What is a website conversion rate?
A conversion rate is the percentage of visitors who take the action your website exists to produce. Divide conversions by visitors, multiply by a hundred: 1,000 visitors and 25 enquiries is a 2.5% conversion rate. For most UK service businesses the action that matters is an enquiry, because everything after the enquiry happens in a conversation rather than on the site.
Before you can ask what a good website conversion rate looks like, both halves of that fraction need pinning down. Which visitors count: all sessions, or only real people on pages that could plausibly convert? Which actions count: a submitted form, a tapped phone number, a booked call, a newsletter signup? Change either definition and the same website produces a completely different number, which is exactly why comparing your rate against someone else's goes wrong so quickly.
Why published conversion rate benchmarks mislead
Search for benchmarks and you will find confident tables of averages by industry. Treat them as entertainment rather than evidence, for three reasons.
They mix industries and business models. An "average" that pools e-commerce checkouts, SaaS free trials, and plumbing enquiries is averaging things that have nothing in common. Even within one sector, a national firm with brand recognition and a two-person local business face entirely different buying decisions, and their rates reflect that long before website quality enters the picture.
They mix traffic sources. Visitors from a search for "emergency electrician near me" convert at a multiple of visitors who wandered in from a social post. A site with mostly high-intent paid search traffic will post a rate that a content-heavy site can never match, and neither number says which website is better. Benchmarks flatten this into one figure and throw the most important context away.
They mix conversion definitions. One study counts qualified enquiries. Another counts any form fill, including newsletter signups and brochure downloads. A third counts add-to-basket. The studies rarely disclose their definitions clearly, and most are published by marketing platforms reporting on their own self-selected customers, which is not a neutral sample of businesses like yours. An average of inconsistent measurements is not a benchmark. It is noise with a decimal point.
The ranges commonly cited for service businesses
With all of those caveats stated, it would be evasive not to mention the numbers at all. Industry surveys and marketing platform studies commonly cite average website conversion rates somewhere in the 2% to 5% range for service businesses, with dedicated landing pages receiving paid traffic sometimes reported higher, and broad informational traffic often converting below 1%. These figures are attributed to the market, not to us, and we would not stake a decision on any of them.
What the ranges are useful for is a sanity check at the extremes. If your site converts at 0.2% on traffic that includes genuine buyers, something specific is broken and worth diagnosing. If a tool or an agency promises you 15%, be sceptical of the definition being used. Between those extremes, the benchmark tells you almost nothing that your own data cannot tell you better.
How to calculate your own conversion rate properly
A rate you can trust needs a clean numerator and a clean denominator, agreed once and never quietly changed.
Count every enquiry route, not just the form. Form submissions, phone number taps, call bookings, quote requests: each one should be recorded as a conversion event in your analytics. Google's own documentation on events in Google Analytics 4 covers how these are captured. Sites that only track the form routinely understate their real conversion rate, because a decent share of service business enquiries arrive by phone.
Pick one denominator and keep it. Sessions is the usual choice. Filter out your own visits and obvious bot traffic, otherwise the denominator inflates and the rate sags for reasons that have nothing to do with the site.
Segment before you judge. A single site-wide rate hides everything interesting. Split it by traffic source and by landing page and the story changes: the service pages might convert at 4% while the blog converts at 0.3%, and both can be healthy. This is also where problems localise. If one high-traffic page converts far below its siblings, that page is the leak, and our post on why websites fail to convert walks through the usual causes one by one.
Use a window long enough to mean something. Small sites see too few conversions for daily or even weekly rates to be stable. Three months of data gives you a baseline you can defend; one good Tuesday does not.
Why your own trend beats any benchmark
The question a benchmark pretends to answer is "am I doing well?". The question your business actually needs answered is "is this getting better?". Only your own trend can answer the second one, and the second one is the one connected to money.
