Here's an uncomfortable exercise: open your Google Ads account, go to the Search Terms report, and add up what you spent last month on queries that had nothing to do with what you sell. For most small business accounts, that number lands somewhere between 30% and 60% of total spend. Not underperforming spend - pure waste.
Next step on this topic: how to get client budget.
I've seen this pattern across dozens of account audits. Reviewing Google Ads campaigns for home services businesses at Salty's Media, the same handful of money leaks show up in almost every account that “isn't working” - regardless of industry, budget size, or who set it up. The good news: they're fixable in an afternoon, and none of them require an agency retainer to correct.
This isn't because small business owners are bad at marketing. It's because Google's default settings are optimised for Google's revenue, not yours. Broad match by default, auto-applied recommendations, Display Network opt-ins buried in campaign setup - the platform quietly steers new advertisers toward maximum spend, not maximum return.
And the cost compounds. Wasted spend doesn't just burn budget - it feeds bad conversion data back into Google's bidding algorithm, which then optimises toward more of the wrong clicks. A leaky account gets leakier over time.
Leak #1: Broad Match Keywords Spending Blind
If you set up your campaign using Google's guided flow, there's a strong chance every keyword in your account is broad match - because that's the default. Broad match tells Google, “show my ad for anything you think is related.” And Google's definition of “related” is generous when your money is on the line.
In one audit of a roofing company's account, the keyword “roof repair” was triggering ads for searches like free roof repair grants, roof repair jobs near me, and how to repair a roof yourself. Three different searchers - a grant hunter, a job seeker, and a DIYer - and not one of them will ever become a customer. That single match-type problem accounted for roughly a third of the account's monthly spend.
How to spot it in five minutes: Go to Campaigns → Insights and reports → Search terms. Sort by cost. Read the top 50 actual searches that triggered your ads. If you wince more than a few times, you have a broad match problem.
The fix: Restructure your core keywords to phrase match (“roof repair”) or exact match ([roof repair near me]). Phrase match keeps you visible for relevant variations without handing Google a blank cheque. There is a legitimate use for broad match - paired with Smart Bidding and a large volume of clean conversion data, it can find converting queries you'd never think of. But that's a strategy for mature accounts, not a starting point. Google's own match type documentation explains the mechanics; just remember the defaults are not your friend.
Leak #2: No Negative Keyword List (Or a Dead One)
Negative keywords tell Google what you don't want to show up for. Most small business accounts I open have either no negative list at all, or one containing four words that somebody added in 2023 and never touched again.
For a service business, the same waste-words appear constantly. Here's the starter list I add to almost every account on day one:
- free
- cheap
- DIY / “how to”
- jobs / careers / salary / hiring
- course / training / certification
- reviews / complaints (unless you specifically want comparison shoppers)
- grants / government assistance
- wholesale / suppliers / parts
That list alone typically stops 10–20% of wasted spend within the first week.
The fix that actually sticks: Negatives aren't a one-time task. Put a ten-minute weekly recurring event in your calendar: open the Search Terms report, scan for anything irrelevant, add it as a negative. That's the whole routine. Accounts maintained this way get cheaper every month, because the click quality keeps improving.
Leak #3: Auto-Applied Recommendations Left Switched On
Somewhere in your account settings, there's a good chance Google has permission to make changes on its own - adding broad match versions of your keywords, expanding your targeting, raising budgets. It's called auto-apply recommendations, and for most small accounts it's a slow leak disguised as helpful automation.
Google's recommendations aren't malicious, but they're built around one assumption: that spending more will be good for you. Some suggestions are genuinely useful (fixing disapproved ads, adding missing assets). The expansion-flavoured ones - “add broad match keywords,” “remove redundant keywords,” “expand your reach with Display” - routinely undo the exact tightening work from Leaks #1 and #2.
How to check: Go to Recommendations → the three-dot menu → Auto-apply. Review what's ticked.
