Cost per click is the price an advertiser pays when someone clicks an ad, not when the ad is merely shown. Investopedia's explainer and Amazon Ads' guide define it the same way: you pay only when a person clicks through to your site or landing page.
For advertisers that is a budget line. For anyone researching keywords it is something more useful: a public trace of what businesses are willing to pay for a visitor who typed a particular phrase. Read carefully, it is one of the best value signals in keyword data.
How the Auction Sets CPC
CPC stands for cost per click. PPC, pay-per-click, is the advertising model that charges that way. Amazon Ads puts the difference in one line: PPC is the system, CPC is the metric that measures it.
The price itself is usually set by auction. Each advertiser sets a maximum bid, the most it will pay for one click, and the platform weighs that bid together with competition for the keyword, demand for the term, the quality of the ad, and how visible the placement is. Amazon Ads states the rule plainly: the more in-demand the keyword and the more visible the spot, the higher the cost.
Three terms travel with the auction in Amazon Ads' guide:
- Average CPC: the mean spent per click, which shifts as positions change.
- Maximum CPC: the most that can be charged for one click, the base bid plus any adjustments from settings such as automated bidding or cost controls.
- Manual vs automated bidding: with manual bidding you set each bid; with automated bidding you set a budget and the platform adjusts bids for you, an option the guide notes across Google Ads, Microsoft Advertising, and display networks.
The CPC Formula
CPC is total ad cost divided by total clicks. Spend $500 and get 100 clicks, and each click cost $5. Reddit for Business's glossary works the same example with $600 over 100 clicks, which comes to $6.
The arithmetic is trivial, but the inputs are not. Whenever you log a CPC for keyword work, keep the two inputs beside it: what spend and what click count the average came from, or, when no spend sits behind it, which tool's estimate you are quoting. A CPC with no source is a number you cannot defend later.
CPC vs CPM, CPA and CTR
Amazon Ads frames CPC against CPM, cost per mille, the price per 1,000 impressions, as clicks versus views. You pay for CPC when someone acts and for CPM whenever the ad is seen. Its guide ties CPC more closely to purchases and CPM to awareness, and suggests using both for a fuller picture. CTR, click-through rate, links them: it is the share of views that turn into clicks. CPA, cost per acquisition, goes one step further and prices the sale or sign-up itself.
For keyword research the point of the list is simple. CPC is the metric that reflects intent to act. CPM tells you what attention costs; CPC tells you what a visitor who wanted something costs.
Weighing CPC with Volume
Among people wiring keyword data into scripts and agents, CPC works as a value operand next to volume. Volume counts searches. CPC says what advertisers pay for a click out of that pool. Multiply the two and a small, expensive phrase can outweigh a big, cheap one.
The best-known version is a scoring prompt that weights volume times CPC against inverted difficulty and sorts the results into labeled buckets, and how keyword difficulty scores work walks through why the difficulty half of that formula varies by vendor. The lesson carries over whatever weights you choose: set price against ease, and name what each keyword you keep is meant to earn.
An ads-workflow write-up from late August runs the same logic as a filter. Ask an AI model for keyword ideas, then run every idea through a SERP API for real volume and CPC before anyone trusts it. The author's numbers: 50 ideas in 30 seconds, filtered in an hour to the 10 that matter, and campaign research that took two days finished in two hours. The model suggests; the data decides.
DIY keyword tools shared by builders tend to carry the same five columns: suggestions (the completions Google offers as someone types), volume, CPC, competition, and the results page for each query, with sorting, saved lists, and CSV export around them. Pull the columns together and CPC stops being an ads number and becomes a ranking key.
CPC Thresholds and What They Can and Cannot Tell You
Some researchers turn CPC into a gate. One public sweep of 85 local-service niches across 510 live Google results pages kept only niches clearing four tests at once, one of them a CPC above $15, and found that 10 passed. Its top example was foundation repair, at about $60 a click.
Another thread filtered a list of domain names by the CPC of the phrase each one spells out. A high CPC means businesses already pay to reach that phrase; paired with volume, it says both how valuable and how large the audience is.
The gap between categories shows why a threshold needs context. One niche analysis logged fortune-telling terms at $0.04 to $0.36 a click against about $4.80 for car insurance, more than a tenfold difference between curiosity and insured risk. A $15 gate means something different on each side of that gap.
Why the Same Term Prices Differently by Source and Month
Two tools can show two CPCs for the same keyword, and neither is necessarily wrong. SEO Utils' release notes describe the problem and its fix: after comparing volume, difficulty, and CPC across sources, the tool added six switchable data sources, each labeled with its source and month. Estimates built from a modeled database, a live ads pull, and panels of real browsing behavior will not agree to the cent.
Copy the labeling habit. Write the source and the month next to every CPC you record, and a disagreement becomes information instead of confusion.
From CPC to a Decision
Build the shortlist from CPC-weighted rows with a source and month on each. Hear the survivors in searchers' own words by checking their suggestions. Then open the live results page for each, in the places you care about, and note who holds the organic spots and who is buying ads.
Without the dates, this month's bargain and last quarter's peak look like the same opportunity.
