Real Google Ads Account Teardown — Week of Oct 5

Source: YouTube auto-generated captions, lightly cleaned (repeated caption lines removed, misheard brand names corrected).

0:00 — Thirty days of waste

One live account spent $2,21 over the last 30 days. $912 of that went to search terms with zero conversions. Those are 72 distinct terms, all negative keyword candidates, and the average CPA sits at $184. A weekly report can look fine, while much of the month is still buying queries that never convert. That is the

0:25 — This week versus last

problem. We start from the week of September 28th through October 4 spent $479 against $413 the week before, up 16%. Conversions moved from $1.0 to $5.0 and CPA fell from $413 to $96. The 30-day figure says the account has a query problem. The weak says whether that problem is still buying traffic right

0:52 — Why the order matters

now or whether a different campaign just moved the numbers. Read pacing first, then find what moved, then explain why, then name the fix. That order matters because a cheaper CPA can come from one campaign while another spends more and converts nothing. Name the fix too early and you pause the wrong campaign or you raise budget on the one already wasting money. Averaging the week hides that

1:17 — Step 1 — Read pacing

split. Start with pacing, not with keywords. Spend went from $413 to $479, up 16%, while conversions went from 1.0 to 5.0, and CPA dropped from $413 to $96. You read pacing first because a spend increase with a CPA drop is not automatically healthy. It only says the account delivered more for less. It does not say which campaign paid for it. Hold pace next to efficiency before you celebrate the CPA. Spend rose 16% and conversions rose from 1.0 to 5.0. So efficiency improved faster than spend.

That is why you do not cut budget off one bad prior week. A CPA of $413 on one conversion is a thin base. Five conversions at $96 is better, but the

2:11 — Step 2 — Find what moved

sample is too small to scale. Once pacing is clear, split the account by campaign. The total improved, so the job is to find what moved, not to assume every campaign improved with it. A blended CPA of $96 can hide one campaign that spent more and converted nothing. You separate them because the fix for a winner is not the fix for a campaign that added spend without a conversion. Campaign A spent $272 and recorded 2.0 zero conversions with spend up 10% versus the prior week. That is the largest share of the $479 and it did convert. You mark it because a campaign that spent more and still converted is not the leak. Leave it alone while you hunt the one that took a bigger increase and returned zero.

Campaign B spent $192 and recorded 0.0 conversions with spend up 28%. That is the move that matters. It took a larger increase than campaign A and return nothing. Which is why the account CPA is not as clean as the headline. You isolate B because cutting waste here changes the week. Touching A does not fix a campaign that converted zero. Campaign C spent $10 and recorded 1.0 conversion with spend down 14%. It is small and it did convert. So, it is not the leak and not the growth engine. You still name it because skipping the small converter makes the account look like only two campaigns. C can convert on little spend. B can spend $192 and convert nothing. Those are different problems. Now, explain why B moved the wrong way. Over 30 days, this account

3:58 — Step 3 — Explain why

sent $912 to search terms with zero convergence across 72 terms. A campaign that raises spend 28% and converts nothing may be buying more of those terms, not failing at the bid. You explain the query before you touch budget because a lower bid on a junk term still buys junk. The week still improved and that needs an explanation, too. Conversions went from 1.0 to 5.0, while B added spend and added zero. A and C carried the conversions. CPA fell from $413 to $96 because the converting campaigns outweighed B, not because B got healthier. Say that out loud so the account CPA is not treated as proof that every campaign is fine. Do not blame the bid strategy until the queries are checked. Average CPA for the month is $184 and this week it printed $96. So bidding is not obviously broken. B failed because spend rose 28% into terms that

do not convert. The same pattern as the $912 with zero conversions. Rule the bid out so the fix stays on the query list. Name the fix as a negative list, not a budget cut. 72 terms with zero

5:18 — Step 4 — Name the fix

conversions absorbed $912 of the $2,21. add them as negatives where they can match. Starting with campaign B because that is where new spend is still landing on nothing. A blanket budget cut would also shrink A which just produced 2.0 conversions. The waste is the terms on campaign B cap the spend increase until the negatives are in. It rose 28% to $192 with 0.0 conversions. So an open budget buys another week of the same terms. Cap it because a negative takes time to prove and the budget will not wait. Do not pause B. If converting themes inside are real, remove those terms, then judge what remains. Leave campaign A and campaign C on their current budgets. A spent $272, up 10% with 2.0 conversions. C spent $10, down 14% with 1.0 zero conversion.

Leave them because the fix was named for the leak, not the account. Raising A would scale a two conversion week. Cutting C would delete a cheap conversion. Neither action explains the $912. Check the same comparison next week before you call it fixed. You want B spend flat or down, and you want conversions on A and C to hold. If B still spends near $192 with zero conversions after the 72 negatives, the match is wider than the list. So add phrase negatives, not a new bid strategy. Recheck because an unmeasured fix is an opinion. The week spent $479 against $413.

Conversions moved from 1.0 to 5.0 and

7:05 — What the week showed

CPA fell from $413 to $96. Under that sat $912 of 30-day search spend with zero conversions and 72 terms still need negatives. B spent $192 up 28% with 0.0 conversions. A and C produced the conversions. Read pacing. Find what moved. Explain why. Then name

7:32 — Recap

the fix. Pull search terms before you change a bid. Put the 720 conversion terms on the negative list. starting in the campaign that spent more and converted nothing. That is how a 96 CPA stops hiding a $192 leak. The blog is on ahmeego.com. Body by Amigo, a company of It All Started with an idea.