Facebook Ads Benchmarks 2026: CPM, CTR, Cost per Purchase
Every public Facebook ads benchmark for ecommerce we could open and check, in one place: CPM by quarter, click-through rate by industry, cost per purchase by month. Plus the one number on your own store that decides whether any of them matter.
"Is a $20 cost per purchase even possible?"
A dropshipper asked exactly that in the Shopify Community. Product priced around $37. Cost per purchase somewhere between $37.57 and $80.10. Losing about $10 on every order.
The question underneath it is the one every founder asks when the ad account turns red: are my ad costs normal, or am I doing something wrong? Here are the Facebook ads benchmarks ecommerce brands should compare against in 2026, every number pulled from a public report I opened and checked. Median ecommerce CPM: about $13.88 (Superads, July 2025 to July 2026). Click-through rate: about 2% across three separate datasets. Cost per purchase: about $34.34 (Superads ecommerce median).
And the part nobody puts in the benchmark table: your cost per purchase is decided twice. Once in Meta's auction. Once on your product page.
The auction sets the price of the click. Your page sets the price of the customer.
Benchmarks tell you what the click costs everyone. Only your page decides what the customer costs you.
What are the Facebook ads benchmarks for ecommerce in 2026?
Short answer first, then the detail.
Most public datasets put an ecommerce Facebook ad at a CPM between $13 and $15 outside the holidays, a click-through rate around 2%, a cost per click between $0.60 and $0.70, and a cost per purchase in the mid $30s. Q4 pushes all of it up. January pulls it back down.
Here's every headline number side by side, with where it came from:
| Metric | Superads, ecommerce (Jul 2025 to Jul 2026) | Triple Whale, full-year 2025 (via Rule1) | LocaliQ 2026, traffic campaigns (all industries) |
|---|---|---|---|
| CPM (cost per 1,000 impressions) | about $13.88 | $14.19, up 20% year over year | not reported |
| Click-through rate | about 2.05% | 2.19%, up 13.5% | 1.93%, up from 1.71% |
| Cost per click | about $0.69 | not reported here | $0.60, down from $0.70 |
| Cost per purchase | about $34.34 | not reported here | not reported |
| Conversion rate | not reported | 1.60%, up 8.3% | not reported |
| Return on ad spend | not reported | 1.86, up 1.3% | not reported |
Sources: Superads ecommerce CPM, CTR, CPC and cost per purchase pages; Triple Whale's 2025 data on about 35,000 ecommerce brands as compiled in Rule1's Facebook ads benchmarks roundup; LocaliQ's Facebook advertising benchmarks for 2026, last updated September 23, 2026.
A note on how I built this. I only kept numbers I could open on the publisher's page this week. Triple Whale's own report blocked my access, so its figures appear here exactly as Rule1 compiled them, and I say so every time. WordStream's page blocked me too, which is why its partner LocaliQ's 2026 edition is the one cited. If a number you've seen elsewhere isn't in this post, that's why.
Each of these datasets measures something slightly different. Superads reports medians across thousands of ad accounts and more than $3B in ad spend. Triple Whale reports medians across its ecommerce customers. LocaliQ reports traffic and leads campaigns across 20 industries, most of them service businesses. So the right way to use this table is to pick the column that looks most like you, then compare against that column only.
A benchmark from a blend of 20 industries won't tell you much about a $64 tub of protein powder. Pick the dataset that looks like your store, and ignore the rest.
What is a normal Facebook CPM in 2026, quarter by quarter?
CPM is what you pay for 1,000 impressions. It's the price tag on attention, and it's the one number you control the least.
Rule1 published a 13-month table of global median CPM ending January 2026. I grouped the months into quarters and averaged them. Q1 2025 is left out because the table only includes March.
Quarter averages of the monthly global medians in Rule1's compiled table (March 2025 to January 2026). Jan 2026 is a single month.
Here's the math so you can check it. Q2: $18.57, $19.79 and $19.67 average to $19.34. Q3: $19.58, $20.38 and $19.96 average to $19.97. Q4: $21.69, $25.22 and $22.04 average to $22.98. Then January 2026 lands at $15.74, a 38% drop from November's $25.22 peak.
That's the all-industry line. Ecommerce runs cheaper and swings harder.
Superads' ecommerce CPM data puts the median at about $13.88 across July 2025 to July 2026, roughly 33% below its global baseline of about $20.59. Ecommerce started the window at $14.88 in July 2025, bottomed out at $11.21 in September 2025, spiked to $21.96 in November 2025, and finished at $12.90 in July 2026. The October to November jump was 65%. The November to December correction was 42%. April to June 2026 sat in a tight band between $13.25 and $13.70.
