Planning a website · guide 4 of 7
Conversion basics: small changes that turn visits into leads
Get more enquiries from the visitors you already have: clearer calls to action, shorter forms, real trust signals, a faster mobile page, and honest measurement in GA4.
28 min readUpdated 9 Nov 2026
- 11 questions to test yourself
- 2 calculators
- 11-item checklist
On this page12 sections
Before you buy more traffic, check the bucket
Picture two owners. One spends the next six months chasing more visitors. The other spends one afternoon making the contact page easier to use on a phone. Which one gets more enquiries in month one?
Often the second, and the reason is arithmetic. Traffic poured into a page that loses people fills a leaky bucket faster: you pay for every drop, and most of it still runs out of the bottom. Fixing the holes makes every visitor you already have, and every one you win later, worth more.
This guide finds the holes in order: what to count, the button, the form, the proof, the phone, and how to tell whether a fix worked when you only get a few leads a month. At the end you score your own page and leave with your first three fixes.
One more thing to keep in mind as you read. There is one mistake, saved for near the end, that quietly cancels every other fix here. Most owners who make it never notice. See if you can guess it before you get there.
Try the numbers
The leaky bucket, with your numbers
Move the sliders to your own figures. If you do not know them yet, the defaults are illustrative; the measuring section shows where to find yours. For the value of an enquiry, a rough guess is fine: what an average job is worth to you, times the share of enquiries you usually win.
Extra enquiries, same visitors
8 more enquiries
What they could be worth
about €375 a month
Visitors needed to match it without fixes
500 more visitors a month
Arithmetic on your own inputs, not a forecast. Nobody can promise what a change will do to your rate; the point is to see what a small difference in the rate is worth next to chasing more traffic.
With the defaults, going from 1.5% to 2% sounds tiny. It is 7 or 8 more enquiries a month, and matching it with traffic alone would take 500 extra visitors every month, for as long as you want the result.
Now look at your own third number and ask one question: how long did it take you to gain that many monthly visitors last time, and what did it cost? If the honest answer is “a year” or “I never have”, the page is the cheaper place to start. That is why the rest of this guide is about small, testable changes, not a redesign.
What counts as a conversion for your business
A conversion is the moment a visitor does the thing that can turn into money for you. For an online shop that is usually a purchase. For most service businesses it is an enquiry: a call, a form, a booking, a message on WhatsApp, or a visit to the shop after checking the directions.
Write your list before you change anything. Separate the main actions, the ones that start a sale, from the small steps on the way, such as viewing the price list or starting the form. The small steps are useful clues, but if you optimise for them you can end up with more clicks and the same number of customers.
Quick quiz
Your analytics show 40 taps on the phone number this month. How many phone leads did the website bring?
Pick one. Your first answer is the one that counts.
Calls to action: one obvious next step, labelled with what happens
A call to action is the button or link that asks the visitor to take the next step. The most common problem is not the colour. It is that the visitor cannot tell what will happen when they press it, or has to choose between five buttons that all look equally important.
Give each page one main action and make it the most visible thing after the headline. Others can stay, but quieter: a text link instead of a second big button. Then label it with the outcome, in the visitor's words. “Submit” describes what the computer does. “Get my quote” describes what the visitor gets.
Add one line beside or under the button that answers the question people have at that moment: how fast you reply, whether it costs anything, whether they are committing to something. Only promise what you actually do. If the page you are fixing is your homepage, the guide on the homepage that turns visitors into enquiries applies the same rule to the first screen a stranger sees.
Before and after
The same button, two ways
Vague
[Submit]
Contact us for more information.
Clear
[Send my request]
We reply within one working day with a price range. No obligation, no call unless you ask for one.
- The label says what the visitor gets, not what the form does.
- The line underneath answers the three silent worries: how long, how much, and what happens next.
- Every promise in it is one the business keeps. If you reply in two days, say two days: a missed promise costs more than a modest one.
Forms: every field is a question the visitor has to answer
Each field on a form is a small cost: more typing on a phone, one more chance to hesitate, one more reason to wonder why you need to know. Fewer fields usually means more people finish. The trade-off is that you may learn less up front, so the art is keeping only what you genuinely need to reply well.
A good test for each field: if it were missing, could you still answer the enquiry? If yes, cut it or make it optional and say so. If you need it only for some enquiries, ask for it in your reply instead.
Flip each card to see the usual verdict for a small-business enquiry form. Decide first whether you would keep the field.
Test yourself
Think of your answer, then flip the card.0 of 8 flipped
Trust signals: proof that removes a reason to doubt you
Before someone gives you their phone number, they want to know you are real, competent and fair. Saying “we are trusted and professional” does not answer that, because every business says it. Evidence does.
Test your instincts first: two of these four beliefs are wrong, and both are common on small-business sites.
