The Google Ads numbers a home-services business needs

Which Google Ads metrics does a contractor actually need?

You need four numbers, and they all tie to booked jobs. Cost per booked job, which is what a real job costs you in ad spend. Conversions, meaning tracked calls, forms, and bookings. Conversion rate, the share of clicks that become leads. And impression share lost, which tells you how much room you have to grow. Everything else is detail.

Most metric guides hand you a list of 16 numbers to watch daily. That is noise. A contractor running a business does not need 16 numbers. You need the few that answer one question: is this making me money, and can I get more of it? The four above answer that.

The rest of this doc explains each one in plain terms. It shows you the booked-job math, and it tells you which popular metrics you can safely ignore. The goal is a dashboard you can read in five minutes and trust.

What’s a vanity metric versus a money metric?

A vanity metric looks good in a report but does not track money. Impressions, click-through rate, and raw clicks are vanity metrics. They go up when your ads get seen and clicked, which feels like progress. A money metric tracks dollars and jobs. Conversions, cost per booked job, and impression share lost are money metrics. They tell you what you earned and what you can still win.

Here is the split at a glance:

Vanity metric	Why it misleads	Money metric to watch instead
Impressions	Being seen is not being hired	Conversions (tracked calls, forms, bookings)
Click-through rate	High clicks can still be wrong clicks	Conversion rate (clicks that become leads)
Clicks	You pay for these, they do not pay you	Cost per booked job
Top of page rate	Top spot on junk searches wastes money	Impression share lost (room to grow)

What is a conversion in Google Ads (calls, forms, bookings)?

A conversion is a valuable action a searcher takes after clicking your ad. For a home services business that means a phone call, a form fill, or a booking. It is the bridge between a click, which costs you money, and a job, which makes you money. If your account is not counting conversions, none of the money metrics below can exist.

Not all conversions are equal, which is the whole reason to track them well. A tracked call that lasts long enough to be a real conversation is worth far more than a click on your phone number. The best accounts go one step further and feed back which leads booked, so the bidding aims at jobs. That setup lives in two docs: how to count calls and forms properly for the base, and offline conversion tracking for the booked-job loop.

How do you calculate cost per booked job (not just cost per lead)?

Cost per booked job is your cost per lead divided by your close rate. Say leads cost you $100 each and you book one in three. Each booked job costs about $300 in ad spend. That is the number to budget against, because a lead sitting in your inbox never paid a bill. Cost per lead alone hides how good, or bad, your leads really are.

Work a quick example. You run two campaigns and both report a $100 cost per lead. Your gutter and storm-damage campaign books 1 in 5, so $500 a job. Your water damage callout campaign books 1 in 2, so $200 a job. Same lead cost, wildly different truth. Cost per lead said they were equal. Cost per booked job showed the storm campaign is more than twice as expensive. This is why we bid to jobs, not leads. It is also why storm work carries its own target instead of the account average.

Two related terms come up a lot, and neither one replaces the number above. Cost per acquisition is the platform’s cost-per-conversion label. Customer acquisition cost is the fuller all-in cost of winning a customer. Both get their own page, so we will not redefine them here. To set a target, you first need your own numbers: job value, margin, and close rate. Feeding booked jobs back to Google is how the bidding starts using them.

What is impression share (and why “lost to budget” matters)?

Impression share is the share of searches where your ad showed, out of all the searches it could have. If your ad was eligible 100 times and showed 60, your impression share is 60%. The powerful part is the two reasons Google gives for the 40% you missed. Lost to budget means you ran out of money. Lost to rank means your ad or bid was not strong enough.

Those two “lost” numbers point at different fixes:

• Lost impression share to budget. Demand exists and you can afford more of it. Raising the budget captures real searches you are already missing.
• Lost impression share to rank. More budget will not help. Your ad relevance, bid, or Quality Score is holding you back. Fix those first.

Impression share is also your growth-ceiling gauge. Once you hold more than about 80% of the impressions for your searches, expect costs to climb. You are near the top of that market. The only ways up from there are to pay more per job or open a new service line. That is a strategy decision, not a bidding one.

What is Quality Score telling you about your money?

Quality Score is Google’s 1-to-10 rating of how relevant your keyword, ad, and landing page are to a search. It is not a vanity grade. A higher Quality Score lowers what you pay for the same position. So it is really a discount lever. Better relevance means cheaper clicks, which means a lower cost per booked job for the same work.

Read it as a diagnostic, not a target. A low score usually points at a mismatch. The keyword, the ad, and the page are not telling the same story. The fix is message match: the search term in the ad, and the ad’s promise on the landing page. Chasing the number itself is a waste. Fixing the relevance behind it is where the savings are. Full detail in the Quality Score guide.

How do you read a Google Ads report without drowning?

Add the columns that track money and hide the rest. For a contractor, show conversions, cost per booked job, conversion rate, cost, and impression share lost. Hide the display and video columns you do not use. Then read it on a routine, weekly, not daily. Daily reading tempts you to react to noise, and reacting to noise is how good accounts get wrecked.

One column pair is worth adding when a campaign underperforms: the “Top vs Other” segment. It splits your ads into those shown above the organic results and those shown below. If most of your money is going to spots below the organic results, raising your bid just buys expensive position. The better fix is relevance and the landing page. If you sit mostly above the results already, you have room to push for more volume.