So if your CPM is $13 in May, you're normal. If it's $22 in November, you're also normal.
Sale days are their own animal. Gupta Media's CPM tracker measured a $16.85 CPM on Black Friday 2024 and $17.70 on Cyber Monday 2024, which it says was 138% above that year's average. Different dataset, lower absolute numbers, same direction. The week everyone wants to sell is the week everyone pays the most to be seen.
CPM also depends on what you sell. In the Triple Whale data compiled by Rule1, Health and Wellness had the highest median ecommerce CPM at $20.70 and Automotive had the lowest at $10.01, both as of January 2026. Same platform. Same month. Double the price of attention.
If Q4 is what's spooking you, I broke down the holiday auction in why the Q4 CPM increase happens and how to plan for it. The short version: buy customers in October, sell to them in November.
What click-through rate should an ecommerce ad get?
Around 2%. Three datasets, three slightly different answers, one neighborhood.
Superads' ecommerce click-through rate averaged about 2.05% across its 13-month window, basically level with its all-industry baseline of about 2.04%. Triple Whale's 2025 ecommerce median, per Rule1, was 2.19%. LocaliQ's 2026 traffic-campaign average across all industries was 1.93%.
But the ecommerce line moved a lot more than the average suggests. Here are the months Superads called out:
Superads ecommerce medians, July 2025 to July 2026. Only the months named in Superads' published analysis are shown.
June 2026 at 1.22% to July 2026 at 2.93%. That's a 140% swing in one month, per Superads. Nobody's creative got 140% better in 30 days. Market-wide numbers move for market-wide reasons.
Which is why I tell founders not to panic over a single month's click-through rate. Watch the three-month trend against your own history.
Now the industry view. This is the Triple Whale 2025 ecommerce data as Rule1 compiled it, with year-over-year change in brackets:
| Ecommerce industry | Click-through rate | Conversion rate | Return on ad spend |
|---|---|---|---|
| Health & Wellness | 2.70% (+22.8%) | 1.72% (+5.0%) | 1.50 (-2.8%) |
| Books & Music | 2.34% (+25.5%) | 1.72% (+17.5%) | 1.65 (+2.8%) |
| Lifestyle & Boutique | 2.28% (+10.2%) | 1.74% (+7.6%) | 1.93 (+2.7%) |
| Beauty | 2.27% (+17.8%) | 1.94% (-1.1%) | 1.57 (-1.1%) |
| Apparel & Accessories | 2.25% (+15.7%) | 1.46% (+9.4%) | 2.18 (+3.9%) |
| Home & Garden | 2.22% (+9.2%) | 1.32% (+18.5%) | 2.18 (+7.0%) |
| Automotive | 2.22% (+14.1%) | 1.30% (+15.8%) | 2.54 (+1.7%) |
| Media & Publishing | 2.21% (+24.2%) | 1.56% (+21.0%) | 1.17 (-2.2%) |
| Toys, Art & Collectibles | 2.19% (+8.0%) | 1.52% (+14.9%) | 1.93 (+2.7%) |
| Electronics | 2.19% (+8.0%) | 1.20% (-1.6%) | 1.92 (+1.5%) |
| Travel Accessories | 2.19% (+17.1%) | 1.29% (-16.7%) | 2.25 (-0.8%) |
| Pets & Animals | 2.13% (+11.0%) | 1.80% (+11.1%) | 1.58 (+7.1%) |
| Sports & Outdoors | 1.91% (+13.1%) | 1.28% (+11.0%) | 2.28 (+3.8%) |
| Baby | 1.91% (+12.7%) | 1.85% (+0.9%) | 2.17 (+1.6%) |
| Food & Beverage | 1.85% (+5.5%) | 2.02% (-0.1%) | 1.56 (+7.2%) |
Look at Health & Wellness. Highest click-through rate in the table at 2.70%. Highest CPM at $20.70. And one of the lowest returns on ad spend at 1.50.
The ad wins the scroll. Then something downstream loses the sale.
Now Automotive. Middle-of-the-pack click-through rate, a 1.30% conversion rate, and the best return on ad spend in the table at 2.54. Cheap attention ($10.01 CPM) carried it. Click-through rate on its own predicts almost nothing about profit.
