I went through your Google Ads account. Here is where the $2,738 went.
You spent $2,738.78 on Google Ads over about nine months, across five search campaigns and your Local Services Ads. I pulled the account and read it line by line. There are two problems, and they are separate. The tracking that was supposed to tell you whether any of this worked was set up to count the wrong things. And the ads themselves were aimed at searches that were never going to become a customer. Below I walk through both, and I explain how each one happened, because none of it is obvious and most of it is not your fault.
First: the tracking
The numbers in your account are not real.
A "conversion" in Google Ads is supposed to mean a real result: someone called you, or filled out your form. That number is the whole point of running ads, because it is how you tell what is working. In your account, the conversion number has been set up to count things that are not results at all. Here is what I mean, with three examples straight from the account.
Defect 1
A plain visit to your website is counted as a "conversion"
One of your active conversions is literally called "Website visits," and it is marked as a primary goal. So every time someone lands on your page and does nothing else, the account records a conversion. Nobody called. Nobody booked. But the report goes up. This is why the numbers can look busy while the phone stays quiet.
MEASURED · "Local actions – Website visits" set as a primary conversionDefect 2
There is a "purchase" conversion built for online stores
Sitting in your account is a conversion named "kirklandrooter.com (web) purchase." That is the kind of tracking an online shop uses when someone checks out a shopping cart. You are a plumber. No one checks out a cart on your site. It does nothing except sit there, and it tells me the setup was copied in from a template, not built for your business.
MEASURED · a GA4 "purchase" conversion action is presentDefect 3
A search from Oklahoma got counted as a lead
One of your recorded conversions came from the search "nickerson plumbing ada ok" — someone looking for a specific plumber in Ada, Oklahoma, roughly 2,000 miles away. Here is how that happens: your ads run on the loosest matching setting Google offers (more on that below), so Google served your ad because it saw the word "plumbing" and ignored the rest. The person clicked, and because "any click that then visits the site" was set up as a conversion, the account logged it as a lead. If a search from Oklahoma counts as a Kirkland lead, then the conversion number cannot be trusted at face value for anything.
MEASURED · "nickerson plumbing ada ok" recorded as a conversionSo the honest summary is not "you have no tracking." It is worse than that: you have tracking, you have been paying for the traffic it measures, and it has been quietly reporting numbers that do not mean what they say.
Now the ads
Two settings sent most of your budget to the wrong searches.
Almost everything below comes down to two settings whoever built the account chose, and then never corrected. On their own they sound harmless. Together they are the most common way a small ad budget gets burned.
Setting 1
Every keyword is on "broad match"
A keyword can be matched three ways: exact (show my ad only for this phrase), phrase (show it for close variations), or broad (show it for anything Google decides is related). Broad is the widest net there is. All 133 of your keywords are on broad match, and not one is on phrase or exact. So when you added the keyword "drain cleaning," you did not buy the search "drain cleaning." You told Google to show your ad for anything it thinks is in the neighborhood of drain cleaning — a competitor's name, a different service, a different town.
MEASURED · 133 of 133 keywords on broad match · 0 on phrase or exactSetting 2
There is not one negative keyword in the account
A negative keyword is a search you tell Google to never show your ad for. It is the brake on broad match — the list where you put competitor names, services you do not offer, towns you do not serve, and tire-kicker words like "free" or "jobs." Your account has zero of them. So you had the widest net turned all the way on, and nothing pulling anything back out of it.
MEASURED · 0 negative keywords in the entire accountThis is the answer to "how am I paying for my competitors?" You never added "Roto-Rooter" or "Bob Oates" as keywords. You did not have to. Broad match volunteers your ad for those searches on its own, because it sees "plumbing" and matches. The only thing that stops it is a negative keyword, and you had none. So Google was free to spend your money on every plumber's name in the region. Here is what that looked like.
Where the money went
You paid to send people to your competitors.
These are real searches from your account, the ten most expensive ones. Most are people typing a competitor's name into Google. Your ad showed up anyway, they clicked, you paid, and they were looking for someone else the entire time.
| The search someone typed | You paid | Clicks | Calls |
|---|---|---|---|
| drain pro (competitor) | $135.60 | 1 | 0 |
| aurora plumbing phone number (competitor) | $101.08 | 2 | 0 |
| abv plumbing (competitor) | $95.97 | 1 | 0 |
| roto rooter (national chain) | $86.77 | 2 | 1 |
| toms plumbing sequim (~2.5 hrs away) | $74.11 | 1 | 0 |
| bob oates plumbing (competitor) | $66.12 | 1 | 0 |
| wallner plumbing (competitor) | $48.24 | 2 | 0 |
| ferguson plumbing seattle (supply house) | $38.57 | 1 | 0 |
| noble plumbing snoqualmie (competitor) | $37.66 | 1 | 0 |
| nickerson plumbing ada ok (Oklahoma) | $24.86 | 3 | 1 |
| These 10 searches alone | $708.98 | 15 | 2 |
Pulled straight from your account's search-terms report. "Calls" is the conversions Google recorded for that search.
