Summary: Most local service businesses pull the plug on Google Ads too early, then assume “ads don’t work for my industry.” In reality, a new search campaign needs roughly 4 to 6 weeks and 30 to 50 conversions before you can tell signal from noise. This post explains what’s actually happening inside the account during that window, when it’s fair to call a campaign a failure, and the few warning signs that mean you should stop sooner.
You hand a credit card to Google, set a $50-a-day budget, and three weeks later you have nine clicks, zero calls, and a sinking feeling that you just lit $1,000 on fire. Every plumber, HVAC tech, and lawn care owner I’ve talked to has been there. The question is not whether you should panic. The question is whether three weeks is even enough time to know if it’s working in the first place. The honest answer is almost always no.
The Short Answer: Probably Longer Than You’d Like
For a typical local service business spending $1,500 to $4,000 a month on Google Ads, you need somewhere between 4 and 6 weeks of real data before you can draw any conclusion that isn’t a guess. The number that actually matters is not days on the calendar. It’s conversions. You need roughly 30 to 50 booked calls, form fills, or whatever you count as a real lead before the patterns in your account stop being random.
If you are getting one or two leads a week, that’s eight weeks minimum. If you’re spending too little to generate that volume at all, you’ll never get there. That’s not the ads failing. That’s a math problem with the budget.
What’s Actually Happening Inside the Account
The first two weeks of any new Google Ads campaign are called the learning period, and that name is literal. Google’s bidding algorithm is trying every combination of audience, time of day, device, and ad copy it has access to, watching which ones convert, and rebalancing your spend toward the winners. Until you’ve fed it about 30 conversions, it is largely guessing.
During this window, you’ll see weird things. One day your cost-per-click is $4.20, the next day it’s $11.40. Half your spend goes to a search term you never thought to target. Your ad shows on a Saturday night when your shop is closed. None of that means the campaign is broken. It means it’s still figuring out who your customer is. Touching it too much during this period, by pausing keywords, swapping ad copy, or changing the budget, resets the learning and makes it worse.
A Realistic Timeline for a Local Service Business
Here’s what a fair evaluation window actually looks like on a $3,000/month account for, say, a residential plumber.
Weeks 1 and 2: Learning period. Cost-per-click is jumpy, conversions trickle in. Don’t make structural changes. Add obvious negative keywords as you spot them in the search terms report, but otherwise leave it alone.
Weeks 3 and 4: Patterns emerge. You should now have 15 to 25 conversions and a clearer cost-per-lead. If a plumber is paying $80 to $140 per booked job at this stage, that’s normal. Under $60 is great. Over $200 is a flag, but not yet a kill order.
Weeks 5 and 6: Decision point. By now you have enough data to say whether the cost per lead is sustainable for your ticket size. A plumber whose average job is $450 can absorb a $150 cost per lead. A handyman whose average job is $180 cannot. This is the math that decides if you keep going.
Signs You Should Pull the Plug Early
There are a few situations where six weeks is too long to wait and you should cut it sooner. If your account has spent $1,500 and you have zero conversions tracked, something is broken. Usually it’s conversion tracking itself, not the ads. Check that calls and form fills are actually firing. If they are, and you still have zero leads, your landing page or phone-answering setup is the problem, not the campaign.
If your search terms report is full of completely irrelevant queries, like “free plumbing advice” or job listings, your keyword match types are too loose and the account is wasting money on traffic that will never convert. Fix the targeting before you spend another $500. And if you’re getting leads but every one is a price shopper from two counties away, your geo-targeting or audience setup needs work, not your patience.
Conclusion
Three weeks of disappointing Google Ads results is not a verdict. It’s a learning period. Give a new campaign 4 to 6 weeks and 30 to 50 conversions before you judge it, but watch closely for the specific failure modes that mean something is genuinely wrong. Reach out to Aragon Group if you want a second pair of eyes on an account you’re about to pause.
Frequently Asked Questions
How much should I budget per month to give Google Ads a fair shot?
For most local service businesses, $1,500 a month is the practical floor. Below that, you typically can’t generate enough conversion volume in 6 weeks to learn anything. $2,500 to $4,000 is where most campaigns can actually be evaluated honestly.
What if I see no leads at all in the first week? Is that normal?
A handful of leads in week one is normal. Zero leads in week one can be normal too, especially in a new account or a small geographic area. What matters is whether you’re getting clicks. If you’re spending money and getting clicks but no leads, look at your landing page and your conversion tracking before you blame the ads.
Should I keep tweaking the campaign during the learning period?
No, with one exception. You should add obvious negative keywords when you see clearly irrelevant search terms wasting money. Everything else, including bid changes, budget changes, and new ad copy, should wait until the learning period is over. Constant changes restart the learning and make the data unreadable.
What’s the difference between a campaign that’s failing and one that just needs more time?
A failing campaign has a clear, fixable problem: broken tracking, wrong geo-targeting, garbage search terms, or a landing page that doesn’t match the ad. A campaign that just needs more time looks unremarkable but is steadily accumulating conversions, with cost-per-lead trending in roughly the right direction.
Leave a Reply