PPCleaner

Google Ads Cost Per Click Too High? Check the Query Mix

Published November 17, 202612 min read
An ecommerce owner checking advertising invoices beside a laptop, calculator and packed orders.
The short answer

A high average cost per click is often not a price problem. Average CPC is total cost divided by total clicks across every search term you paid for. If broad match or Smart Bidding sends you into other industries' auctions, like "plumber liability insurance" or "plumbing franchise for sale", those clicks can cost two or three times your real keywords and drag the average up.

To check, open the search terms report, sort by cost, and compare each term's CPC with your account average, weighted by clicks. Block the expensive terms that never convert. Judge the result by cost per conversion, not by CPC. Blocking cheap junk can push your average CPC up even when the account is doing better.

Why your average CPC can be high when your keywords aren't

Your average CPC is an average over search terms, not keywords. If a few off-target queries land in expensive auctions, they can push the blended number well above what your real keywords cost. Before you lower bids or chase Quality Score, find out which search terms are paying the high prices.

Google calculates average CPC as total cost divided by total clicks. That number mixes together every query your keywords matched: the ones you wanted, the ones you put up with, and the ones you never knew about. With exact match only, the mix was narrow and the average told you something. With broad match, AI Max, Dynamic Search Ads or Performance Max, the mix is wide. The average now describes a portfolio, and you can't tell what's in a portfolio from its average.

The pages that rank for this question mostly give the same list: raise Quality Score, use long-tail keywords, adjust bids, tighten ad schedules and locations, add negative keywords. Each tip is fine on its own. All of them treat CPC as one price you pay for one thing. When the price is a blend, you first need to know what is in the blend. Otherwise you can spend a month improving the Quality Score of a keyword whose CPC was never the problem.

A worked example: six search terms, one misleading average

Here is an illustrative plumbing account over 30 days. The numbers are made up to show the mechanics, but the pattern is real: a few off-target terms in another industry's auction, plus some cheap junk.

Search termClicksAvg CPCCostConversions
emergency plumber near me250$6.00$1,50025
24 hour plumber100$5.00$50010
water heater replacement50$8.00$4005
plumbing franchise for sale60$15.00$9000
plumber liability insurance40$20.00$8000
plumber salary100$1.00$1000
Total600$7.00$4,20040

The account's average CPC is $7.00 ($4,200 ÷ 600). Cost per conversion is $105 ($4,200 ÷ 40).

Now look at what the owner believes they are buying. The three terms that convert cost $5 to $8 a click. None of them costs $7 on its own. The $7 average comes from two terms from other markets, franchise sales and business insurance, where the bidders are franchisors and insurers. Those two terms took 100 of the 600 clicks (one in six) and $1,700 of the $4,200 cost (about 40%). They produced no conversions.

If this owner searches "google ads cost per click too high" and follows the usual advice, they will try to lower the CPC on "emergency plumber near me". That term is already their cheapest source of leads.

Don't average the CPC column

A quick mistake: export the report and average the CPC column. The six CPCs above average to about $9.17 ($55 ÷ 6). That figure means nothing. It gives "plumber liability insurance" (40 clicks) the same weight as "emergency plumber near me" (250 clicks). Always divide total cost by total clicks. Weight every term by its clicks.

Blocking junk can make your average CPC go up

Negative keywords lower your cost per conversion. They do not always lower your average CPC. If the terms you block are cheaper than your average, removing them raises it. This is the part most CPC advice leaves out, and it causes real trouble when a client or manager watches CPC as the main KPI.

Run three scenarios on the account above. Assume the auction prices of the terms you keep stay the same. That is a simplification, and we come back to it below.

Scenario 1: block the two expensive off-target terms.

500 clicks, $2,500 cost. Average CPC falls from $7.00 to $5.00, a drop of about 29%. Cost per conversion falls from $105 to $62.50.

Scenario 2: block only the cheap junk ("plumber salary").

500 clicks, $4,100 cost. Average CPC rises from $7.00 to $8.20. Cost per conversion falls slightly, from $105 to $102.50. The account is a little better and the CPC chart looks worse.

Scenario 3: block all three non-converting terms.

400 clicks, $2,400 cost. Average CPC is $6.00, which is higher than in scenario 1. Cost per conversion is $60, the best of the three.

The best outcome does not have the lowest CPC. If you report CPC alone, scenario 1 looks like the winner and scenario 3 looks like a step backwards. Report cost per conversion next to CPC, and explain the mix change whenever CPC moves after a round of negatives.

The assumption to watch

In the example, the kept terms keep their price after the blocks. In a real account, that depends on your bidding:

  • Manual CPC or Enhanced CPC: prices on kept terms mostly stay where they were. The example holds fairly well.
  • Maximize Conversions with no target: the budget you free up is spent again. The bid strategy may bid higher on the terms you kept, or find new queries to fill the gap. Your spend may not fall at all. You will get more of the good traffic, and maybe some new junk.
  • Target CPA or Target ROAS: the strategy bids against its target, so freed budget is not spent as aggressively. Its conversion data also gets cleaner, which affects future bids. We cover that in how junk queries skew Smart Bidding.

