What is a conversion rate, and what is a good one?
The short answer
A conversion rate is the share of visitors who do the thing you wanted — enquire, book, call, buy. For most small business sites two to five per cent is normal and anything above that is good, but the number matters far less than the direction. A site converting one per cent of a hundred relevant visitors beats one converting five per cent of twenty.
On this page
The definition, and the part people skip
Conversion rate is: out of everyone who arrived, how many did the thing you wanted.
Twenty enquiries from a thousand visitors is two per cent. That is the whole calculation.
The part people skip is deciding what "the thing" is. For a shop it is an order. For most small businesses it is not a sale at all — it is an enquiry, a booking request, a phone call, a directions tap, a menu view. If you have not named your conversion, you cannot have a rate, and most businesses that ask this question have not.
Name one primary action per page. If a page has three equally important goals it has none — see what is a call to action.
What counts, when you do not sell online
For a business that takes enquiries rather than payments, these are the honest conversions:
- A form submission — the easiest to count
- A phone call from the site — countable if the number is a tap-to-call link
- A WhatsApp or message tap
- A booking request
- A directions tap, which for a local business is close to a purchase
Two of those never appear in website analytics unless you deliberately track them, which is why so many local businesses believe their site does nothing. The site is working; the outcome is happening on a phone call nobody counted.
Your Google Business Profile reports calls and directions for free — see what is Google Business Profile.
Why the benchmark you were quoted is useless
Somebody has probably told you the average is two per cent, or three, or five. The number is close to meaningless, and here is why.
It depends entirely on the traffic. A page reached by people searching your business by name might convert at thirty per cent. The same page reached from a broad social campaign might convert at half a per cent. Same page, sixty-fold difference, and neither number tells you whether the page is good.
It depends on the ask. "Ring us for a quote" and "buy this now" are not comparable actions.
And industry averages are averages of everyone, including businesses whose sites are broken. Being average is not the goal.
The number worth tracking is your own, over time, from the same source. Whether this quarter beats last quarter is a real signal. Whether you beat an internet average is not.
How to work yours out without any software
You do not need a tracking setup to start.
- Count the outcome for a month — enquiries, bookings, calls, whatever you named. You probably already know this number.
- Get visitor numbers for the same month from whatever analytics you have.
- Divide. That is your rate.
- Write it down and do it again next month.
Two months of that tells you more than any benchmark, because you are comparing like with like: the same business, the same site, the same audience.
What actually moves it
In rough order of how much they change the number for a small business:
The contact form actually working. Test it. A silently broken form is a conversion rate of zero that looks like a quiet month — see my contact form stopped working.
A price, or a range. The single biggest lift available to most small business sites. People leave because they could not tell whether they can afford you.
Speed. Under three seconds on a phone on mobile data. Above that you lose people who never saw anything you wrote.
One clear action per page, worded as what the visitor gets rather than what the form does.
Answering the question they arrived with. Somebody who searched a specific thing and landed on a generic page leaves — which is the argument for a page per service.
Notice that four of those five are edits rather than projects. That is the real finding of most conversion work.
What this looks like in Helm
Two halves, and only one is a platform's job.
Counting it is in the same dashboard as your content: entry pages, sources, and form submissions, so you can see how many arrived and how many enquired without a second tool. Date ranges with a comparison overlay make this month against last month two clicks, which is the comparison that actually matters.
Changing it is the half that decides the number. Every item on the list above — the price, the button wording, the page text, the images that slow the page — is a field you edit and press Go Live on, with a live preview. Conversion work is iterative by nature: you change one thing, wait a month, compare. That only happens if changing one thing is a two-minute act rather than a support ticket.
The short version
Name the one thing you want a visitor to do, count how many do it, and divide. Ignore industry benchmarks and compare against your own last month. Then work the list: test the form, publish a price, fix the speed, ask for one thing per page. Most of that is editing rather than building, which is why the businesses that improve are the ones that can edit.