Guide · landing-pages · conversion · diagnosis
Clicks arrive, then disappear. The four post-click failures to check in order: message mismatch, friction, mistrust, and broken routing.
The account did its job. The click arrived. Now the click has to survive the page.
This guide is the second half of a diagnosis that starts in what to do when ads spend but the phone does not ring. That page works the account side: whether the searches were right, whether the clicks were real, whether the counting is honest. This page starts where that one ends. The visitor has landed. Something between the landing and the phone is eating them.
There are only four things that eat a click after it arrives: the visitor was promised the wrong thing, the visitor could not act, the visitor did not believe you, or the visitor arrived at the wrong page entirely. In that order.
The short answer
Clicks that never contact you fail in one of four places: the promise did not continue from the ad to the page, the contact path was buried or broken, the proof was too thin to trust, or the ad sent the click to the wrong page. Check them in that order, from a phone, as a stranger, before touching the budget.
The account side delivered a visitor.
Does the page repeat the ad's promise within seconds?
Number or form reachable, working, honest about hours.
Proof, specifics, and a page that respects the visitor's urgency.
A call, a form, or a request. Everything before this is a chance to lose.
The ad said one specific thing. The page said something general.
“Emergency electrician tonight” landing on “Family owned and operated since 1998” is not a mismatch of style. It is a broken sentence. The visitor finishes it by pressing back. Message mismatch is the cheapest loss to create and the cheapest to fix, and it hides well, because everyone involved reads the page knowing the brand, the town, and the services, while the visitor knows only the sentence they typed.
The test never changes: read the ad, then the first screen of the page. If they are not visibly the same promise, this is loss one, and nothing else matters until it is fixed.
The visitor is convinced, or convincing enough, and the page makes acting harder than leaving.
The usual mechanics: the phone number below the fold, or inside a menu, or as an image that does not dial. A form with eleven fields and no statement of what happens after submission. An after-hours ad running against a phone nobody answers. A booking widget that assumes an account. Each one converts a ready customer into a statistic.
There is a second failure hiding inside this loss: the path exists but does not match the demand. An urgent visitor staring at a quote form, or a considered project buyer forced through a phone tree, both lose to their own impatience. Choosing between call, form, and booking request is a real decision with its own framework in choosing the conversion path.
The page worked, the path was open, and the visitor still left, because the page asked for trust without offering any.
For a local service business, thin proof is usually some version of the same few things: stock photos instead of real work, testimonials with no names or dates, credentials implied but not stated, prices avoided so completely that the visitor assumes the worst, and a design that looks like it was last considered the year the business stopped caring. None of these are fatal alone. Together they tell a stranger that the details are not your habit, and strangers notice habits, because your work happens inside their house.
Proof does not need to be loud. It needs to be checkable: a photo of a job with a month, a review with a name and a platform, a licence number where licensing applies. The landing-page audit covers the full set.
Everything about the page is fine, and the page is wrong.
It happens quietly. A new service ad inherits an old page. A city-specific ad points at the general site. A repair campaign sends traffic to a page selling replacement. The page itself passes every check, for a different promise than the one that bought the visitor. Routing failures are a campaign-structure problem wearing a landing-page costume, which is why the diagnosis always confirms which ad, which keyword theme, and which promise produced the click.
The fix is mapping, not redesign: each genuinely different promise gets a page that continues it, and the campaign’s final URLs get checked against that map like any other setting.
You do not need exotic analytics to order these losses. Three ratios do most of the work, and all of them need honest tracking first, which is its own audit.
| Ratio | What it tells you | Which loss it points to |
|---|---|---|
| Clicks to calls dialled | Whether visitors who landed can or bother to call | Losses one and two |
| Clicks to forms started | Whether the written path is being found at all | Loss two |
| Forms started to forms submitted | Where in the form the visitor gives up | Loss two, specifically form friction |
| Contact to booked job | Whether the page’s promise attracted the right work | Often loss one, misread as a sales problem |
Read the ratios together, not alone. Many calls dialled but few answered is an operations problem, not a page problem, and the honest diagnosis says so. Everything dialled and answered but nothing booked is a promise problem: the page sold something the business does not profitably deliver. The ratios narrow the loss. The audit from the phone confirms it.
A painting company’s ads generated steady clicks and almost no contact for a defined month.
| Stage | Count | First read |
|---|---|---|
| Clicks | 180 | Healthy |
| Calls dialled | 6 | The page is losing nearly everyone |
| Forms started | 4 | The written path is invisible too |
| Forms submitted | 1 | Of the few who tried, most stopped |
The account-side audit had already passed: search terms were repair and repaint queries, locations were right, tracking was verified with test calls. The phone review found loss two: the number appeared below two screens of gallery images, and the form asked for a budget figure most visitors could not answer. The fix was mechanical. Number above the fold, budget field optional, one photo row instead of a gallery. The following month, on similar spend, calls rose to 31 and submitted forms to 9.
The account was never broken. The page was never redesigned. The losses were found in order and closed in an afternoon.
Fix message match first, then the contact path, then proof, then routing. Each fix is cheap; each changes what the next one can measure. Stop and re-check the numbers after each change rather than rebuilding everything at once, because two simultaneous changes argue about credit and the lesson is lost.
When the page passes the full nine-point audit and the ratios still disappoint, the page is probably not the problem anymore. At that point the question moves back to the account, or to the offer itself, and that verdict deserves the same discipline this page demanded.
The Profit Ledger
Clicks and visits are activity. The Ledger counts contacts and booked work, and says where each one came from. I count conservatively. Every untracked job is your upside, not my credit.
Open your highest-spend ad, then the page it lands on, on a phone, and read them as one sentence. If the sentence breaks, start with the nine-point landing-page audit. If the page holds up, verify the counting with the conversion tracking audit before believing the ratios, and if you would rather have the page and the account read together, ask for the free audit and I will start with whichever side is leaking.
That guide diagnoses from the account side: search terms, keywords, eligibility, and tracking. This guide starts where the account side ends, at the click. The same symptom can come from either side, which is why both get checked before the budget moves.
Sometimes, and sometimes it just means the visitor got the phone number from the search results or the ad itself and never needed the page. Judge the page by its contact outcomes, calls, forms, and requests, not by bounce rate alone.
Message match and the contact path. When the headline does not repeat the ad's promise, or the number is buried, every other improvement is working with the door still closed.
There is no universal count. A small number of clicks with zero contact from a page that also fails the nine-point audit is already evidence. When the audit passes and the numbers still disappoint, more clicks make the verdict sharper before they make it different.
Usually not, when the ad promises something specific. The homepage serves everyone, so it answers no one in particular. A page that continues the ad's promise is the default for paid traffic, with the homepage kept for people looking for the business itself.
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.