Most business websites lose money in the same place: an ad or a search result sends someone who is ready to act, and the page they land on asks them to browse. A high-converting landing page fixes that by doing one thing — turning an interested stranger into an enquiry — and doing nothing else. This is the anatomy I build for clients, the copy that fills it, and the tests worth running once it is live.
It goes deeper on the page itself than the lead generation funnel guide, which covers the whole path from offer to nurture. If the funnel is the plumbing, this is the tap.
The rule that underpins everything
One page, one audience, one offer, one action. The moment a page tries to serve both homeowners and landlords, or offers a quote and a newsletter and a brochure, conversion falls. If you have three services worth advertising, build three pages. It is less work than it sounds — the structure below is reusable.
The anatomy, section by section
-
Above the fold: headline, sub-headline, action.
On a phone, the visitor sees roughly one screen. It must contain: what you do and for whom, the outcome or offer, and either the form or a button that jumps to it. Nothing else. A hero image is optional; a hero video that autoplays and pushes the form below the fold is a conversion tax.
-
The trust strip.
Immediately under the fold: review score and count, accreditations, years in business, a number (“1,400 boilers serviced in Greater Manchester”), recognisable client or brand logos. Three to five items, small, factual. This is the cheapest lift on most pages.
-
What you get.
Three to five bullets or icon blocks describing the outcome, not the features. “A written quote within 24 hours, fixed price, no obligation” beats “professional service and competitive rates”.
-
How it works.
Three steps with times attached: you submit → we do X by Y → you get Z. This removes the fear of an unknown sales process, which is the real reason many people do not fill in forms.
-
Proof in detail.
Two or three testimonials with names, places and photos where possible; a before/after; a short case note with a number in it. Generic five-star quotes without attribution do very little.
-
Objection handling.
The four things people privately worry about — price, disruption, timescale, being sold to. Answer them in plain sentences, with a price range if you possibly can. Pages that name prices convert fewer but far better leads.
-
The form, again.
Repeat the form or a button after the proof. A returning scroller should never have to scroll back up.
-
A minimal footer.
Phone, email, company details, privacy and terms links. No menu, no blog, no social icons pulling people away.
The five copy blocks that carry the page
| Block | Formula | Example |
|---|---|---|
| Headline | Outcome + who + constraint | “Fixed-price boiler repairs in Salford, usually same week” |
| Sub-headline | What happens and what it costs | “Send three photos, get a fixed quote within two hours. No call-out fee for quotes.” |
| Button | First person, specific | “Get my fixed quote” (not “Submit”) |
| Micro-copy under the form | Remove the last doubt | “No obligation. We never share your details. Most quotes sent within two hours.” |
| Thank-you page | Next commitment | “Your quote is on its way — pick a time for the survey while you wait.” |
Write the button and the micro-copy before the rest. They force clarity about what you are actually asking for.
Form design, where most pages leak
- Ask for what you need to reply, plus two qualifiers. Name, phone or email, and two dropdowns (what do you need / when). Five or six fields is fine when the offer is worth it; nine is not.
- Dropdowns and buttons over typing, especially on mobile.
- Multi-step for longer forms: qualifiers first, contact details last. Completion rates rise because the first click is easy.
- Correct input types so phones show the right keyboard, and inline validation that explains errors next to the field.
- No CAPTCHA unless spam is genuinely a problem — use a honeypot and server-side filtering. CAPTCHAs punish real customers on bad connections.
- The form posts to the CRM, not just an inbox, so the instant follow-up starts on its own (how to wire that up).
- A phone number as the alternative, tracked, for people who would rather call.
Speed and mobile are conversion features
Paid traffic is overwhelmingly mobile, and a page that takes four seconds loses a large share of clicks you already paid for. Keep the hero image small and modern-format, load fonts locally, avoid sliders and heavy page-builder sections above the fold, and test on a real phone on mobile data rather than office wifi. The speed checklist and Elementor fixes apply doubly here: on a landing page, every 100ms is money.
Message match
The page must repeat the promise that got the click. If the ad says “boiler service £85”, the headline says “boiler service, £85” — same words, same number. Mismatched pages produce the pattern every ad account has seen: decent click-through, terrible conversion, and a manager blaming the audience. For each ad group or campaign, one matched page (the Google Ads setup and local Facebook ads guides both hinge on this).
What to test, in order
- The offer. Free quote vs fixed price vs free audit vs discount. Multiples, not percentages.
- The headline. Outcome-led vs problem-led vs price-led.
- The form. Fields, order, single vs multi-step.
- Proof placement. Reviews above the fold vs below.
- The page length. Short vs long for the same offer.
Test discipline: one change at a time, run until each variant has a few hundred visitors and a couple of dozen conversions, and judge on booked jobs rather than form fills — a variant that doubles leads and halves quality is a loss. Below roughly a thousand visitors a month, skip A/B testing and simply improve the page using this list; the maths will not support a split test.
Benchmarks to judge against
| Traffic and offer | Visitor → lead |
|---|---|
| Search ads, matched page, quote/audit offer | 8–20% |
| Social ads, quote offer | 3–10% |
| Any traffic, free guide or checklist | 20–40% (colder leads) |
| Home page used as a landing page | 1–3% (the reason this article exists) |
| Lead → booked appointment with instant follow-up | 20–40% |
The mistakes that quietly halve results
- The full site menu at the top.
- Carousel heroes that move before anyone reads them.
- “Contact us” as the offer.
- Stock photos of models in hard hats.
- A form that emails a mailbox nobody watches on Saturdays.
- No thank-you page, so nothing is tracked and no second commitment is asked for.
- Redesigning weekly instead of running one version for a month.
Further reading: Core Web Vitals (web.dev) · PageSpeed Insights.
FAQ
What is a good landing page conversion rate?
For service businesses, 5–15% of visitors becoming leads is normal for a quote or audit offer and 20–40% for a free guide. Ad traffic to a matched page sits at the higher end; cold traffic to a general page at the lower. Measure your own for a month before deciding a page is broken.
Should a landing page have navigation?
No. Remove the menu and any link that is not the call to action — a landing page exists to complete one action, and every other link is an exit. Keep a logo (not linked, or linked to the same page), the phone number, and the legal links in a small footer.
How long should a landing page be?
As long as it takes to answer the questions that stop someone acting, and no longer. Simple, low-risk offers (a callback, a quote) convert well on short pages; expensive or unfamiliar services need proof, process and objection handling, which means a longer page. Length is a result of the offer, not a style choice.
What should I test first on a landing page?
The offer, then the headline, then the form, then the page’s proof. Changing the offer moves conversion rates by multiples; button colour moves them by rounding error. Only test one thing at a time, and only when you have enough traffic for the result to mean something.
Want the page, the form-to-CRM wiring and the follow-up built as one unit, so the traffic you are already paying for converts? See the funnel packages — each one ships with tracking, so you can see which section to improve next.
