Playbook · landing-pages · conversion · audit
A nine-point landing-page audit for local-service Google Ads: message match, service and area clarity, call routing, proof, speed, mobile, tracking.
A click is a question. The landing page is where the answer has to be.
When a local service business pays for a click and the page fails, the money is lost twice: once for the click, and once for the job the click was about to become. The account can be structured perfectly, the keywords can be exact, and the bid strategy can be flawless. If the page does not answer, route, and reassure within seconds, the verdict will blame the advertising for a failure the advertising paid to deliver traffic to.
This is an audit you can run on your own phone in an afternoon. It has nine points, in the order a visitor experiences them.
The short answer
Open the page on a phone, as a stranger, with a real problem. Nine checks: does the page repeat the ad’s promise, say which services and areas you serve, show a working contact path, tell the truth about availability, prove the work, load fast, work on mobile, count what happens, and ask for one clear next action? Anything that fails is where clicks go to die.
| # | Audit point | The question it answers |
|---|---|---|
| 1 | Message match | Does the page repeat the promise the ad made? |
| 2 | Service and area clarity | Is it obvious what you do and where you do it? |
| 3 | The call path | Can a person in a hurry reach you in one tap? |
| 4 | The form path | Does the form match how your customers actually buy? |
| 5 | Availability and response | Does the page tell the truth about when you answer? |
| 6 | Proof | Does the page show work, reviews, and numbers with dates? |
| 7 | Speed | Does it load fast enough for someone standing in water? |
| 8 | Mobile first | Is the phone experience the design, not an afterthought? |
| 9 | One next action | When the visitor is convinced, what exactly do they do? |
Points 1 and 2 decide whether the visitor stays. Points 3, 4, and 5 decide whether they can act. Points 6 and 7 decide whether they trust you. Points 8 and 9 are the frame around all the others. Tracking runs underneath everything: a page that converts in ways you cannot count is a page you cannot defend in a budget conversation.
The ad promised something specific. “24 hour plumber”, “roof repair this week”, “oven cleaning near me”. The visitor arrives with that sentence in their head, and the page has roughly three seconds to repeat it back.
Message match is not a design preference. It is the continuation of the ad. A generic homepage headline like “Quality service since 2009” answers a question nobody asked. The page that says “Emergency drain unblocking in Leeds, answered day and night” matches the click, holds the visitor, and earns the next one.
The test is literal: read the ad out loud, then the headline out loud. If a stranger could not connect them in one breath, the match has failed.
Two questions every local-service visitor asks before anything else: do you do this work, and do you come here?
Vague answers cost real money. “Serving the wider area” is not an answer. A list of towns, or a plain map statement, is. “Full-service” is not a service list. Repair, replacement, maintenance, and emergency work behave differently, price differently, and convert differently, and the page should say which ones it is selling.
For urgent work, the phone number is the product. Everything else on the page is context for that number.
The audit checks four things. The number is visible without scrolling on a phone. It is tappable, once, and dials. It is a real business number that gets answered in the hours the ad runs. And if the call is not answered, something honest happens: a callback promise, a message path, or an after-hours note, rather than silence.
A call path that depends on the visitor copying a number into their keypad, hunting for it under a menu, or calling outside answered hours is a leak with a design budget.
Not every visitor calls. Project work, quotes, and scheduled services often start as forms, because the customer is comparing, not drowning.
A good form is short, states what happens after submission, and asks only for what the business needs to respond usefully. Name, contact, service, location, and one line of description is usually enough. Long forms with required fields the customer does not understand are a quiet way of asking people to leave.
The deeper question is whether a form is the right primary path at all. That is a decision about urgency and capacity, not taste, and it has its own framework in choosing the conversion path.
The page states when the business answers, and the business answers then. That is the whole point.
Availability dishonesty takes two common forms. The ad promises 24-hour service the operation cannot staff. Or the page shows a number that goes to voicemail during the hours the ad budget is spending. Both buy calls that become complaints, and both are invisible in a click report. The audit compares promised hours to actual answering, including what happens on a Friday night, before the budget scales anything.
A stranger is about to let your crew into their house. The page has to give them a reason to do it.
Proof for a local service business is concrete: photos of work you actually did, reviews with names and dates, years in trade stated plainly, credentials where they matter, and prices or price logic where honesty permits. Stock photography of handshakes is not proof. Neither is a wall of five-star reviews with no dates or source.
The strongest proof is specific. “Replaced a storm-damaged roof in Round Rock, June 2026, documented in the client’s Ledger review” beats “hundreds of happy customers”, because it can be checked.
