Facebook ads cost what the auction charges to reach your audience, so the question worth answering is what a purchase will cost you. The calculator below opens on Triple Whale’s 2026 ecommerce medians, a $15.06 CPM, a 2.39% click rate and a 1.53% conversion rate, and turns any monthly budget into impressions, clicks, purchases, cost per purchase and ROAS.
Facebook ads cost calculator
Defaults are Triple Whale’s 2026 Meta medians (40,000+ brands, Aug 2025 to Jul 2026). Replace them with your own Ads Manager figures once you have a clean month of data.
Formulas: impressions = budget ÷ CPM × 1,000; clicks = impressions × CTR; purchases = clicks × conversion rate; revenue = purchases × AOV; contribution = revenue × gross margin − budget. Benchmarks and planning only; your account’s rates decide the real result.
How the calculator works
One chain of arithmetic produces every figure on the screen. Divide the budget by the CPM and multiply by a thousand to get impressions. Multiply impressions by the click through rate to get clicks. Multiply clicks by the conversion rate to get purchases. Divide the budget by purchases to get the cost of each one. Multiply purchases by the average order value to get revenue, and divide that revenue by the budget to get ROAS. Move any single input and the whole chain moves with it, which is what makes the tool useful: it shows which lever in your account is worth pulling first.
| Output | Formula | At the 2026 medians with a $20,000 monthly budget |
|---|---|---|
| Impressions | Budget ÷ CPM × 1,000 | 1,328,021 |
| Clicks | Impressions × CTR | 31,740 |
| Purchases | Clicks × conversion rate | 486 |
| Cost per purchase | Budget ÷ purchases | $41.18 |
| Revenue | Purchases × average order value | $35,625 |
| ROAS | Revenue ÷ budget | 1.78 |
Two optional fields turn it from a generic ad calculator into one built for an ecommerce brand. The first is new customer share, the proportion of purchases you expect from first time buyers. With it, the calculator reports cost per new customer, which is the figure a brand trying to grow should be managing to. The second is gross margin, what each order contributes once the cost of goods is paid. With it, the calculator reports contribution after ad spend, which is the figure that says whether the budget is making money or only making revenue. The calculators that currently rank for this search stop at ROAS, and ROAS on its own will call a campaign a success while the store loses money on every discounted order it ships.
Use the preset buttons to load the all category medians or the apparel and accessories medians, then overwrite them with your own rates from Ads Manager once you have a month of clean data. The medians are the right place to start and the wrong place to stop.
The 2026 benchmarks behind the defaults
The default values come from Triple Whale's 2026 Meta ads benchmark, published on 18 August 2026 and drawn from more than 40,000 brands between August 2025 and July 2026. Two cautions apply before you lean on them. They are medians, so half of all brands sit on the wrong side of each one. And they are platform reported, so they carry the same attribution generosity your own Ads Manager does. They work as a sanity check for an account and fail as a target for one.
| Metric | 2026 median | Year over year | What it does inside the calculator |
|---|---|---|---|
| CPM | $15.06 | +13.24% | Fewer impressions per dollar than a year ago |
| CTR | 2.39% | +15.97% | Creative is earning more clicks from each impression |
| Conversion rate | 1.53% | -4.73% | A smaller share of those clicks becomes an order |
| CPA | $38.99 | +3.14% | Cost per purchase rose, though far less than CPM did |
| Average order value | $73.36 | +3.50% | Larger baskets absorbed part of the increase |
| ROAS | 1.88 | +0.57% | Flat, the net result of everything above |
Category moves the numbers further than most people expect. The same report puts the median cost per purchase at $36.98 for apparel and accessories, $39.31 for beauty and $40.53 for health and wellness, and for apparel and accessories it reports click through at 2.44%, conversion at 1.47% and ROAS at 2.24. If you sell in one of those categories, begin from the category figures rather than the all category ones. Then replace the average order value with your own, because order value changes the answer more than any benchmark does. A $110 basket makes a $45 purchase cost comfortable. A $55 basket makes the same cost fatal.
Why your cost per purchase will not match the median
CPM belongs to the auction. It reflects how many advertisers want the same people you want, in the same week, and your influence over it is small. Meta's Q2 2026 earnings call disclosed a 12% year over year increase in the global average price per ad alongside a 14% increase in impressions served, which is the whole market paying more for more inventory at once. Nothing you do inside your account changes that trend.
The two rates between impressions and orders are a different story. Click through rate is mostly creative. With broad audiences and Advantage+ settings in play, the ad decides who sees it, so an opening that stops the right person raises click rate and lowers the price of every click behind it. Conversion rate is mostly offer, landing page and product, plus the quality of the purchase signal Meta learns from. On the same call Meta attributed a 15.7% lift in conversions on Facebook to improvements in its ranking models and said Advantage+ revenue was now running above $75 billion a year. Accounts that send one clean, deduplicated purchase event per order are the ones positioned to collect gains like that. Accounts that send duplicates and missing values pay the same rising CPM and get none of the lift.
Average order value is the multiplier on all of it, and the calculator treats it that way. Before you compare your cost per purchase with any benchmark, compare your basket with theirs. Most of the gap between two accounts with the same CPM is explained by the basket, not by the media buying.
How much should an ecommerce brand spend per month?
