (4–31 October 2026, and “Remind me” requests 4 October–8 November 2026), a B2B manufacturer (5–31 October 2026) and a luxury fashion shop’s checkout (17 days after fixes went live on 15 October 2026).

Links checked on 8 November 2026. Baymard and Google checked on 9 November 2026.

Written by Alex Coull

Coullweb is a small website team in Basingstoke. Since 2015 we’ve designed, built and looked after sites for 58+ businesses. For us, launch is the start. We measure where visitors get stuck. We fix the biggest problem first. Then we check it worked before we move on to the next. About us.

P.S. If you only do one thing from this guide, make it the 2-minute check. In GA4, go to Admin, then Data display, then Events, and open Key events. If phone and email clicks aren’t there, your real rate is higher than the one you’re worried about. If they are there and your rate still looks low, book a Friction Audit and we’ll find the leak.