Speed matters most at the moment of urgency, on the device of urgency: the phone.
A homeowner standing in water will not wait for a hero image. The audit does not chase a universal benchmark, because none applies honestly across markets and devices. It asks a narrower question: on a typical phone, on a typical connection, does the page appear and become usable fast enough that a person in a hurry stays?
Most local-service search that matters happens on a phone, often one-handed, often stressed, sometimes wet.
The mobile audit is unglamorous: text readable without zooming, the number reachable with a thumb, the form usable without fighting the keyboard, no horizontal scrolling, no popups covering the contact path, and no pinch-to-read tables. The desktop version can be lovely. If the phone version argues with the visitor, the audit fails on this point regardless.
A convinced visitor still needs to be told what to do.
Pages fail here by offering everything at once: call, form, chat, social links, newsletter, and six service menus, all at equal volume. The visitor who was ready chooses nothing, because choosing became work. The audit asks: what is the one action this page exists to produce? That action gets the prominence. Everything else is secondary, smaller, and lower.
Open the page as a stranger, on a phone
Not as the owner. Not on your office connection. Ideally with a real problem in mind and three seconds of patience.
Score the nine points honestly
Pass, fail, or not applicable. A point that is not applicable, like a form on an emergency page, is a decision to write down, not a gap to hide.
Fix in the order the visitor experiences them
Message match before proof, proof before speed. Fixes that reorder the visitor's first three seconds come first.
Change one thing, then watch the stage that moved
Calls, forms, and bookings are counted separately before and after. The conversion-tracking audit covers how.
Re-run the audit quarterly and after any site change
Pages drift. Redesigns, new widgets, and new services each quietly retest all nine points.
The nine-point landing-page audit
0 of 12First three seconds
The contact path
Trust and mechanics
An emergency plumbing page, audited and repaired, spends $1,200.00 in a defined month.
| Stage | Before the audit | After the audit |
|---|---|---|
| Clicks | 240 | 236 |
| Calls dialled | 41 | 68 |
| Calls answered | 26 | 52 |
| Booked jobs | 11 | 21 |
| Cost per booked job | $109.09 | $57.14 |
The click count barely moved, because the account was never the problem. The page was. Moving the number above the fold, matching the headline to the ad, and publishing real answering hours changed the stages the click report cannot see. That is the shape of a landing-page win: same clicks, more jobs.
This audit finds page-side failures. Two neighbours complete it. When clicks arrive and nothing happens at all, why clicks do not become calls works the diagnosis in order. When the page converts but the counting is doubtful, the conversion tracking audit is the next read. And when the account itself is the suspect, what to do when ads spend but the phone does not ring starts from the campaign side.
The Profit Ledger
The Ledger does not credit a page for clicks, visits, or engagement. It credits booked jobs and confirmed revenue. I count conservatively. Every untracked job is your upside, not my credit.
Run the nine-point checklist on your highest-spend landing page this week, from your own phone. If three or more points fail, fix the page before the budget scales. Then decide deliberately between the call and the form with the conversion-path framework, and make the numbers behind it defensible with the conversion tracking audit. If you want a second pair of eyes on the page and the account together, start a conversation or ask for the free audit.
One per genuinely different promise. If repair, replacement, and maintenance pull from the same ad with the same headline, one page can serve them. When the search intent, price, urgency, or sales process differs, separate pages keep the message match honest and the measurement readable.
It depends on the urgency of the demand and how the business can respond. Emergency work usually earns the number first; project and estimate work usually earns the form first. The full decision framework is in the conversion-path guide, and the honest answer on any single page is to test it.
It matters most on mobile, at the moment of urgency. A slow page loses the caller who needed help now, and that caller is often the most valuable one. No universal benchmark applies; what matters is whether your page loads fast enough for your customer on the phone they actually have.
The ad promises a specific thing. The page repeats that specific thing, in the same words, above the fold. A click is a question. The landing page answers it before it asks for anything.
Most of this audit is judgment, not code. You can check message match, service and area clarity, the call path, availability honesty, proof, and the next action from your own phone. Speed and tracking usually need help, and the audit says where that line sits.
The tracking audit asks whether the numbers are counted correctly. This audit asks whether the page deserves to convert at all. Both run before budget moves, because more spend on a leaking page buys more leaks.
Start here
Send what you have. A missing number is not a blocker. Working it out is part of the audit. No contracts, ever. I reply within 12 hours.
Send the numbers
Spend, leads, and what you suspect is going wrong. I reply within 12 hours.