How much do Facebook ads cost, and what actually sets the price

There is no rate card. You are bidding against other advertisers for the attention of specific people, and the price gets settled thousands of times a day in an auction you never see. Learn how that auction works and you can calculate your own number, which is worth more than anyone else's average.

What you are actually buying

Almost every lead or sales campaign bills the same way: you pay for impressions. Not clicks, not leads, not sales. Telling Meta to optimize for leads does not buy leads at a price, it tells delivery which people to buy impressions for, and you are charged for those impressions whether or not anyone fills in the form.

That is why CPM, the cost of a thousand impressions, is the only figure in the account that behaves like a price. Cost per click and cost per lead are spend divided by results after the fact. A campaign can post a good cost per lead one week and a bad one the next with the price of impressions unchanged: what moved was how many people responded.

A few objectives let you choose click-based billing, and there you do pay per click. On default settings for a conversion campaign you pay per impression, and every other cost figure in the report is derived from that one.

How the auction decides what you pay

Every time a person could be shown an ad, Meta runs an auction for that single impression, and every advertiser who wants that person is in it. The winner is not the highest bidder, and that one fact explains most of what you can do about your costs.

Meta ranks the entries by total value, which by its own description comes from three things: your bid, how likely Meta estimates that person is to do what you are optimizing for, and quality signals about the ad itself, including how often people hide or report it. A modest bid behind an ad people respond to can beat a bigger bid behind a weak one.

The winner then pays the minimum it took to win rather than what they were willing to pay. That is why ad quality shows up in your CPM at all: a better ad raises your total value, so a smaller bid clears the runner-up. You pay less for the same slot, you do not buy a better one.

On the default bid strategy you never enter a bid. Meta derives one from your budget, your schedule and how fast it needs to spend. A budget is not a bid, but it sets the pace, which is why raising it can raise CPM: the system now has to win more auctions per hour and gets less picky about which ones.

What actually moves your CPM

This is why "the average CPM in my industry" is a dead end. Your CPM is set by demand for the specific people in your targeting during the specific hours you run, not by your industry. Two plumbers in one city with different radius settings are in different auctions.

Read the third column before you act on anything in the first. Half of a CPM increase is a problem to fix and half is an entry price to plan around, and mistaking the second for the first is how people end up rewriting a perfectly good ad in November.

Why the end of the year costs more

The fourth-quarter rise is not a surcharge and not a policy. It is the same auction with more money in it: retailers who advertise once a year arrive between Black Friday and Christmas, and budgets that have to be spent before a fiscal year closes get spent. Attention did not grow to meet them. So a lead can cost meaningfully more in late November than in September with nothing wrong in the account. Budget for the higher entry price, or spend the quarter on audiences other advertisers are not fighting over, starting with people who already know you.

A daily budget is not what you will spend today

You set $50 a day, you open the account, yesterday reads $61. Nothing is broken. A daily budget is an average Meta paces toward, not a cap: it spends more on a day when the auction looks good and less on a day it does not. Meta's own guidance has put that overshoot at up to 25% above the daily amount, balanced across the calendar week. Treat the percentage as Meta's current rule rather than a constant, and check the unit that holds either way: seven days of spend divided by seven, never yesterday.

The charge on your card is a third number again. Meta collects when your outstanding balance reaches your billing threshold or on your monthly billing date, whichever comes first, and that threshold starts small and rises as you build payment history. A charge that does not line up with a day's spend is normal. A charge that does not line up with total spend for the period is worth a look.

Lifetime budgets pace differently again: the amount is spread across the whole schedule rather than per day, so daily spend swings wider by design. Neither budget type is a per-day guarantee.

The formula that gives you your own number

Here is the whole chain, and all of it is arithmetic you can check in your own reports. Impressions cost money, a fraction of impressions become clicks, a fraction of clicks become leads.

Cost per lead = (CPM ÷ 1,000) ÷ (click-through rate × landing page conversion rate)

Placeholder numbers, so you can see where yours go. A thousand impressions cost $18, 1.2% of them produce a link click, and 8% of the people who land on the page fill in the form. That thousand impressions buys 12 clicks at $1.50 each, those 12 clicks produce about one lead, so a lead costs roughly $18.75. None of the three inputs is a market figure. All three are numbers your own account already reports.

Invented inputs, real arithmetic: a CPM increase you do not control moves the number less than two improvements you do.

What a lead costs and whether you can afford it are different questions, and the second runs on figures Meta never sees: gross profit per customer times the share of leads you close. At $400 of profit per job and one win in five, a lead is worth $80 to you and $18.75 is comfortable. One win in twenty and the same campaign is break-even at best.

One more number belongs next to the price. Meta's benchmark is about fifty optimization events per ad set per week for delivery to learn properly, so a weekly budget near fifty times your cost per lead is the floor: about $940 in the example above. Cost per lead tells you whether ads can work for you, that threshold tells you whether you can afford to find out, and sizing that starting budget is its own subject with its own piece here.

Working out your own price, in order

1. 1.Pull the last 30 days and read CPM on its own, then against the 30 before it. That is your entry price and its direction, before any interpretation.
2. 2.Divide spend by link clicks and by leads for the same window. Both are derived numbers, and they mean something only with CPM beside them.
3. 3.Compare the daily budget with a full seven days of spend, never with yesterday. Uneven days are the pacing working as designed.
4. 4.Write down the two figures Meta cannot see: gross profit per customer, and the share of leads you close. Their product is your ceiling.
5. 5.Hold that ceiling next to your cost per lead. Room to spare turns the question into volume. No room means the fix is the offer or the funnel, not the bid.

Every step is a comparison against the account's own history, and that is the work that disappears in a busy week. It is also what Aevin's built-in analyst does on a schedule: it reads the account, reports which of these numbers moved and by how much, and shows the figures the finding came from.