Enough to buy a readable number of purchases, and no more than the margin can carry at the cost per purchase you actually get. The table runs four monthly budgets through the calculator at the 2026 medians, then shows the conversion rate each would need to reach a 3.0 ROAS. At a 50% gross margin, a 3.0 ROAS leaves about 50 cents of contribution for every dollar of ad spend once the cost of goods is paid, which is the point at which growth starts to fund itself.
| Monthly budget | Purchases at the medians | Cost per purchase | Revenue | ROAS | Conversion rate needed for 3.0 ROAS |
|---|---|---|---|---|---|
| $5,000 | 121 | $41.18 | $8,906 | 1.78 | 2.58% |
| $20,000 | 486 | $41.18 | $35,625 | 1.78 | 2.58% |
| $50,000 | 1,214 | $41.18 | $89,062 | 1.78 | 2.58% |
| $100,000 | 2,428 | $41.18 | $178,125 | 1.78 | 2.58% |
The cost per purchase column repeats the same number because the calculator holds every rate fixed as spend rises. A real account will not oblige. Under roughly $5,000 a month most ad sets never leave the learning phase, so the rates they report are noise and no plan should be built on them. Past a few hundred thousand a month the campaign has to reach people further from its best audiences, CPM climbs, and the last purchase of the month costs more than the first. That is why we raise spend in increments, reading the blended cost per new customer after each increment before committing the next, instead of projecting a strong week across a quarter. The full logic for moving money between channels is on the paid media management page.
A brand weighing whether Meta deserves a larger slice of the budget can borrow the same publisher's other reports for context. Triple Whale's 2026 figures show a median cost per purchase of $28.14 on Google Ads and $17.07 on TikTok against $38.99 on Meta. The order of those three says mostly how close each platform sits to the moment of purchase. It says nothing about which one finds customers the other two cannot reach, and only a blended read of your own account can settle that.
Prefer the inputs filled in from your own Ads Manager rather than the medians? The free 30 minute growth audit reads your signal quality, structure and creative, and hands you the real numbers to plug in.
What the calculator cannot tell you
It cannot tell you whether a purchase was new. A campaign reaching last week's site visitors and last year's buyers will report a cost per purchase far under the median and add very little, because many of those people were going to buy without the ad. Put a defensible new customer share into the calculator, typically somewhere between 40% and 70% for an account weighted toward prospecting, and manage to the cost per new customer it returns rather than to the headline cost per purchase.
It cannot tell you whether Meta caused the order. Google, TikTok, email and affiliate partners each take credit for many of the same orders inside their own attribution windows, so adding up every platform's reported revenue produces a number larger than what the store took. Every week, line Meta's reported purchases up against the orders the store actually took and the first time buyer count in your analytics or in Triple Whale, and only then let the calculator's output move the budget. That weekly reconciliation is a standing part of any Facebook ads management service worth paying for, and in the accounts we audit it is more often missing than present.
It cannot tell you about the learning phase either. A fresh ad set, or a large change to an existing one, sends delivery back into learning, where cost per purchase runs high and swings from day to day. Read any change across at least a full week, and ideally two, before you decide what it did.
A worked example from a real account
Here is the calculator used the way it should be, on blended numbers rather than on Meta's own. From January to August 2026, a womens fashion brand I work with acquired each new customer for $67 across its paid channels, at a new customer ROAS of 2.17x. Those are blended figures, no incrementality test separated Meta's contribution from Google's, and so they do not prove what Meta alone achieved. What they give the calculator is something true to start from, which is more than a benchmark can offer, because a benchmark describes the median brand and this describes the account in front of you.
Enter $67 as the cost per new customer, with the brand's own average order value and gross margin, and the output is contribution after ad spend per new customer. That single number decides whether the next $10,000 of budget is worth spending. In this account it was, repeatedly, which is why Facebook spend grew 56% across the same window. The decision came from that calculation, run every week, and not from the ROAS printed in Ads Manager.
If Meta is the largest line in the budget and the dashboard looks healthier than the bank balance, that is precisely the account the audit was built for.
Common questions
Meta's own [pricing page](https://www.facebook.com/business/ads/pricing) makes the point that there is no fixed cost and that ads can run on a budget of a few dollars a day, but a budget only becomes readable once it buys a few hundred purchases a month. At the 2026 medians, $20,000 a month buys roughly 486 purchases at about $41 each. Brands putting $50,000 or more a month into paid media are the ones where senior management of the account changes the outcome.
One that still leaves a profit after the goods, the shipping, the discount and the returns have all been paid for. The 2026 median is $38.99 across categories and $36.98 for apparel and accessories, but a brand with a $110 average order can pay more than that and still profit, while a brand with a $50 basket cannot. Measure it against your margin, not the median.
CPM reflects competition for the people your ads reach, the quality Meta assigns your creative, and the season. The 2026 median is $15.06, up 13.24% in a year, so part of any rise is simply the market. A CPM far above that usually points to narrow audiences, tired creative, or a purchase signal Meta cannot learn from. Fixing creative and signal lowers the effective cost even when the auction price stays put.
Yes. Instagram ads are bought through the same Meta Ads Manager and priced by the same auction, so the math is identical. The medians above cover Facebook and Instagram together. For a planner that starts from a monthly target and splits the budget by Instagram placement, see the Instagram ads cost calculator.
Related reading
- Instagram Ads Cost Calculator: Plan a Monthly Budget by Placement
- Meta Ads for Black Friday: Structure and Budget Pacing
- BFCM Creative Testing: What to Test in October
- Ecommerce Attribution: Models, Tools and What to Trust
- Facebook Ads for Shopify: Setup, Tracking and Scale in 2026: connecting Shopify and structuring campaigns so each ad set learns