Myth or fact?
Four beliefs about trust signals
1.Only showing five-star reviews makes you look best.
2.Showing a price range, even a wide one, can increase enquiries from the right people.
3.Stock photos of smiling people are as good as real photos.
4.Proof works harder next to the button than on a separate page.
Signals that usually work are specific and checkable. Put the strongest one right next to the main button, where the doubt happens:
- Reviews with first names and details, ideally linked to your Google Business Profile.
- Photos of your real team, van or premises instead of stock images.
- A street address or the area you cover, and opening hours.
- Prices or price ranges.
- The guarantee or warranty you actually give, and the registration or insurance you hold.
Spot the friction: one contact page, many small leaks
Small problems add up. None of them alone seems worth fixing, which is why they survive for years. Here is an illustrative page; read it the way a hurried customer on a phone would, and count the leaks before you look at the answer.
Make a guess
How many separate leaks did you count on that page?
Anything within ±1 leaks counts as right.
Here are the seven, with the section that fixes each one:
- A slow hero photo that pushes everything useful down (speed).
- Nine required fields, including first and last name split, company and full address (forms).
- A button that says “Submit” (calls to action).
- A phone number that cannot be tapped or copied (calls to action).
- No opening hours, no area covered, no reply time (trust and the line under the button).
- No proof of any kind near the form (trust signals).
- A cookie banner covering a third of the screen (speed and phones).
What would you do?
You have one hour this week
Now choose. The plumber can only make one change before the busy season. Most of the site's visitors arrive on phones, many with an urgent leak.
Which change would you make first?
Speed and phones: the leaks you cannot see on your desktop
Most owners check their site on a fast office computer. Many customers see it on a phone, on mobile data, often in a hurry. A page that feels instant on your desk can take several seconds on theirs, and every second is another chance for them to give up and go back to the results.
Guess before you read the number below.
Make a guess
In Google's research on mobile pages, as load time grew from 1 second to 3 seconds, by how much did the probability of a visitor bouncing (leaving) increase?
Anything within ±8% more likely counts as right.
You do not need to understand every metric to fix the usual culprits. On small-business sites they are almost always the same few things:
- Huge images: a 5 MB photo straight from the camera where a 200 KB one would look the same on a phone.
- Too many plugins, widgets and tracking scripts, each loading its own code. Remove the ones nobody uses.
- A slider or video at the top that has to load before anything useful appears.
- Pop-ups and banners that cover the button on a phone, or jump in after the visitor has started reading.
- Buttons and links too small or too close together to tap without zooming.
Measuring conversions in GA4, so you know what actually changed
If you do not count enquiries before you change something, you cannot know whether the change helped. Google Analytics 4 (GA4) can count them for you once you tell it which actions matter. In 2024 Google renamed GA4's “conversions” to “key events”, so that is the word you will see in the menus.
There are several ways to capture an enquiry, each with a weakness. Pick the most reliable one for each type of lead, and check it against reality now and then: your inbox, your phone log, your booking calendar.
Side by side
| Method | What it counts | Watch out for |
|---|---|---|
| Thank-you page after the form | Each visit to a page shown only after a successful send | People can reload or bookmark it; it must not be reachable any other way |
| GA4 form interactions (enhanced measurement) | Form starts and submits, switched on in your web stream settings | It does not detect every form correctly, so test it; a submit is not always a successful send |
| Phone number taps | Clicks on your tappable number, set up as an event | A tap is not a call; compare with the calls you received |
| Call tracking numbers | Real calls, from a number shown only on the website | A monthly cost, and the number differs from your listings; keep your main number consistent elsewhere |
| Asking every new customer | How people say they found you | Memory is unreliable, but it catches what analytics miss |
Step by step
Count form enquiries as a key event in GA4
Step 1 of 5
Make the form land on a thank-you page
Ask your developer or form plugin to send people to a page such as /thank-you after a successful send, and to keep that page out of your menus. It is the simplest signal to trust.
Did the change work? Why small numbers lie
Big websites run A/B tests: half the visitors see version A, half see version B, and software decides the winner. With a few dozen enquiries a month, that kind of test can take many months to tell you anything reliable. You can still learn; you just have to be careful what you conclude.
Try this one before reading on.
Quick quiz
A business averages 20 enquiries a month. After a new contact page goes live, the next month brings 25. What can it conclude?
Pick one. Your first answer is the one that counts.
How wide is “normal”? When enquiries arrive independently of each other, statistics gives a rough rule: most months land within about two square roots of your average either side of it. For an average of 20 that is roughly 11 to 29. Put in your own average to see your range.
Try the numbers
Your normal month-to-month swing
A quiet month that is still normal
about 11
A busy month that is still normal
about 29
A rough rule of thumb that assumes enquiries arrive independently (a Poisson pattern). Seasonal businesses, or months with holidays, adverts or press, swing more. A month inside this range after a change has not proved anything yet, in either direction.
