Your homepage was built for a different visitor. It's designed for someone who already knows your name, wants to browse services, maybe check your About page. A paid click is not that person. A paid click is a stranger who typed "emergency water heater repair" ninety seconds ago and is deciding whether to call you or the next result.
Sending that click to your homepage isn't fatal. Plenty of service businesses run profitable campaigns off a homepage. But you're paying full price for the click and getting a page that makes the visitor do work — scanning a nav bar, guessing whether you cover their town, hunting for a phone number. On a $9 click, that guesswork is expensive. Published 2026 benchmarks put average cost per click at $9.87 for attorneys and legal services, $8.33 for home and home improvement, and $8.00 for dentists and dental services (WordStream). At those prices, a page that converts two percentage points better is real money.
What Google says it's actually grading
Google's own description of landing page experience is short and worth reading literally. It's determined by "the usefulness and relevance of information provided on the page, ease of navigation for the user, the number of links on the page, and the expectations users have based on the clicked ad creative" (Google Help).
Four things. Useful information. Easy navigation. Fewer links. And no gap between what the ad promised and what the page delivers.
That last one is where homepages lose. If your ad says "24/7 Emergency Plumber — Springfield" and the page headline says "Family Owned Since 1987," you've broken the expectation. The visitor has to translate. Some of them won't.
Landing page experience is one of the three inputs to Quality Score, alongside expected CTR and ad relevance — we covered how that machinery works in the Quality Score article, so I won't repeat it here. The practical point: the page affects what you pay per click, not just what happens after the click.
The things the page needs
A headline that repeats the search. Not clever. Literal. "Emergency AC Repair in Mesa — Same-Day Appointments." The visitor should confirm they're in the right place without reading a second line.
The phone number, huge, at the top, tappable. Most local service traffic is mobile and most of your good leads want to talk to someone. The number should be a tel: link so a tap dials, and it should stay visible as they scroll on mobile.
A one-line answer to "do you cover me?" List the towns, neighborhoods, or counties. This single element kills a surprising amount of hesitation and it costs you nothing.
One next step, repeated. Call, or request an estimate. Pick the primary one, put it above the fold, and repeat it two or three times down the page. Don't offer a call, a form, a chat widget, a text option, and a newsletter and let the visitor choose.
A short form. Name, phone, service needed, ZIP. Every extra required field is another chance to abandon. You can qualify on the phone; you cannot qualify someone who never submitted.
Proof that's specific to the job. License and insurance numbers, years in business, manufacturer certifications, review count with the star rating, warranty terms. For a law firm, case types handled and jurisdictions. Named awards beat vague adjectives.
Real photos. Your trucks, your crew, your actual office. Stock photography of a smiling man in a hard hat reads as a lead-gen farm, because that's what lead-gen farms use.
What happens next. "Call and we'll give you a two-hour arrival window" or "Submit and we call back within 15 minutes during business hours." Reducing uncertainty about the next 20 minutes converts better than any headline tweak.
The things it doesn't need
A full navigation menu. Google explicitly counts links on the page and ease of navigation. Every nav item is an exit. Keep a clickable logo if you want people to be able to verify you're a real company — that's enough.
Your company history. Nobody with a flooded basement cares that you were founded in 1987. Move it below the fold or cut it.
Sliders and hero videos. They push your phone number below the fold and they slow the page down.
A wall of services. If the ad group was about drain cleaning, the page is about drain cleaning. Other services get their own pages.
Pop-ups on arrival. You paid for the visit; don't cover it with a modal. Google's Destination experience policy specifically targets destinations that are "unnecessarily difficult or frustrating to navigate" (Google Ads Policy).
Prices you can't honor. "$49 tune-up" that turns into $400 on the phone is the fastest way to earn one-star reviews and a policy problem.
Mobile isn't a version of the page. It is the page.
Build the mobile layout first and treat desktop as the adaptation.
For speed, use Google's own published thresholds as your bar: Largest Contentful Paint within 2.5 seconds, Interaction to Next Paint of 200 milliseconds or less, and Cumulative Layout Shift of 0.1 or less (web.dev). Those are measured on real visits at the 75th percentile, not on your laptop over office wifi. The usual culprits on service business sites are unoptimized hero images, chat widgets, four tracking scripts, and a page builder loading everything for a page that shows almost nothing.
Inside Google Ads, the Landing pages report has a "Mobile-friendly click rate" column — the share of mobile clicks going to a page Google's Mobile-Friendly Test considers mobile friendly. Note that this data can be up to 14 days old, so wait two weeks after a fix before judging it (Google Ads Help).
Policy tripwires that get pages disapproved
Under Destination requirements, your page has to return a working response for common browsers and devices, has to be crawlable by Google's AdsBot, and the domain in your display URL has to match where the click actually lands — redirects from the final URL to a different domain are a disapproval trigger. Google also states that violations of this policy don't cause immediate suspension without warning; you get at least 7 days' notice (Google Ads Policy).
Two more that hit local service advertisers specifically. If you use call assets, call-only ads, message assets, or location assets, the number must be verified, must be local or domestic for the country you target, must connect to your business, and must have active voicemail. Fax lines, vanity numbers, and virtual numbering services are listed as unacceptable. And if your landing page is thin — mostly built to route traffic somewhere else, or duplicated from another source without adding anything — that falls under Insufficient original content.
If you block bots aggressively at the firewall, or your dev put the campaign page behind a staging password, or your "track my clicks" tool bounces through a third domain, check these first. They look like mysterious low traffic. They're usually a policy flag.
One page per service, not one per keyword
Most local businesses need somewhere between three and eight landing pages, not thirty. Group by what the customer wants done: emergency repair, installation and replacement, maintenance plans. Each of those is a different urgency, a different price point, and a different headline. Keyword-level pages are a maintenance nightmare and rarely beat a well-built service page.
If you'd rather work from a build order than a checklist — page structure, form fields, tracking for calls and forms, and the campaign structure that feeds it — that's what the Launchpad system walks through step by step.
How to know whether it's working
Measure conversion rate per landing page, and count calls as conversions. If you only track form fills for a plumbing account, you're grading the page on maybe a third of its output.
For context, published benchmarks put the 2026 average Google Ads search conversion rate at 8.18% across industries (WordStream). Home services specifically averaged 7.33% in LocaliQ's 2025 data, with wide spread by trade — cleaning and maid services at 17.65% and roofing and gutters at 3.70% (LocaliQ). Treat those as rough orientation only. Your market, your keyword mix, and whether you count calls all move the number more than any benchmark does.
Then do the thing almost nobody does: pull up your own ad on your phone, click it, and try to become a customer. Time how long it takes to load. Tap the number and see if it rings the right phone. Submit the form and see where the lead lands and how fast someone responds. Half the landing page problems I'd expect you to find are not design problems. They're a form emailing an inbox nobody opens.