For traffic campaigns, LocaliQ's 2026 report breaks out the categories closest to a Shopify store:
| LocaliQ category (traffic campaigns) | Click-through rate | Cost per click |
|---|---|---|
| Beauty & Personal Care | 2.73% | $0.50 |
| Furniture | 1.90% | $0.64 |
| Shopping, Collectibles & Gifts | 1.88% | $0.59 |
| Health & Fitness | 1.86% | $0.64 |
| Home & Home Improvement | 1.76% | $0.63 |
| Apparel/Fashion & Jewelry | 1.51% | $0.71 |
| All industries | 1.93% | $0.60 |
A high click-through rate means the ad made a promise people wanted. Your conversion rate tells you whether the page kept it.
What is the average cost per purchase on Facebook?
About $34 in ecommerce, if you take Superads' cost per purchase data at face value. Its ecommerce median averaged about $34.34 per purchase from July 2025 to July 2026, with the all-industry baseline at about $47.55. Ecommerce ran roughly 28% cheaper.
But look at the path it took:
Superads ecommerce medians. Months shown are the ones named in Superads' published analysis; values marked approximate there are rounded to the cent.
$55.75 in June 2026. $16.51 a month later. Same platform, same category, a $39 swing.
That tells you two things. First, a single month of cost per purchase is noise. Second, and this matters more, cost per purchase is a stacked number. It sits on top of CPM, click-through rate and your page's conversion rate. Move any one of them and it moves.
Here's how they stack. This is the formula I'd tattoo on every media buyer's forearm if they'd let me:
Cost per purchase = CPM ÷ (1,000 × click-through rate) ÷ conversion rate. The first half is the price of a click. The second half is your page.
Run Triple Whale's 2025 ecommerce medians (via Rule1) through it. CPM $14.19. Click-through rate 2.19%. That's $14.19 ÷ 21.9 clicks per 1,000 impressions, so a click costs about $0.648. Call it $0.65, which sits right between Superads' $0.69 and LocaliQ's $0.60.
Now divide by the 1.60% median conversion rate. $0.65 ÷ 0.016 = $40.63 per purchase.
Medians don't multiply perfectly (the median store on CPM isn't the median store on conversion), so that won't match any single report to the cent. But $40.63 lands in the same neighborhood as Superads' $34.34. The formula holds.
And that dropshipper from the Shopify Community? A $20 cost per purchase at a $0.65 click needs a 3.25% conversion rate ($0.65 ÷ 0.0325 = $20.00). Double the 1.60% median. Possible, but not from the ad account.
Why do my numbers look nothing like the benchmarks?
I get this message a lot. Founder pulls up a benchmark, pulls up Ads Manager, and the two don't even look like the same sport.
Usually it's one of these six things.
1. Median vs average. Superads explains on its pages that it uses medians because a small share of campaigns with extreme values would inflate an average. If you compare your account's average to someone's median, you'll look worse than you are.
2. All clicks vs link clicks. Superads' own definition notes that Facebook lists Clicks, Link Clicks and Outbound Clicks separately, and "Clicks" includes things like profile and comment clicks. Your link click-through rate will always look lower than an all-clicks benchmark.
3. Objective mix. LocaliQ's numbers are traffic and leads campaigns. Most Shopify ad accounts spend on sales campaigns, which usually carry higher CPMs because Meta is hunting buyers, not clickers. Superads says as much: conversion campaigns usually have higher CPMs than traffic campaigns.
4. Geography. A US-only account pays more than a global blend. Superads calls the US one of the more expensive markets.
5. Price point. A $240 weighted blanket will never hit the cost per purchase of a $19 phone case, and it shouldn't. Superads makes the same point: higher-priced products typically carry a higher cost per purchase. What matters is whether each product's profit per product can carry its own cost per purchase.
6. Attribution. What Ads Manager counts as a purchase depends on your attribution window and your tracking setup. A store with a broken pixel will show a terrible cost per purchase on great ads.
Here's the thing. None of those six are your fault, and none of them are a reason to change your ads. They're reasons to compare carefully.
Before you fire your media buyer over a benchmark, make sure you're comparing medians to medians and link clicks to link clicks.
What do these benchmarks mean for your product page?
Everything. And this is the part every benchmark report skips.
Look back at the formula. Three numbers decide your cost per purchase. Meta's auction controls the first one. Your creative influences the second. Your page owns the third.
Watch what happens when you change only the page.
Run the math on a store like this, a hypothetical: a ceramic aroma diffuser brand with an $85 average order value, paying the benchmark $0.65 per click, buying 10,000 clicks a month. That's $6,500 in ad spend.
At the 1.60% median conversion rate, the page turns 10,000 visitors into 160 orders. Conversion rate 1.6%, average order value $85, so revenue per visitor is $1.36. On 10,000 visitors that's $13,600 in revenue. Cost per purchase: $6,500 ÷ 160 = $40.63. Return on ad spend: 2.09.