Whichever strategy you use, the first effect is the same: the money stops going to auctions that were never going to pay back.

How to split your average CPC by search term

Take a 30 to 90 day export of the search terms report with clicks, cost and conversions. For each term, calculate how far its CPC sits from the account average, multiplied by its clicks. The terms with the biggest positive values are the ones pulling your average up.

Step by step:

  1. Open the search terms report at campaign level for Search and Shopping, and for Performance Max if you run it. Add the Clicks, Cost, Avg. CPC and Conversions columns. Pick a range long enough that a term with 10 or 20 clicks has had a fair chance to convert. For most lead-gen accounts that means 60 to 90 days.
  2. Export to a spreadsheet. Don't filter yet. If you want to cut the list down before exporting, the filters that surface waste can help, but keep the totals from the full export for the next step.
  3. Compute the true average: total cost ÷ total clicks. In the example, $7.00.
  4. Add a "CPC drag" column: (term CPC − account average) × term clicks. In a sheet: =(D2 - $H$1) * B2, where D is the term's CPC, B its clicks and H1 the account average.
  5. Sort by CPC drag, largest first. Positive values pull the average up. Negative values pull it down.

For the example:

Search termCPC drag
plumber liability insurance+$520
plumbing franchise for sale+$480
water heater replacement+$50
24 hour plumber−$200
emergency plumber near me−$250
plumber salary−$600

The drag column always sums to zero. Here: 520 + 480 + 50 − 200 − 250 − 600 = 0. That's a useful check that your formula is right. The two off-target terms sit at the top. "Water heater replacement" also pushes the average up a little, but it converts, so that is a normal cost of doing business.

  1. Add the conversion column back in. CPC drag tells you what makes the average high. Conversions tell you whether that matters. A high-drag term that converts is an expensive good term. A high-drag term with zero conversions over a meaningful click count is the problem you came looking for.

Group by word, not only by term

Single terms are often too small to judge. "plumber liability insurance", "plumbing business insurance cost" and "insurance for plumbing contractors" might have 8 clicks each. Separately they look like noise. Together they are 24 clicks at a high CPC with no conversions. In the spreadsheet, tag each term with the theme words it contains ("insurance", "franchise", "salary", "jobs", "course", "software") and sum clicks, cost and conversions by tag. The low-volume terms add up article shows this grouping in more detail.

If you do this through the API instead of the interface, Search, Shopping and Performance Max search terms are reported in campaign_search_term_view. Cost comes back in micros, so divide by 1,000,000 before you compute averages.

Why off-target queries cost more than your real keywords

Off-target queries are often expensive because you are bidding in someone else's market. When broad match or a Smart Bidding strategy maps "plumber" to "plumber liability insurance", you enter an auction set by insurers. A lead is worth a lot to them, so they bid high. You pay close to their prices for a click that is worth nothing to you.

Some patterns we see again and again in the accounts we audit:

  • B2B terms around a consumer service. Insurance, software, franchise, wholesale and supplier queries attached to a trade. "dental practice management software" in a dentist's account. "HVAC franchise opportunities" in an HVAC contractor's account.
  • Career and training terms in some fields. Most job queries are cheap ("plumber salary" at $1). Some are not. Nursing schools, trucking schools and coding bootcamps bid hard, so "become a dental hygienist" or "CDL training near me" can cost more than the service terms you meant to buy.
  • Legal and finance words. Any query containing "lawyer", "attorney", "settlement", "loan" or "financing" can match into some of the most expensive auctions in Google Ads, even if your business only shares a noun with them.
  • Competitor and brand terms. Bidding on another company's brand usually means a lower Quality Score and a higher price. If that is a choice you made, fine. If broad match made it for you, it belongs in this audit. For your own brand inside Performance Max, see measuring branded search spend in PMax.

Notice that the usual advice doesn't touch any of these. A better ad and landing page for "emergency plumber" won't lower what you pay for "plumber liability insurance". Quality Score is calculated for your keywords, and the auction price for an off-target query is set by advertisers who actually want that query. The only fix is to stop entering the auction.

Blocking the expensive terms without breaking anything

Use phrase or exact match negatives built from the theme words you found, add them where they apply, and check the list against your converting terms before you save. The goal is to remove the expensive non-converters without touching a term that brings in leads.

A few practical rules:

  • Prefer phrase and exact. Broad negatives behave unpredictably across match types. A phrase negative like "liability insurance" blocks the theme without risking "water heater" traffic. For single bad terms, exact match negatives are the safest block. For theme words, there are phrase match examples you can copy.
  • Check against conversions first. Before adding "insurance" as a negative, search your converting terms for that word. A plumber who works on insurance claims may convert well on "water damage insurance claim plumber". In that case use a narrower phrase like "liability insurance", or exact negatives.
  • Use shared lists for Search and Shopping. Standard Shopping accepts negative keywords and shared lists, like Search does, so one "Other industries" list can cover both.
  • Performance Max is different. A PMax campaign accepts up to 10,000 campaign-level negative keywords, but no shared negative keyword lists. You have to add the negatives to each campaign. If they seem to do nothing, this is usually why.