The fix: Untick everything under “Grow your business” categories. Keep, at most, the maintenance items like fixing broken ad destinations. Then treat the Recommendations tab as suggestions from a salesperson: occasionally right, never neutral.
Leak #4: Sending Every Click to Your Homepage
You can fix all your targeting and still lose the money one step later - on the landing page. The classic version: an ad promising “Emergency Boiler Repair - Same Day Service” that clicks through to a generic homepage where the visitor has to hunt for the right service, scroll past the company history, and find a contact form with nine fields.
Every mismatch between what the ad promised and what the page delivers is paid traffic evaporating. It also drags down your Quality Score, which means you pay more per click for the privilege.
The fix - message match: Every ad group should point to a page that continues the exact conversation the ad started. For a service business, the minimum viable landing page has four things: a headline that mirrors the ad's promise, a click-to-call button visible without scrolling (most local service searches happen on phones), one short form as the backup option, and one or two trust signals - review count, accreditation, years in business. If you want to go deeper on the fundamentals, these are the 5 key things all landing pages should have. That's it. You don't need beautiful; you need matched.
Leak #5: Location and Schedule Settings Nobody Checked
Two settings quietly drain local business budgets more than almost anything else.
The “presence or interest” trap. By default, Google can show your ads to people who are merely interested in your area - not physically in it. That's how a plumber in Manchester ends up paying for clicks from someone in another country researching Manchester. In location options, switch targeting to “Presence: people in or regularly in your targeted locations.” For most local service businesses, this single toggle recovers meaningful budget instantly.
Ads running when nobody answers. If your ads run 24/7 but your phone is answered 8am–6pm, you're paying premium rates for calls that go to voicemail - and a missed call from a paid click is the most expensive kind of missed call. Check when your ads actually spend (Reports → time of day), then either schedule ads around your answering capacity or put a proper after-hours system in place. Either is fine; paying for calls into the void is not.
Want AI doing the heavy lifting in your marketing?
I build the systems that handle the boring 80 percent, so you get your week back. Done properly, with the human kept in.
Leak #6: No Call Tracking Means You're Optimising Blind
Here's a pattern that fools a lot of service businesses into killing campaigns that were actually working - or scaling ones that weren't: they measure only form submissions, while most of their real leads arrive by phone.
For home services especially, phone leads routinely outnumber form fills two to one. If your account only counts forms as conversions, two bad things happen. First, you underestimate what's working and make cuts in the wrong places. Second - and this is the part people miss - Smart Bidding only sees half your results, so the algorithm optimises toward form-fillers and away from callers, who are often your best, most urgent customers.
The fix, cheapest first: At minimum, enable Google's call reporting on your call assets and set “calls from ads” as a conversion (with a sensible minimum duration, say 60 seconds, to filter pocket-dials). The step up is a proper call tracking tool that swaps numbers dynamically and ties each call back to the keyword that produced it. Even the basic version transforms your data - and everything in Leak #7 depends on this being right.
Leak #7: Smart Bidding Sitting on Bad Data
Smart Bidding strategies like Maximise Conversions or Target CPA are genuinely powerful - they're also amplifiers. They take whatever conversion data you feed them and pursue more of it. Feed them clean data, and they compound your wins. Feed them junk, and they scale the junk.
I've audited accounts where the “conversion” being optimised for was a click on the contact page. Not a form submission - a page view. Google's algorithm was dutifully finding more people who liked looking at contact pages. The account had hundreds of “conversions” and almost no actual customers.
The conversion hygiene checklist before you trust any automated bidding:
- Every conversion action tracks a real business outcome - a form submitted, a call over 60 seconds, a booking made. Not page views, not button clicks.
- Phone calls are being counted (see Leak #6).
- There's only one “primary” conversion goal per campaign purpose - mixed goals confuse the algorithm.
