You're ranking. Search your service and your town and there you are, near the top, and the visit count in your analytics has been flat and healthy for months. The phone rings about as often as it did before.
Here is the short answer to why. Getting found and getting chosen are two different jobs, and only the first one is search engine optimization. The second one happens in the few seconds between a stranger landing on your page and deciding whether you're worth the effort of a call. That is what "converting" means: the visit turning into an actual call, form, or booking. A page converts when a first-time visitor can tell, without scrolling, what you do, who you do it for, roughly where, and what to do next. It also needs nothing on it that gives them a reason to doubt you're real.
The rest of this guide is the checklist behind that sentence.
Being found and being chosen are separate jobs
I'm not arguing that rankings don't matter. They matter enormously, and if nobody can find you, nothing else on this page is going to help you. Being found is covered in the local SEO basics guide; this one starts where that one stops, at the moment someone is already on your page.
The part that surprises people is that you don't need more visitors to get more calls. Say 100 people land on your services page this month and two of them get in touch. Getting to four is a job for the page, working on the same 100 you already paid for, and it is entirely within your control in a way that climbing two more positions in Google is not.
What this guide is not. It is not a redesign brief. Every fix below is a wording change, a settings change, or a deletion, and none of them need a developer.
The four-second question
Picture walking into an unfamiliar hardware store holding a broken faucet fitting. You want one of two things: the right aisle, or a person who knows where the right aisle is. If neither is obvious within a few steps, you turn around and drive to the other store, and you don't read a single sale banner on your way out.
A first-time visitor to your website is that person. They're holding a problem and scanning for whether you're the aisle.
So the test for your homepage and your main service pages is blunt: can a stranger tell what you do, who you do it for, and roughly where you do it, without reading a paragraph? Within a glance, before any scrolling happens.
Headlines fail this when they were written to sound established rather than to be understood. "Advocating for families since 1998" sounds like something. "Family law in St. Catharines" is something. One of those tells a worried person on their phone at 11pm that they're in the right place. The other makes them keep looking for the aisle.
"Family law in St. Catharines" is something.
The fix is free. Rewrite the first line of the page in the plainest words you'd use if a neighbor asked what you do, then put the town in it. If the plain version sounds nothing like the rest of your site, that is a separate problem, and the brand voice guide is about exactly that.
Give people one obvious next step
The opposite failure looks like this. A page ends up with a contact form, a newsletter sign-up, a "download our free guide," a chat bubble, a request-a-quote button and a link to the blog, and every one of those was added by a reasonable person for a reasonable reason. Together they hand the visitor a decision they didn't want to make.
Pick the action that actually earns you money and make that action obvious. If you're a roofer in Welland, that's a call or a quote request. If you're a family-law practice, that's booking a short consultation. Everything else gets demoted or removed.
Then check the button itself. Is it visible without scrolling on a phone, and does it say what happens next? Does it appear again at the bottom of the page, so someone who read the whole thing doesn't have to scroll back up to act?
| Button says | What the visitor learns | Verdict |
|---|---|---|
| Learn more | Nothing about where the link goes or what it costs them | Weakest button on the internet |
| Contact us | That a form exists, somewhere | Better, still vague |
| Get a free quote | What they get and that it costs nothing | Good for trades |
| Book a 15-minute call | What they're agreeing to and how much of their day it takes | Good for consultations |
| Call (905) 555-0123 | Exactly what happens when they tap | Best on a phone, if the number is a real tappable link |
If you copy one line out of that table, copy the 15-minute call.
Trust, and the things that fake it
A first-time visitor is doing a quick and slightly unfair risk assessment: are these people real, and will I regret contacting them?
What answers that is unglamorous. Real photographs of your actual shop, your van, the people who will show up at the door, even when the lighting isn't great. Stock photography of a smiling headset woman does the reverse of what it's meant to do, because your visitor has already seen her on the last dentist's site. A phone number written as text, sitting where a thumb can reach it, and never baked into an image where it can't be tapped or copied. A real street address, or at least a clearly named service area. Reviews with names attached, ideally on Google or another platform the visitor can go and check.
And some signal about money. You don't have to publish a full price list, and in a lot of trades you genuinely can't. But a line like "kitchen projects start around $12,000" does two jobs at once: it saves you the calls from people who were never going to be able to hire you, and it earns credibility from the ones who can. Silence on price reads as expensive and evasive, in that order.
What does not build trust: countdown timers, "3 people are viewing this right now," and manufactured scarcity on a service that isn't scarce. Those patterns migrated here from discount e-commerce, and a visitor who recognizes one learns something true about you, just not the thing you were hoping to convey.
Every field you add costs you something
Think about arriving at a new doctor's office and being handed a clipboard with two pages of questions before anyone has told you whether they can even see you this month. That's what a long contact form is.
Someone who was thirty seconds ago mildly interested in a quote is now being asked for their address, their budget range, how they heard about you, and whether they'd like the newsletter. Some of them close the tab. You never see them in any report, because a visitor who leaves without acting doesn't leave a trace you'll notice.
Ask for the least you need to have a useful conversation. Name, a way to reach them, and a sentence about the problem is the whole list. Everything else you can ask once they're talking to you, when the asking is warranted.
Two smaller kinds of friction sit beside the form. On a phone, your number should be a real tel: link, so calling costs one thumb press instead of a memorize-and-retype exercise. And a slow page loses people before it says a word. Google's own threshold for the main content appearing (the measure it calls Largest Contentful Paint) is 2.5 seconds or less for "good" and over 4 seconds for "poor". Its free PageSpeed Insights tool reports where your page sits. The technical SEO guide covers what to do about a slow number.
Most of your visitors are using a thumb
Local searches happen on phones, and the person is doing something else at the same time: standing in a driveway with the engine running, or half-watching television with one hand in a bowl of chips.
So run this test yourself, today, on your own phone, not the monitor where you approved the design:
- Can you reach the call button with one hand, without shifting your grip?
- Does tapping the phone number actually dial, or just highlight the text?
- In the phone-number field of your form, do you get a number keypad or the full letter keyboard? (A field of type
telis what asks the phone for that keypad, per MDN's reference page; ask whoever built the form to use it.) - Is the headline readable before you scroll?
- Are the tap targets big enough? WCAG 2.2 sets the floor at 24 by 24 CSS pixels for a pointer target, and a link crammed against the next link fails a thumb even when it passes a mouse.
- Does the page still make sense with every image blocked? Your phone number should survive that.
Expect to find at least one item on that list you have never seen, because until now you only looked at the site on the big screen where you approved it.
Change one thing, then wait
The temptation, once you start noticing this stuff, is to fix everything at once with a full rebuild. Resist that for a month. If you change nine things and calls go up, you've learned nothing about which of the nine did it, and you'll carry all nine forward forever, including the ones that hurt.
Change one thing. Give it two or three weeks. Count calls and form submissions, since those are the numbers that actually pay you, and write the count down somewhere you'll find it again.
Start with the headline. It's the cheapest thing on the page to change and the first thing every single visitor reads. If you can't get it down to something your neighbor would understand without asking a follow-up question, that's your first week's work.
If you rewrite it and can't tell whether the new line passes, send me the old one and the new one. I'll tell you which I'd keep.
Checked against Google, W3C and MDN help pages on 2026-09-21: Largest Contentful Paint (LCP), About PageSpeed Insights, Understanding Success Criterion 2.5.8: Target Size (Minimum), input type="tel"