Now the page converts at 3.2%. Same ads. Same audience. Same $6,500.
Conversion rate 3.2%, average order value $85, so revenue per visitor is $2.72. On the same 10,000 visitors that's $27,200. That's 320 orders, so cost per purchase falls to $20.31. Return on ad spend: 4.18.
10,000 clicks at $0.65. Only the page's conversion rate changes.
Same auction. Half the cost per customer.
Now push it further. Say November hits and your CPM climbs to the $25.22 peak from Rule1's table while your click-through rate holds at 2.19%. A click now costs $1.15 ($25.22 ÷ 21.9). On the 1.6% page, cost per purchase is $71.88. On the 3.2% page, it's $35.94.
Compare that to the 1.6% page in January, when CPM falls to $15.74 and a click costs $0.72. Cost per purchase: $45.00.
So the page converting at 3.2% buys customers more cheaply in the most expensive month of the year than the 1.6% page does in the cheapest month. Let that sit for a second.
Try your own numbers here. The defaults match the diffuser example:
This is the work we do every day, so let me show you what it looks like on a store that isn't hypothetical. A bedding brand came to us with a conversion rate of 1.0% and an average order value of $125, so revenue per visitor was $1.25. On 10,000 visitors, that's $12,500. After the page rebuild: conversion rate 3.5%, average order value $231, revenue per visitor $8.10. On the same 10,000 visitors, that's $81,000. See the full case study numbers. Real client numbers, not typical results, and not a promise of what your store will do.
Now run their conversion rates through the benchmark $0.65 click. To be clear, this is an illustration, not their ad account. At 1.0%, a purchase costs $65.00. At 3.5%, it costs $18.57. Their ad costs didn't need to get any better. The page did the heavy lifting.
The auction is rented. The page is owned. Every dollar you save on the page, you keep in every month, at every CPM.
How is AI changing what these benchmarks mean?
Here's the honest version, without the hype.
Meta's ad delivery is now mostly a machine decision. Meta's Advantage+ campaign documentation describes campaigns that automate audience expansion, placements across every available slot and campaign-level budget with automated bidding. It also says that from version 25.0 of its Marketing API, the old Advantage+ shopping campaign endpoint can no longer create new campaigns. Sales, app and leads all run under one unified Advantage+ structure now.
Under the hood, Meta's engineering team says its Andromeda retrieval system narrows tens of millions of candidate ads down to a few thousand relevant ones for each person. Meta reported a 6% recall improvement and an 8% ads quality improvement on selected segments.
What that means for benchmarks:
- Targeting stopped being your lever. When the machine picks the audience, two stores in the same category are bidding on the same people with the same tools. CPM converges. Your edge isn't there anymore.
- Creative became the targeting. The ad itself tells the system who to show it to. That's why click-through rate keeps climbing year over year in the Triple Whale data (+13.5% in 2025, per Rule1). I wrote the full playbook in the 2026 Facebook ads creative testing framework, and the Andromeda side in what Meta's Andromeda update changed for ecommerce ads.
- The page is the last manual lever. Meta's AI can find the buyer and pay a fair price for the click. It can't make your product page answer "why is this $85 and not $30?" That's yours.
I'll admit something. I used to believe the next audience test was the one that would fix an account. Most of those tests were noise, and the machine now does that job better than any of us did by hand. What it can't do is fix a page that loses 98 out of 100 people it sends.
And here's the twist. The same AI wave that took targeting off your plate makes the page faster to fix. A product page that used to take a copywriter, a designer and three weeks can now be rebuilt from the product URL in less than 15 minutes. That's what RevenueFlows AI does, and it's why I care more about your revenue per visitor than your CPM.
If you're running Advantage+ already, the Advantage+ sales campaigns setup guide for Shopify brands walks through the settings.
How do you benchmark your own store against these numbers?
Benchmarks are a map. Your account is the territory. Here's the method I use on every profit audit, and you can run it yourself in about an hour.
Step 1. Pick one comparison source. Ecommerce-only and median-based beats everything else. From this post, that's Superads or Triple Whale (via Rule1). Don't mix columns.
Step 2. Pull 90 days, not 30. Ecommerce click-through rate swung 140% in a single month in Superads' data. One month tells you nothing. Use the last full quarter and the same quarter last year if you have it.
Step 3. Match the definitions. In Ads Manager, use CPM, CTR (link click-through rate) and cost per purchase. Note your attribution setting. Compare medians to medians where you can.
