A homeowner with a new EV types "level 2 ev charger installation" into Google. Two electricians' ads come back.
The first says "Licensed Electricians, Call 24/7" and opens a home page that lists every service the company offers, from ceiling fans to generators. The second says "Level 2 EV Charger Installation" and opens a page about exactly that job: the charger, the circuit it needs, the permit, how the price works, a button to book.
Google can put the second ad above the first even when the first company bids more. That surprises a lot of owners, because they assume the top spot goes to whoever pays the most. It doesn't. This chapter explains how Google actually orders the ads, what Quality Score is and isn't, and why the cheapest way up the page is an ad group and a page built for one job. The keyword numbers are national. The example websites come from our example market, Columbus, Ohio. Your competitors will be different, and the mechanics won't.
The Bid Gets You Into the Auction. Relevance and the Page Decide the Seat.
A homeowner hitting search on "panel upgrade near me" sets off a split-second contest among every electrician whose ad could appear. It decides who shows and in what order with a number it calls Ad Rank, worked out fresh for every single search. Google actually calculates it twice: once to decide whether your ad is eligible to show at all, and again to decide where it lands.
Google lists six things behind that number:
- Your bid.
- How good your ad and its page are for this particular search, scored on the spot, including whether the page delivers on the ad's headline.
- Thresholds, a floor an ad must reach before Google lets it appear anywhere on the page.
- How competitive the auction is.
- The searcher's situation: what they typed, where they stand, whether they're on a phone, what hour it is, and what else is competing on that results page.
- How much your extras are likely to help, meaning the call button, the address, the links to other service pages.
You control three of those six directly: the bid, the ad and page, and the assets. There's no public recipe for how the six combine. Any article that multiplies your bid by your Quality Score is working from a model Google never confirmed. What Google does say is that higher-quality ads typically pay less per click than lower-quality ones. And position matters to you because the ads near the top get seen and clicked far more than the ones further down.
Quality Score Grades Three Things, and Ad Relevance Moves Fastest
Inside your account, each keyword gets a Quality Score from 1 to 10. It's useful, but Google says plainly that the number is a diagnostic and plays no part in the auction itself. The auction uses its own real-time judgment of your ad, not the number on your keyword. Treat the score as a report card on recent searches, not as a lever.
Google builds that number from three grades, and each grade compares you with everyone else who appeared for the same searches:
- Expected clickthrough rate. Would people tap this ad if it showed for this keyword? Google answers that with your ad's position and extras filtered out, so a bigger bid or another sitelink won't raise it. A sharper headline will.
- Ad relevance. Does the ad talk about the job the homeowner typed, or about the company in general?
- Landing page experience. Once someone taps, does the page get them closer to booking, or send them hunting through menus?
Relevance is the fastest of the three to fix, because it depends almost entirely on which searches you put together in one ad group. Google's own help pages say it directly: for each ad group, "pick a narrow theme," and split ad groups whose keywords "can't be easily addressed by the same ad."
An EV Charger Ad Group Holds EV Charger Installs and Nothing Else
The searches around EV charging look similar, and they aren't. Keyword Planner splits them into different intents at very different prices. These are US monthly searches with the top of Google's estimated top-of-page bid range:
- Someone hiring an installer: "ev charger installation" (14,800 searches, top estimate about $34), "level 2 ev charger installation" (5,400, about $26), "ev charger installation near me" (4,400, about $41).
- Someone with a Tesla who wants Tesla's charger installed: "tesla wall connector installation" (12,100, about $18) and "tesla charger installation" (3,600, about $28).
- Someone shopping for the hardware: "electric car charger for home" (12,100, about $5) and "tesla charging at home" (27,100, about $6).
Advertisers already pay several times more for the install searches than the shopping ones, because install searches turn into jobs. The same split shows up around panels. "Electrical panel upgrade" (9,900, about $28) and "breaker box replacement" (5,400, about $24) are people hiring. "200 amp panels" (8,100, about $2) is mostly people pricing the part at the hardware store.