A fair before-and-after test in five steps
So change one thing at a time, compare periods of equal length, and watch for things that move the numbers for other reasons: holidays, the weather, a competitor closing, your own adverts starting or stopping. Notice in the calculator that the swing shrinks relative to the average as numbers grow, which is why a longer test period tells you more.
Put it in order
Put the steps of a fair before-and-after test in order
Use the arrows to move each item, then check.
- Record a baseline: visitors and enquiries over several weeks
- Compare the rate, not just the count, and check for outside causes
- Keep the change, undo it, or test longer if the result is unclear
- Change one thing, and write down the date
- Wait the same number of weeks as the baseline
Mistakes that make conversion work backfire, and the one that cancels the rest
- Fixing what you like rather than what the data or testers show. Your favourite section may be the one nobody reads.
- Changing five things at once, then not knowing which one helped or hurt.
- Chasing small steps, like clicks on “learn more”, while real enquiries stay flat.
- Using pressure tactics: fake countdowns, “only 2 slots left” that is not true, pop-ups that block the page. They can lift clicks briefly and cost you trust, reviews and repeat customers.
Quick quiz
The start of this guide promised one mistake that cancels every other fix. Which is it?
Pick one. Your first answer is the one that counts.
Score yourself
Score your own contact page, honestly
Open your site on your phone and answer for the page where most enquiries start.
I know which actions count as an enquiry, and I can see how many happened last month
My phone number is a tappable link near the top of the page on a phone
There is one main button, labelled with what the visitor gets
Next to it, a line says how fast I reply and what happens next, and I keep that promise
The form asks only for what I need to reply
Real, specific proof (a named review, real photos, a guarantee) sits near the button
Opening hours, the area I cover and a price range are easy to find
I have sent myself a test enquiry on mobile data in the last three months, and it worked smoothly
0 of 8 answered. Answer them all to see your result.
Checklist
Your fix list, roughly in order of effort
Saved on this device only
Questions owners ask
Questions people ask
What is a good conversion rate for a small business website?
There is no single good number: it depends on the industry, the price of what you sell, how urgent the need is and where the visitors come from. A plumber's emergency page and a wedding photographer's portfolio will never look alike. The useful comparison is your own page against itself, before and after a change.
Should I add a chat widget?
Only if someone can answer it quickly during the hours you show it. An unanswered chat is worse than none. Many local customers are happy with a tappable phone number and a WhatsApp link that go to someone who replies.
Will these changes help my Google rankings?
Nobody can promise that. Some overlap, such as a fast page that works well on a phone, also helps Google's assessment of page experience. But the main point of conversion work is getting more value from the visitors you already have, whatever your position in the results.
Do I need special testing software?
Not to start. GA4, a written baseline, one change at a time and three people watching on their phones will take most small businesses a long way. Testing tools make sense once you have enough traffic and enquiries for their results to mean something.
What if my website platform will not let me make these changes?
Most common platforms let you change button labels, form fields and phone links yourself, or with a plugin or app. If yours does not, or every small change needs a paid developer, that is a real cost to weigh. Try the free fixes in this guide first, then read the next guide in this series, on choosing between WordPress, Shopify, Wix and a custom build: it covers what each one lets you control and what leaving costs.
You reached the end
Your summary
This guide has 11 quick questions
Try them to see how much you already know. Your first answer is the one that counts.
Planning a website: 0 of 7 read
- 1. What a website brief should contain, and the gaps that stall projects: not read yet
- 2. Planning a website redesign: what to keep, what to change: not read yet
- 3. The homepage that turns visitors into enquiries: not read yet
- 4. Conversion basics: small changes that turn visits into leads: this guide, not read yet
- 5. WordPress, Shopify, Wix or custom? Choose once, not twice: not read yet
- 6. What a website really costs: the build, the bills and the extras nobody quotes: not read yet
- 7. Website launch checklist: the checks before, during and after go-live: not read yet
0 of 57 guides read · Saved on this device only
New guides by email when they are published.
Unsubscribe in one click.
What to do next
- Decide today who answers website enquiries and how fast, and put that promise in one line under your main button.
- Send yourself an enquiry from your phone on mobile data, and fix the first snag you hit: often an untappable number or a long form.
- Set up a form enquiry as a key event in GA4, test it in Realtime, and note this month's baseline before your next change.
Not sure which build fits?
Answer five questions and get a recommended build, add-ons and matching demos.
Open the plannerNext in Planning a website
Planning a website
WordPress, Shopify, Wix or custom? Choose once, not twice
Google does not care which platform you pick. Your future self will, on the day you try to leave.
21 min readStart reading
Where we can help
See it on a real build
Every brand in the demos is fictional.


