If you search for A/B testing advice, you'll find a lot of articles about eCommerce, telling you to test your product page, checkout flow, or add-to-cart button. This is useful if you're selling shoes online but not so useful if you're a plumber hoping someone clicks the "Call Now" button.

Service businesses have different websites with different goals. You're not trying to get someone to buy something on the spot - you're trying to get them to pick up the phone, fill out a contact form, or book an appointment. The visitor might be comparing three dentists in their suburb. They might be a homeowner whose pipe just burst and they need someone in the next hour, or they might be a small business owner looking for an accountant and reading through five websites before shortlisting two.

The testing principles are the same (try two versions, see which one gets more people to do the thing), but what you test and how you measure it looks different. This is that guide.

What "conversion" means for a service business

In eCommerce, a conversion is a sale. Simple. The money shows up in the dashboard.

For service businesses, conversions are messier. Someone contacts you, and then you have a conversation, and then maybe they become a client. The website's job is to start that conversation, not close the deal. That means your conversion actions are usually one of these:

  • Phone calls. The visitor taps your phone number on their mobile or picks up the phone and dials. This is the most common conversion for emergency services (plumbers, electricians, locksmiths) and for businesses where people want to talk to a person before committing (lawyers, financial advisors).
  • Contact form submissions. The visitor fills out a form with their name, number, and a brief description of what they need. Common for businesses where the enquiry isn't urgent (accountants, architects, marketing agencies).
  • Online appointment bookings. The visitor picks a date and time through a booking widget. Common for dentists, physiotherapists, hairdressers, personal trainers, and other appointment-based businesses.
  • Quote requests. A more detailed form where the visitor describes a project and asks for pricing. Common for tradespeople, renovation companies, and agencies.

When you set up an A/B test, you need to pick one of these as your goal. "More people visited my website" isn't a conversion. "More people tapped the Call button" is.

Tracking conversions on a service business website

Before you can test, you need to know what your current conversion rate is. This is where service businesses often get stuck, because phone calls and form submissions are harder to track than online purchases.

Form submissions are the easiest to track. Most form tools (Gravity Forms, Contact Form 7, Typeform, Jotform, even a simple mailto link) either redirect to a thank-you page or show a confirmation message. You can set either of those as a conversion goal. SplitPea can track a click on the submit button or a visit to the thank-you page.

Phone calls are trickier. If your phone number is a clickable link (which it should be on mobile), you can track taps on that link as a conversion. That's not a perfect measure because some people will tap and then not actually call, and some people will read the number off their screen and dial manually. But it's close enough to be useful for testing.

Booking widgets usually have a confirmation step. If the widget redirects to a confirmation page or triggers a success event, you can track that.

The key point: you don't need perfect tracking to run a useful test. If taps on your phone number go up by 30% with your new headline, that's a meaningful signal, even if not every tap turns into a completed call.

What to test (service business edition)

The high-impact tests are the same ones that matter for any website, but the specifics look different when your goal is a phone call instead of a purchase.

Your headline

This is still the best first test. But the winning headline formula for a service business is different from ecommerce.

Ecommerce headlines tend to be about the product. Service business headlines should be about the customer's problem or the outcome they want.

Weak headline: "Johnson & Sons Plumbing — Quality Service Since 1987"

This tells the visitor your name and how long you've been around. It doesn't tell them whether you can help with their specific problem.

Stronger headline: "Same-Day Plumbing Repair in the Northern Suburbs"

This tells the visitor three things: what you do, how fast you do it, and where you work. If they're in the Northern Suburbs with a leaking pipe, they know they're in the right place.

Even more specific: "Blocked Drain? We'll Be There Today."

This speaks directly to the problem the visitor probably has. It creates an immediate connection.

Some headline angles to test for service businesses:

  • Problem-focused vs. outcome-focused. "Got a Toothache?" vs. "Pain-Free Dental Care in Carlton."
  • Location-specific vs. generic. "Family Lawyer in Parramatta" vs. "Expert Family Law Services."
  • Speed or urgency vs. quality or trust. "24-Hour Emergency Electrician" vs. "Licensed Electricians — 15 Years Experience."
  • Question vs. statement. "Need a New Roof?" vs. "Melbourne's Most Trusted Roofers."

The right answer depends on your customers. That's why you test.

Your phone number placement

For service businesses where calls are the primary conversion, where your phone number sits on the page matters more than almost anything else.

Test these variations:

Phone number in the header vs. phone number in a sticky bar. A sticky bar stays visible as the visitor scrolls. If your number is only in the header, it disappears the moment someone scrolls down. A sticky call button on mobile keeps it one tap away at all times.

Phone number as text vs. phone number as a button. A styled "Call Now" button is more visible and more tappable than a plain text phone number.

Phone number visible immediately vs. behind a "Call Us" button. Some businesses hide the number behind a tap, usually to track clicks more accurately. Test whether showing the number directly gets more calls.

If you're a business where people call (tradies, emergency services, medical practices), this test can have a bigger impact than almost anything else on the page.