So build the account the way you'd label a panel: one circuit per job. An EV install ad group holds five or six install searches and nothing else. Its ad says "EV Charger Installation" in the headline, not "Licensed Electricians." The Tesla installer searches get their own group, with an ad written for Tesla owners. The panel upgrade and breaker box searches get theirs. The hardware shoppers stay out of the account, or wait for a separate, deliberate test once the install groups are paying for themselves.
When every keyword in a group asks the same question, one ad can answer all of them, and that's what ad relevance measures.
Here's what that looks like in the ad itself. A responsive search ad allows up to 15 headlines of 30 characters each, and Chapter 8 builds them properly from your reviews. For the EV install group, a few of them might read:
- EV Charger Installation
- Level 2 Chargers Installed
- We Check Your Panel First
- Permit and Inspection Handled
Every one of those is about the job the homeowner searched. None of them would make sense in the panel upgrade group, and that's the point. If a headline fits every ad group in your account, it isn't helping relevance in any of them. "Licensed Electricians, Call 24/7" is true, and it's a fine line for your company-name campaign. In the EV group it's a missed chance to repeat the words the homeowner just typed.
A Panel Upgrade Ad Should Open a Panel Upgrade Page, Not a Menu of Every Service
The second half of the match is the page. Google rates pages higher when they're organized and their wording echoes what the person searched for. A homeowner who typed "breaker box replacement" should see those words, or close cousins, near the top of the page. In plain terms: the page should finish the sentence the ad started.
For each ad group, that means one page that covers:
- The job, named the way the homeowner searched it. "Level 2 EV charger installation," not "electrical services."
- What's involved: the circuit, a panel check, the permit and inspection.
- How pricing works, even if the answer is a range or a flat diagnostic fee. Chapter 3 showed how often surprise fees cost electricians the job.
- The area you serve and how fast you can come.
- One clear way to call or book, counted the way Chapter 1 set up.
The good news from our example market: at least 8 of the 12 electricians we studied already have a dedicated EV charger installation page, and we confirmed each one exists. A page existing isn't the same as an ad using it, so check where your ads actually send people. It takes two minutes: look at each ad's final URL.
Multi-trade companies have one extra problem. On one example site, the EV charger page sits inside the electrical section of a heating, cooling and plumbing website. That can work, as long as the page a homeowner lands on reads like an electrician's page, not a furnace company's.
Five of the Seven EV Charger Pages We Tested Took Over Four Seconds to Show on a Phone, Twice
Speed is part of landing page experience, and Google gives you a free way to check it. PageSpeed Insights (pagespeed.web.dev) loads your page on a simulated phone and reports, among other things, Largest Contentful Paint: how long it takes for the main content to appear. Google's guidance puts a good result at 2.5 seconds or less, and anything over 4 seconds is poor.
We ran the test twice on seven of the example EV charger pages. Five of the seven took longer than 4 seconds on both runs, and the slowest took more than 18 seconds both times. The other two moved around a lot between runs: one took 2.1 seconds the first time and 6.2 the second.
That swing is a lesson in itself. A single lab test is a snapshot from Google's servers, not what every homeowner sees. Run it a few times, and on your service pages as well as your home page. If a page lands over 4 seconds again and again, the usual culprits are oversized photos, chat and review widgets that load before the content, and a pile of tracking scripts. Fix those and the page tends to get faster for everyone.
Speed is one ingredient, though, not the recipe. Shaving three seconds off a home page won't rescue an EV ad that opens it. Get the match right first, then make it quick.
Next: Your Competitors' Websites Already Hold Your Keyword List. Chapter 5 Pulls It Out and Sorts It.
Tight ad groups need the right raw material. Chapter 5 shows how to seed Keyword Planner with the websites already winning in your market, then sort what comes back into groups like the ones above, including the plumbing and HVAC searches you'll need to throw out. If you're new to the guide, start with Chapter 1 on tracking and Chapter 3 on what a click is worth.
Every chapter, from tracking to emergency calls, is on the guide's home page.
When we take over an electrical contractor's Google Ads management, the first build splits the account into one ad group and one page per job. Doing it yourself? Open your keyword list, find the ad group with the most unrelated searches in it, and split it today.