For the full decision process, how many clicks before a term counts as a non-converter and what to do with ambiguous ones, see adding negatives from search terms without guessing.

When the high CPC is real

Sometimes the split shows that your converting terms are the expensive ones. In that case negative keywords won't help. Your market is expensive, and the levers are conversion rate, the value of a lead, and your bid targets. Do the split first so you know which situation you are in.

There are three outcomes from the CPC drag analysis, and each one leads somewhere different:

  1. High-drag terms are off-target and don't convert. This is a mix problem. Negatives fix it, and the rest of this article applies.
  2. High-drag terms are on-target and convert. You are paying market price for valuable queries. A personal injury firm paying a lot for "car accident lawyer" doesn't have a negative keyword problem. Look at landing page conversion rate, lead quality, and whether your target CPA reflects what a client is actually worth. Industry CPC benchmarks won't help much here: they compare your blended average with other accounts' blended averages, and the mix behind each one is unknown.
  3. No term stands out, but CPC rose over time. Compare the same split across two periods. If the same terms got more expensive, competition went up or your bid strategy changed. If new expensive terms appeared, your match types or a new campaign type widened the mix.

There is also a limit to how much the report can show. The search terms report doesn't list every query. Part of the cost sits in rows Google doesn't show, and you can't split what you can't see. We explain how to estimate that gap in where the missing search term cost went. If a large share of your cost is hidden, keep that in mind when you read the drag table.

Display, Video and App campaigns report no search terms at all. If their clicks are part of the average you're worried about, this method doesn't apply to them. Look at their CPC separately, by campaign, and never blend them with Search.

Keeping the mix clean after the first pass

A one-off cleanup fixes the average for a few weeks. Broad match and Performance Max keep finding new queries, so new expensive off-target terms keep showing up. The work is a habit, not a project.

If you do it by hand, repeat the drag analysis every month on the last 30 days, and do a quick pass on the newest terms every week. Analyzing a search terms report in 20 minutes covers the weekly routine.

PPCleaner does the blocking part every night. It reads the search terms each Search, Shopping and Performance Max campaign paid for, judges which ones were never going to convert, and adds them as negative keywords. It never blocks a term that converted in the last 90 days, and every block can be undone in one click for 90 days. It does not touch your bids, budgets, ads or keywords. That means if your high CPC turns out to be outcome 2 above, real market prices on terms that convert, PPCleaner won't change it. That decision is yours.

Whichever way you do it, watch cost per conversion next to CPC, and expect the CPC chart to move in both directions as the mix changes. If the junk you cut was cheap, a higher average CPC after cleanup is the sign the account is buying more of what converts.

cost per clicksearch termsnegative keywordswasted spendgoogle ads

Frequently asked

Will adding negative keywords lower my average CPC?

Only if the terms you block cost more per click than your average. Blocking expensive off-target terms lowers it. Blocking cheap junk, like job or salary searches, raises it. In both cases cost per conversion should improve, so judge the change by that number, not by CPC alone.

What is a good cost per click for Google Ads?

There is no useful universal figure. Industry benchmarks compare blended averages whose search term mix you can't see. A better test is to split your own average by search term, then check whether your converting terms got more expensive over time.

Does improving Quality Score fix a high CPC?

It can lower what you pay on your own keywords. It does nothing for off-target queries that broad match puts into other industries' auctions, where other advertisers set the price. Those need negative keywords, not better ads.

Can I split CPC by search term in Performance Max?

Yes. Performance Max search terms are reported in campaign_search_term_view, along with Search and Shopping, so you can run the same analysis. To block what you find, add campaign-level negatives (up to 10,000 per campaign). PMax doesn't accept shared negative keyword lists.

Why did my CPC jump after switching to broad match or Smart Bidding?

Usually because the mix got wider. Broad match and automated bidding reach queries your old exact and phrase keywords never matched, and some of those queries sit in expensive auctions. Compare the search term split before and after the switch to see which new terms carry the extra cost.

About the author

Alain Abulafya

Founder of PPCleaner · Google Ads consultant & media buyer

Alain has worked in digital marketing for more than 15 years, over 10 of them managing and optimizing Google Ads campaigns. He has worked with more than 100 advertisers in e-commerce and lead generation, and he is a Google Partner. He built PPCleaner to automate the search term reviews he used to do by hand for his clients.

Stop the leak

Clean your search terms. Save time and money.

Every night, PPCleaner reviews the searches your campaigns actually paid for, groups them by intent, and excludes the ones that never convert. No spend threshold to wait for, nothing touched but negatives — and one click undoes any of it.

Start 7 days free

Search, Shopping and Performance Max · Cancel in one click

Related articles

Google Ads CPC Too High? Split It by Search Term — PPCleaner