Step 4. Split the funnel into its three numbers. Write down CPM, link click-through rate, and the conversion rate of the sessions those clicks produce on your store. Recompute your cost per purchase from them: CPM ÷ (1,000 × click-through rate) ÷ conversion rate. If it doesn't roughly match Ads Manager, your tracking has a gap. Fix that first.
Step 5. Find the outlier. Compare each of the three to your chosen benchmark. CPM 40% above benchmark outside Q4? Look at geography and audience size. Click-through rate 40% below? Creative. Conversion rate below 1.6%? The page.
Step 6. Price the gap in dollars. For each outlier, run the formula with the benchmark number swapped in. Whichever swap cuts cost per purchase the most is where your next month goes. Most of the time it's the page, because conversion rate sits in the denominator: double it and cost per purchase halves.
Step 7. Check it against your break-even. A "good" cost per purchase is one that leaves profit. Take your average order value times your gross margin. That's your break-even cost per purchase. The break-even return on ad spend calculator does it for you.
For the hypothetical diffuser brand, an $85 order at a 60% gross margin leaves $51 to spend on acquiring the customer. At the benchmark $40.63 cost per purchase, that's $10.37 of profit per first order. At $20.31, it's $30.69. Same product. Same ads. Three times the profit per sale.
What should you watch for in Q4 2026?
We're in October. Every dataset in this post says the same thing about the next eight weeks: CPM climbs into November, then falls hard in January.
Superads' ecommerce CPM went up 65% from October to November 2025. Rule1's global median peaked at $25.22 in November 2025. If the pattern repeats, your cost per click in late November could be close to double what it was in September. Plan for it in the spreadsheet, not in a panic in week 47.
There's a playbook for this in the Black Friday Facebook ads strategy for Shopify, and the full channel picture in the ecommerce traffic strategy guide. If you want the wider store benchmarks next to these ad numbers, the 2026 DTC benchmarks roundup has them.
But here's the math that matters most for Q4. When the click doubles in price, the page's conversion rate decides whether the order still clears break-even. A 1.6% page at a $1.15 November click costs $71.88 per purchase. That's underwater for most $85 products. A 3.2% page at the same click costs $35.94. Still profitable.
You can't control the auction in November. You can control the page in October.
What to do next
Open Ads Manager. Pull the last 90 days. Write three numbers on a sticky note: CPM, link click-through rate, and the conversion rate of the sessions those clicks produce on your store. Then run the formula from this post and see which of the three is furthest from its benchmark.
If it's the first two, work the ads. If it's the third, the ads were never the problem.
Book Your Profit Audit
On a profit audit we pull those three numbers from your account, show you exactly what each click is worth on your current page, and price the gap against these benchmarks in dollars. Then we'll show you how to rebuild a high-converting product sales page in less than 15 minutes.
Or go here to check it out → revenueflows.ai
P.S. Meta sets the price of the click. Your page sets the price of the customer. Only one of those is yours to fix.
Frequently asked questions
What is a good CPM for Facebook ads in 2026 for ecommerce?
Superads puts the median ecommerce CPM at about $13.88 across July 2025 to July 2026, with a low of $11.21 in September 2025 and a high of $21.96 in November 2025. Triple Whale's full-year 2025 median, as compiled by Rule1, was $14.19. If you're in that range outside of Q4, your CPM is normal and your problem is somewhere else.
What is the average click-through rate for Facebook ads in ecommerce?
Most ecommerce datasets land around 2%. Superads shows a 2.05% average for ecommerce over 13 months, Triple Whale's 2025 median was 2.19%, and LocaliQ's 2026 report puts traffic campaigns across all industries at 1.93%. Check whether your number is link click-through rate or all clicks before you compare.
What is the average cost per purchase on Facebook ads?
Superads' ecommerce median averaged about $34.34 per purchase from July 2025 to July 2026, roughly 28% below its all-industry baseline of $47.55. But your cost per purchase is your cost per click divided by your page's conversion rate, so the same ads can cost $40 a sale on one page and $20 on another.
Why are Facebook ads benchmarks so different from source to source?
Different samples, different math. Some report medians, some averages. Some count all clicks, some only link clicks. Some only include traffic campaigns, others mix in sales campaigns. Compare your store to one source at a time, and to the same metric definition, never to a blend.
When are Facebook ads cheapest for ecommerce?
January, in most public data. Rule1's compiled global median CPM fell from $25.22 in November 2025 to $15.74 in January 2026, a 38% drop. Superads' ecommerce CPM also showed its deepest early-year dip in March 2026.