CPM, CPC and CPA calculator
CPM stands for cost per mille — cost per thousand, because the Latin numeral stuck when the term came out of print advertising. It survives because impressions are counted in numbers too large to be useful one at a time, and because it makes very different media comparable: a billboard, a podcast read and a display banner can all be quoted per thousand pairs of eyes.
Spend, impressions, clicks, conversions and revenue give the whole funnel. A 1,000 budget delivering 100,000 impressions and 2,000 clicks is a CPM of 10 and a CPC of 0.50; 100 conversions from those clicks is a 5% conversion rate and a CPA of 10. Break-even CPC is gross profit per conversion times conversion rate.
How to calculate CPM, CPC and CPA
CPM, CPC, CPA, CTR and conversion rate are five views of the same four numbers, which is why they belong on one page: a visitor who arrived for one of them almost always wants two. CPM measures cost of reach, not value, and that is both its use and its trap: cheap reach against the wrong audience buys nothing, so it compares placements and never judges a campaign. Cost per click is only half a number too; the other half is revenue per click, and the campaign works whenever the second exceeds the first, which is why a 4 CPC is excellent for a business converting at 5% on a 300 order and ruinous for one converting at 1% on a 30 order. CPA is the product of the two, and knowing which component is responsible changes what you do: a 40 CPA from a 2 CPC at 5% conversion is an auction and bidding problem, while a 40 CPA from a 0.40 CPC at 1% is a landing page and offer problem. Reporting CPA without its two components is the most common way a marketing report manages to be accurate and useless at the same time. Click-through rate is the diagnostic that pairs with conversion rate: both high means the ad and the page agree, high CTR with low conversion means the ad is writing cheques the page does not honour, and low CTR with high conversion usually means the targeting is narrow but correct. The break-even CPC tab compresses all of it into one line, profit per conversion times conversion rate, which is why two advertisers can rationally bid twenty times apart on the same keyword and neither be wrong. The budget-split tab sits at the other end of the same argument, dividing a period figure across channels by weight; take any agency fee off the total before splitting it, so the shares cover media alone, and set the platform caps daily rather than monthly.
What people use it for
- Comparing display placements or publishers
- Budgeting a brand awareness campaign
- Checking a media pack quote
- Setting a maximum bid for a keyword
- Comparing keyword or campaign efficiency
- Setting a target CPA for automated bidding
- Deciding whether to fix bids or the landing page
- Testing ad creative against each other
- Diagnosing where a funnel is leaking
- Finding the highest click price a keyword can carry
- Planning a quarterly media budget across channels
- Setting daily caps per campaign
- Reallocating a budget after a monthly performance review
- Setting a manual CPC ceiling, and explaining it to a client
- Checking an agency report line by line
- Modelling a test budget alongside always-on spend
Questions
Divide spend by impressions, then multiply by 1,000.
Cost per mille. Cost per thousand impressions. The M is the Roman numeral for a thousand.
Entirely dependent on platform and audience. From a couple of pounds for broad display to well over fifty for narrow professional targeting.
Divide total ad spend by the number of clicks it produced. It is the headline number in any PPC account.
One below your revenue per click, which is total revenue divided by clicks. Absolute benchmarks are close to useless across different margins and order values.
Cost per acquisition: the ad spend required to produce one conversion. Cost per lead is the same arithmetic where the conversion is a form fill rather than a sale.
Up to the gross profit a conversion generates, or up to lifetime value if the customer is likely to return.
Clicks divided by impressions, multiplied by 100.
Search ads often run 3–6%, display well under 1%, and email 2–5% of delivered. Compare within the same channel only.
Multiply gross profit per conversion by the conversion rate. A 50 profit at 4% gives a 2.00 break-even click, and in Google Ads that is the ceiling a manual CPC bid has to stay under to be profitable. With no conversion history to work from, run a small budget at a low bid until you have thirty or so conversions rather than guessing the rate.
No. Bid below it, or the campaign returns exactly nothing for the effort and risk. It also explains why a competitor outbids you on the same keyword: a better conversion rate or a fatter margin makes the identical click worth more to them.
CPA usually counts media spend against a specific conversion. CAC counts all sales and marketing cost against new customers won.
Split it into CPC and conversion rate. Whichever is out of line against your own history is the one to work on.
Not on its own. An overclaiming ad earns clicks and loses them at the landing page.
Because display impressions are largely unsolicited. A 0.1% display CTR can be perfectly healthy.
On search platforms, yes. CTR feeds quality score, and a higher quality score lowers the cost per click for the same position.
No. Any numbers in proportion work: 2 and 1 splits two thirds to one third. Weight by recent efficiency, but keep a slice for testing, because a budget allocated entirely to proven channels never finds a new one.
Daily, in most platforms, and take each channel figure from its own share rather than from the total. A monthly figure entered as a daily cap is the classic way to spend a month of budget in a weekend.
Take them off the total before you split it, so the shares cover media only and the fee does not distort them.
Monthly is usually enough. Weekly changes rarely gather enough data to justify themselves, and each one restarts whatever learning phase the platform is in.
More competition in the auction, a lower quality score, or a drift in the keyword mix toward more expensive terms. Check the mix first: it moves without anyone changing a bid.