If you get a monthly report from someone running your ads, hold it to the same test. If the first page leads on impressions, click-through rate and a green arrow, it was built to look good. It was not built to tell you what your work cost. Ask for one line instead: what you spent, how many leads it booked, and how many of those became jobs. Anyone running the account properly can answer that. It is also the first thing to ask of someone who reports on booked jobs instead of clicks.

Discipline beats activity here. Give new campaigns and keywords time to gather data before you judge them. Our rule: if you expected a booked job by 50 clicks on a keyword and there is still nothing by 80, that is a real signal to act. Before that, changing things just resets the learning and hides the truth. And a brand-new account is a data-gathering job. Expect a clear picture at about 3 months, not week one.

Which metrics are safe to ignore for a local service account?

Several popular metrics do not apply to a local contractor, so ignore them. View-through conversions, which credit ads that were seen but not clicked, mostly matter for display and video. Cost per thousand impressions, a brand-awareness metric, is not your game. And eCommerce numbers like return on ad spend and average order value assume online checkout, which you do not have.

Here is the honest “skip these” list:

• View-through conversions. Built for display and video branding, not lead-gen search.
• Cost per thousand impressions. An awareness metric. You want calls, not eyeballs.
• Return on ad spend and average order value. Store metrics. Your version is cost per booked job against job value.
• Top of page rate and absolute top of page rate as goals. Position is a means, not a result. Chasing it burns budget on searches that never book. And if you are hunting for the old “average position” number, stop. Google retired it back in 2019 and replaced it with these two rates.

Ignoring these is not laziness. It is focus. Every minute spent staring at a metric that cannot move your revenue is a minute not spent on the four that can.

Which attribution model tells the truth about your booked jobs?

An attribution model is the rule that decides which click gets credit for a conversion. A homeowner might click three of your ads before they call. The model decides how that one booked job is shared out. Google Ads now has two models: data-driven, which is the default on most conversion actions, and last click. Leave it on data-driven.

Google removed the other four. First click, linear, time decay and position-based are gone. The conversion actions that used them were moved over to data-driven.[3] So if a guide still walks you through six models, it is out of date. Your real choice is between two.

Here is why the default is the right one. A homeowner searches “burst pipe”, clicks your ad, then gets busy. That evening they search your business name, click again, and call. Last click hands all the credit to the brand search. Data-driven credits the first click too, which is the one that actually found you the customer. Bid to last click and you can end up cutting the very campaign that starts your jobs, then wonder why the pipeline dried up. Where you point the credit is where you point the spend, which is why it shows up in how you set your budget.

	Data-driven	Last click
Who gets credit	Shared across the clicks that helped	The final click only
Best for	Most accounts, where people compare before calling	Short paths, usually one click
Weakness	Harder to explain to someone else	Undervalues the click that found you
Contractor verdict	Leave it here	Only when you want blunt numbers

This is also why the same booked job can look different in two reports. Google Ads credits the ad click. Google Analytics looks at the whole path, so it can hand that same job to organic search or a direct visit instead. Both can be right at once. Pick one source of truth per action. Use your Google Ads conversions to drive the bidding. Do not import the same conversion from Analytics on top of that. You would count it twice and teach the bidding something false.

Changing the model moves credit between campaigns. Nothing changed in the real world. Only the credit moved. Google’s own guidance is to expect that shift, then check your targets. Smart Bidding learns from whatever the model credits.[4] So if you run Target CPA or Target ROAS, look at the target after any change. And do not flip models on a whim.

One catch matters more than the model choice. Attribution can only credit what it can see, and the job books on the phone or at the kitchen table, not on your website. So even a perfect model ends up sharing credit for form fills instead of jobs. Capturing the click ID with the lead and sending the booked outcome back to Google is what fixes that. Without it you are attributing leads. With it you are attributing jobs.

Stop reading · start fixingSee where your pipeline is leaking.Free 15-minute Leak Finder. We pull up your numbers, name your biggest leak, and hand you a 90-day plan. No sales script, no hard close.Find My Google Ads Leaks

FAQ

Is a high click-through rate good? Not on its own. A high click-through rate only helps if those clicks book jobs. Plenty of accounts post strong click rates and still lose money because the clicks are the wrong searches. Judge it against conversion rate and cost per booked job.

Which is more important, return on ad spend or cost per acquisition? For a local contractor, neither is the headline. Return on ad spend suits online stores, and cost per acquisition is a platform label for cost per conversion. Your headline number is cost per booked job, because that is the one you can hold a budget against.

Which metrics should I check daily? Almost none. Daily checking pushes you to react to noise. Read the money columns weekly, watch spend for anything wildly off, and leave the bidding alone between reviews. Ten minutes on a Monday is enough.

What is a good conversion rate? It depends on your trade and your landing page, so treat benchmarks loosely. A focused service page converts far better than a homepage. The real test is your own cost per booked job against your margins, not a blog average.

What is a conversion window? It is how long after a click a conversion still counts, and you may see it called a lookback window. For clicks the default is 30 days, and you can set it anywhere from 1 to 90 days.[5] Match it to how your customers actually buy. A blocked drain gets booked the same hour. A roof replacement can take weeks of thinking, and a short window will miss the click that started it.