A trend measured on a consistent definition is immune to every flaw the benchmarks suffer from. Your industry mix does not change month to month. Your conversion definition stays fixed because you fixed it. When your rate moves, something real moved: the traffic, the season, or the site. Compare months against the same month last year where seasonality bites, and annotate the dates you shipped changes, so that movement can be traced to causes rather than argued about.
A flat trend is information too. If traffic grows quarter after quarter while the conversion rate stays put, the website is holding steady rather than improving, and every pound spent on more traffic is buying more of the same leak.
The factors that move a conversion rate up or down
When a rate shifts, or refuses to, one of a fairly short list of factors is usually responsible. This is the pattern we see most often across the sites we audit:
| Factor | Pushes the rate up | Drags the rate down |
|---|---|---|
| Traffic intent | Visitors searching for the service you sell | Broad informational or social traffic |
| Message clarity | Service, audience, and next step stated plainly | Slogans and abstractions above the fold |
| Page speed | Passing Google's Core Web Vitals thresholds | Slow loads that lose visitors before the pitch |
| Proof | Named testimonials and specific results near the call to action | Anonymous quotes, stock imagery, no evidence |
| Form friction | Three fields and a clear promise | Long forms, required phone numbers, CAPTCHAs |
| Mobile experience | An enquiry that completes cleanly on a phone | Tiny text, mis-tapping, keyboard fights |
| Offer strength | A concrete, low-risk next step | A vague "get in touch" with no reason to act now |
Notice that only one row is about traffic. The other six are properties of the website itself, which is why two businesses buying identical clicks can convert at wildly different rates, and why quick exits deserve their own diagnosis: our post on why people leave websites so fast covers that half of the problem.
How to set a realistic improvement target
Set targets in relative terms against your own baseline, not in absolute percentages borrowed from a survey. Moving from 2% to 2.5% is a 25% relative improvement, and for a business with steady traffic that is a quarter more enquiries without spending another pound on marketing. Framed that way, modest-sounding movements justify serious attention.
A sensible sequence looks like this. First, establish the baseline: three months of clean data across every enquiry route. Second, pick the single most damaged factor from the table above, based on evidence rather than taste. A scored review such as our free website audit is one way to get that evidence quickly. Third, fix that one factor, note the date, and let a month of data accumulate before judging. One change at a time is slower in theory and faster in practice, because you learn what actually worked.
Be suspicious of anyone who guarantees a specific uplift. Honest conversion optimisation is a discipline of diagnosing, fixing, and measuring, and its results are earned per site, not promised in advance. Our guide to what conversion rate optimisation actually is walks through that audit-hypothesis-test-measure cycle in full. What you can reasonably expect is this: sites with an obvious broken factor tend to show measurable movement within a month or two of fixing it, and sites already performing well improve in smaller increments. Both are wins. Both beat chasing a benchmark that was never about your business in the first place.
Frequently asked questions
Is a 2% website conversion rate good?
It depends entirely on what you are counting and where the traffic comes from. Two percent of high-intent paid search visitors enquiring is very different from two percent of blog readers joining a newsletter. Judged against commonly cited service business averages it is unremarkable, but the more useful question is whether your own rate is higher than it was three months ago on the same definition.
Should I compare my conversion rate with competitors?
You cannot do it meaningfully, because you will never see a competitor's real numbers, their traffic mix, or what they count as a conversion. Public claims are marketing, not measurement. The comparison that actually pays is internal: this quarter against last quarter, page against page, and traffic source against traffic source, all measured on one consistent definition you control.
What counts as a conversion for a service business website?
Any action that starts a sales conversation: an enquiry form submitted, a call booked, a phone number tapped, a quote requested. Count every route a customer can take, not just the form, or you will understate your real rate. Softer actions such as newsletter signups are worth tracking separately, but mixing them into one number hides whether the site is producing actual enquiries.
How often should I review my conversion rate?
Monthly is the practical rhythm for most service businesses. Daily numbers are noise, because small sites see too few conversions for a single day to mean anything. Review the monthly rate against the previous month and the same month last year to allow for seasonality, and look at the quarterly trend before judging whether any change you shipped actually worked.