- You have at least 30–50 real conversions in the last 30 days before using Target CPA. Below that, stick with Maximise Conversions and a firm budget cap.
Fix the data first. Then automation becomes your leverage instead of your leak.
How Much Should a Small Business Spend on Google Ads?
The honest answer: enough to buy meaningful data, in a market you can afford. In practical terms, work backwards from click costs. If clicks in your niche cost £3–5 (typical for many local services), a £500/month budget buys roughly 100–150 clicks - enough to generate leads and learn something. If you're in a market where clicks cost £20+ (legal services, some trades in big cities), £500 buys 25 clicks, which is barely a signal. In expensive niches, either concentrate your budget on your single highest-value service and tightest geography, or build your pipeline through SEO and referrals until you can fund ads properly.
The bigger principle: budget size matters far less than budget efficiency. A leak-free £750/month account will beat a leaky £2,000/month account in nearly every audit I've done. Fix the seven leaks above before you add a single pound of spend - otherwise you're just scaling the waste. It is also worth deciding up front whether Google Ads deserves priority in your PPC campaign, rather than splitting a small budget across every channel at once.
The 15-Minute Monthly Audit Checklist
Once the fixes are in, this monthly routine keeps the account tight:
- Search Terms report - scan top 50 by cost, add negatives (5 min)
- Recommendations tab - review manually, apply nothing automatically (2 min)
- Location report - confirm spend is inside your service area (2 min)
- Conversion check - do reported conversions match real enquiries in your inbox/phone log? (3 min)
- Landing page click-through - click your own ad path on a mobile phone; make sure the promise still matches the page (3 min)
Fifteen minutes a month. That's the entire difference between accounts that get cheaper over time and accounts that quietly bleed.
Fix the Leaks Before You Add to Your Google Ads Budget
When a campaign underperforms, the instinct is to conclude “Google Ads doesn't work for my business” - or its equally expensive cousin, “we just need to spend more.” In most audits I've done, neither is true. The platform was working exactly as configured; the configuration was just built on defaults designed to maximise Google's revenue, not yours.
Run the five-minute checks above this week. Find your two or three biggest leaks, fix them, and give the account thirty days of clean data before you judge it. Once the waste is gone, the next lever is the writing itself, and these ad-copy strategies that increase PPC ROI are a sensible place to start. More often than not, the budget you need was already there - it was just going to grant-hunters, job-seekers, and people who like looking at contact pages.
Frequently Asked Questions
Why are my Google Ads not getting conversions?
The most common causes, in order: ads showing for irrelevant searches (match type and negative keyword problems), landing pages that don't match the ad's promise, phone leads not being tracked so conversions are happening invisibly, and conversion tracking measuring the wrong actions entirely. Audit in that order - the fix is usually in the first two.
What is a good click-through rate for Google Ads?
For search campaigns, 4–6% is a reasonable benchmark for local service businesses, and well-targeted branded or emergency-intent campaigns can run into double digits. But CTR is a diagnostic, not a goal - a high CTR on the wrong searches just means you're wasting money faster. Judge the account on cost per real enquiry.
Should small businesses use Performance Max?
Cautiously, and not first. Performance Max can work once you have solid conversion tracking and clean data for it to learn from, but it's a black box that will happily spend across channels you can't fully control. Master a tightly-run Search campaign first; treat PMax as an expansion layer, not a foundation.
How long does it take for Google Ads to work?
You'll see clicks on day one, but judge nothing in the first two weeks - bidding algorithms need a learning period, and every fix you make partially resets it. A fair evaluation window is 30 days of stable settings and clean tracking. What you should see immediately after fixing the leaks above is the waste disappearing from your Search Terms report.
Related reading: The Best AI Productivity Tools for Entrepreneurs in 2026 (What I Use, Daily) and The Best AI Tools for Small Business Owners (What I Pay For).
If you want the full breakdown, here is everything I know about small business.
Related: ai phone answering for small business.