Those ten searches burned $708.98 and produced two "calls" — and both of those are the junk from the tracking section (a click on the Roto-Rooter name and the search from Oklahoma). Every competitor name on that list is a search you could have blocked with a single negative keyword. None were blocked, because there were no negative keywords. And this is just the top ten of a much longer list.
How big the problem is
81 cents of every dollar I can trace produced nothing.
Let me explain these numbers plainly, because they are the heart of it.
What "traceable" means. Google does not show you the exact wording of every search you pay for — a lot of it, especially from the Smart campaigns, gets lumped into buckets you cannot see into. Of your spend, $1,399 is the slice where Google does tell you the exact phrase someone typed. That is the part I can actually inspect. Within that slice, $1,133 — 81% — came from searches that led to no call at all.
Where the 1,328 comes from. You did not choose 1,328 searches. You had roughly 120 keywords, all on broad match. Broad match takes each one and fans it out across every phrasing Google thinks is related, so a handful of keywords turned into 1,328 different real-world searches that Google billed you for — competitor names, wrong services, wrong towns, and everything in between. That number is not a sign of reach. It is a sign of no aim.
The rest of what I found, and why each one happened
Smaller problems, same root cause.
Duplicate keywords
The same keyword is entered twice, so your campaigns compete with each other
13 of your keywords are entered more than once across different campaigns — "drain cleaning," "sewer repair," "hydro jetting," "commercial plumber," and more. When someone searches one of those, Google will not show two of your own ads at once; it picks one campaign and drops the other. So the duplicates do not buy you more reach. They split your budget and your data across two places, which means neither campaign gets a clear enough picture to improve. It happened because the account was built in layers over time and nobody cleaned up the overlap.
MEASURED · 13 keywords duplicated across campaignsOff-target keywords
Someone deliberately added generic and out-of-town keywords
This one is not a default — someone typed these in as targets: "commercial plumber," "bellevue plumber," "bothell plumber." You are a Kirkland drain and sewer specialist, and you rank #1 on the map for exactly that work. Instead, budget was pointed at general "plumber" searches and neighboring cities where you do not lead, and then broad match widened each of those even further. So the aim was off before broad match ever got involved, and broad match made it worse.
MEASURED · keywords like "commercial plumber," "bellevue plumber," "bothell plumber"Location — the part that is mostly OK
Your targeting is aimed at the right area, but it is inconsistent, and it was not what let the far-away clicks in
Credit where it is due: your campaigns are set to show only to people physically in your area, and the areas are local Washington cities. That is set up correctly. Two notes, though. First, no two campaigns target the same list — one covers 16 cities, another 9, another uses a 14-mile radius drawn from your own shopfront (which reaches Seattle, Bellevue, Bothell, and Redmond). That inconsistency is just sloppiness. Second, and this is the important part: the Sequim and Oklahoma clicks were not a location mistake. Those searchers were right here in your area — they were simply searching for a plumber somewhere else, and broad match served your ad anyway. The fix for that is negative keywords and tighter matching, not location.
MEASURED · location set to "presence" (people physically in-area) on all search campaignsLocal Services Ads
Your LSA was switched on but Google was barely showing it
Here is what "barely working" means, specifically. Your Local Services Ad had an average budget ceiling of $2,100 a week — call it $8,000+ a month available. Over its entire life it spent $138. That is under 2% of a single month's budget, so budget was never the thing holding it back. It brought in 2 leads total, and 0 in its final week. When a business with your reviews (4.9 stars, 564 of them) and Google Guaranteed status still barely gets shown, the budget is not the ceiling — the account setup is. The automated bid and the way the profile is configured are keeping you out of the top spots. That is fixable, and your reputation is the hard part, which you already have.
MEASURED · $138 lifetime spend against a $2,100/week budget · 2 leads total · 0 in the last 7 daysThe bottom line
The ads were not broken. They were aimed at the wrong searches, with no honest way to keep score.
You were not throwing money at a platform that does not work. You were running the widest possible matching with no brakes, and a scoreboard that counted page views and out-of-state clicks as wins. Broad match on every keyword, zero negative keywords, and conversion tracking that logs the wrong things. All three are fixable, and none of it means starting over: tighten the matching, build the negative list so competitor and out-of-town searches are blocked, and rebuild the tracking so a conversion means an actual customer called. Then the budget goes to Kirkland drain and sewer searches — the work you already win the map for.