Your contact form

Most service business contact forms ask for too much information upfront. The visitor hasn't decided to hire you yet. They're just reaching out. Asking for their life story at this stage creates friction.

Test a shorter version: just name, phone number or email, and a one-line description of what they need. Three fields. Compare it against your current form.

The concern is always "but we need that information to give them a proper quote." You do. But you can ask for it after they've made contact. The form's job is to start the conversation, not replace it.

Another thing to test: the button text on your form. "Submit" is generic and tells the visitor nothing. Try "Get My Free Quote," "Request a Callback," or "Book My Appointment." The button should tell them what happens when they click it.

Trust signals

Service businesses live and die by trust. Your visitor is about to invite you into their home, trust you with their teeth, or hand you their legal problem. They're looking for reasons to feel confident, and they're scanning for red flags.

Test adding these near your main CTA or phone number:

Google review rating and count. "4.9 stars from 127 Google reviews" is more persuasive than any copy you can write. If you've got good reviews, show them. Test placement: near the headline, next to the phone number, or beside the contact form.

A specific trust line. "Licensed, insured, and police-checked" for a tradie. "Member of the Australian Dental Association" for a dentist. "Free initial consultation — no obligation" for a lawyer. Test whether adding this line near the CTA changes conversion rates.

Photos of real people. Test a stock photo versus a photo of your actual team or your actual work. Service businesses convert better when they feel real. A photo of a real plumber in a branded uniform standing next to a van beats a generic stock image of a handshake every time.

Your service area or location

Local businesses often bury their service area on a separate page or mention it vaguely. But visitors searching for "electrician near me" want to know immediately that you serve their area.

Test adding your suburbs or region prominently near the top of the page. "Serving Melbourne's Inner West: Footscray, Yarraville, Seddon, and Williamstown" tells someone instantly whether you're relevant to them. If you are, they're more likely to call. If you aren't, they leave sooner, which also saves you both time.

Tests that matter less for service businesses

Not every A/B testing best practice from the ecommerce world translates.

Checkout optimisation. You don't have a checkout. Skip any advice about cart abandonment, payment form layouts, or shipping option tests.

Product page layouts. You don't have product pages (unless you're a service business that also sells products, in which case, standard ecommerce advice applies to those pages).

Pricing page experiments. Some service businesses don't show pricing at all, which is its own kind of test (show pricing vs. "call for a quote"). If you do show pricing, the presentation matters, but it's less complex than ecommerce pricing with dozens of SKUs.

Button colour. This advice needs to die. The colour of your "Call Now" button isn't why people aren't calling. The words on it, the placement, and what surrounds it matter more. Especially with limited traffic, button colour tests are a waste of time.

How to think about traffic for service business sites

Service business websites often have less traffic than ecommerce sites, which makes people wonder if testing is even possible. The honest answer is in our guide on traffic requirements, but here's the service-business-specific version:

Most local service business websites get somewhere between 500 and 5,000 visitors a month. That's enough to test if you're strategic about it.

Test your highest-traffic page first. Usually the homepage. Don't start with your "About" page that gets 40 visits a month.

Test big, obvious changes. A completely different headline, not a slightly different one. Adding a sticky call button, not moving it three pixels to the left.

Be patient. A test on a page with 1,500 monthly visitors will take two to three weeks to reach confidence. That's fine. You're not in a rush. The test runs in the background while you do your actual job.

Use a goal that happens frequently enough. If you only get two form submissions a week, it'll take months to reach statistical significance on form submissions alone. Track phone number clicks instead, which happen more often, and use that as your test goal. It's a closer proxy to the action you care about than, say, page views.

A scenario to make it concrete

You're a family lawyer in Sydney. Your homepage gets about 2,000 visitors a month. Your current headline is "Smith & Associates — Experienced Family Lawyers." Your contact form has six fields. Your phone number is in the header but disappears when visitors scroll.

Here's a testing plan:

Test 1 (weeks 1–3): Headline. Test "Smith & Associates — Experienced Family Lawyers" against "Going Through a Separation? Get Clear Legal Advice." Track phone number clicks and form submissions as conversions. The second version speaks to the visitor's situation instead of your firm's credentials.

Test 2 (weeks 4–6): Sticky phone button. Add a sticky "Call Us" button on mobile that stays visible while scrolling. Track phone number taps. This is a simple change that can meaningfully increase call volume for any service where people want to talk to someone.

Test 3 (weeks 7–9): Form length. Cut your form from six fields to three (name, phone, brief description). Track form submissions. If submissions go up without a meaningful drop in lead quality, keep the shorter form.

Three tests, nine weeks, three clear answers about your website. Each one builds on what you learned from the last. That's how this works.

Getting started

If you're a service business owner and you've never run an A/B test, start with the headline. It's the fastest test to set up, it affects every visitor, and the results tend to be clear.

Here's a five-minute walkthrough of setting up your first test. If you're not sure what to test after the headline, here's a broader list of ideas.

Your website is probably the most visited shopfront your business has. More people see your homepage than walk past your office or see your van. Testing it isn't complicated, and the results apply whether you're fixing pipes, filing motions, or filling cavities.