Five conversions on $639: what Plyto's own Google Ads did in 30 days
Nov 17, 2026
Zero conversions on the campaign built to convert
The number that stopped me cold: our "Plyto campaign - Conversions" campaign spent $88, pulled the best CTR of the account at 1.79%, got 73 clicks at the lowest CPC we paid anywhere ($1.21), and converted exactly 0 people.
Meanwhile the scrappy "Already running ads" audience test, which cost more per click and had a worse CTR, produced 4 of our 5 total conversions.
That is the headline of the last 30 days inside Plyto's own Google Search accounts. Here is the full picture, and what we are changing because of it.
The 30-day numbers, unrounded
Total spend: $639. Total impressions: 29,723. Total clicks: 354. Average CPC: $1.81. Total conversions: 5.
Broken out by campaign:
- Oct 30 test, Audience, Already running ads: $228, 11,513 impressions, 121 clicks, $1.89 CPC, 1.05% CTR, 4 conversions
- Oct 30 test, Audience, In-house marketers: $114, 4,833 impressions, 59 clicks, $1.93 CPC, 1.22% CTR, 0 conversions
- Oct 30 test, Audience, Local service owners: $142, 7,389 impressions, 74 clicks, $1.92 CPC, 1.00% CTR, 1 conversion
- Oct 30 test, Audience, Agencies: $67, 1,904 impressions, 27 clicks, $2.48 CPC, 1.42% CTR, 0 conversions
- Plyto campaign, Conversions: $88, 4,084 impressions, 73 clicks, $1.21 CPC, 1.79% CTR, 0 conversions
Signups across the business in the window: 30 total. 26 came from an ad click. 4 did not.
So the ad accounts did produce signups. They just did not produce them where the campaign names suggested they would.
What the landing pages tell us
We tracked three things per landing page: visits, visitors who read past 15 seconds, and visitors who started typing into a form.
- / : 189 visits, 106 read, 11 started
- /go/chatgpt-ads: 166 visits, 89 read, 21 started
- /go/local: 87 visits, 43 read, 2 started
- /go/marketers: 77 visits, 38 read, 1 started
- /go/agencies: 41 visits, 22 read, 1 started
- /lp/cu9oc8cefbem: 26 visits, 22 read, 0 started
The read rate on /lp/cu9oc8cefbem is wild: 22 of 26 visits got past 15 seconds. Nobody typed. Whatever that page says, it holds attention and then fails to ask for anything people want to answer.
/go/chatgpt-ads is the opposite kind of outlier. 21 people started typing on 166 visits. That is a 12.7% type-start rate, roughly double the homepage (5.8%) and six times /go/local (2.3%). The audience that arrives on that page wants to do something.
/go/marketers and /go/agencies sit at 1 type-start apiece. The two campaigns feeding them (In-house marketers and Agencies) also delivered 0 conversions between them on $181 of spend. The pages and the ads are telling the same story: these audiences click, but they do not want what we are showing them.
Three things a small business can do with a read like this tomorrow
1. Separate "cheap clicks" from "paying clicks" before you scale anything. Our Conversions campaign had the best CTR and the lowest CPC in the account and converted no one. If we had scaled on CTR and CPC alone, we would have poured budget into a campaign that produced nothing. Check conversions at the campaign level before you touch bids. A click you can afford is not the same as a click that pays.
2. When one audience is pulling 80% of the conversions, give it 80% of the budget. "Already running ads" was 36% of our spend and 80% of our conversions (4 of 5). The In-house marketers and Agencies campaigns together ate 28% of spend for 0 conversions. The honest move for next month is to cut or pause those two, push the saved budget into "Already running ads," and keep "Local service owners" on a short leash since it has at least shown one.
3. Fix the pages where people read but will not type. /lp/cu9oc8cefbem is the clearest example in the account: a 85% read-past-15-seconds rate with a 0% type-start rate. People are engaged and then nothing asks them a question they want to answer. Same pattern, softer, on /go/marketers and /go/agencies. The fix is almost always a shorter first field, a clearer single ask, and a form that matches the specific promise of the ad that sent them.
The honest summary
$639 in, 5 conversions out, and 26 of 30 business signups attributed to ad clicks. The best-looking campaign on vanity metrics was the worst on revenue. One audience carried the account. One landing page is quietly doing more work than any other, and one is quietly wasting attention we paid for.
That is the whole point of running the loop we run: the numbers tell you where to put the next dollar, and they rarely tell you what you expected before you looked.
Plyto is your AI marketing assistant: it makes and runs your ads
It writes the ad, builds the landing page, captures the leads and shows you which dollar came back. Try it free, no credit card.
Start free