Meta Ads Benchmarks 2026: CPM, CPC, CTR, and ROAS by Industry

This guide gives you real Meta benchmark ranges, including CPM, CPC, CTR, and ROAS numbers pulled straight from the DTC accounts we run, sets them next to the published industry figures, and shows you how to read the gap. By the end you will know what a healthy account looks like in 2026, why two "average CPC" numbers can be a dollar apart and both be right, and which lever to pull when one of your metrics is off.

What Are Good Meta Ads Benchmarks in 2026?

Meta ads benchmarks are the typical cost and performance ranges advertisers see on Facebook and Instagram: how much you pay per thousand impressions (CPM), per click (CPC), the share of people who click (CTR), the share who buy (conversion rate), and how much revenue each ad dollar returns (ROAS). They exist so you can tell a healthy account from a broken one at a glance.

Here is the honest version most benchmark posts skip. There is no single "Meta average" that applies to your account. A prospecting video campaign and a branded retargeting campaign in the same ad account can post CPMs that differ by 3x, and both are normal. So treat every number below as a reference band, not a scoreboard.

The most useful benchmark you have is your own account 90 days ago. If your CPM is up 40 percent but your CTR is up too and your cost per purchase held, you are fine. If your CPM is flat and your cost per purchase doubled, you have a real problem no industry table will diagnose.

Meta Ads Benchmarks: What We Actually See Across Our Accounts

We pulled Meta delivery data across five DTC and ecommerce accounts we manage for the 90 days ending September 2026. These are link-click metrics (Meta reports CTR and CPC on inline link clicks, not all clicks), aggregated and shown as ranges, no account named.

A few things stand out. First, CPM alone tells you almost nothing. Our lowest-CPM account also had one of the lowest CTRs, because it was buying cheap awareness reach at scale. Our highest-CTR account paid a higher CPM and still had efficient clicks, because the creative was doing the work.

Second, link CPC under a dollar is normal for DTC when the creative earns the click. If your CPC is climbing, the fix is almost always the ad, not the audience. Since the Andromeda update, Meta rewards creative that holds attention and quietly penalizes creative that does not, so the creative is now the targeting.

We are not publishing a single blended ROAS number from these accounts, and that is deliberate. Purchase tracking is configured differently across them, and only a couple have clean, sales-grade conversion data over this window. Publishing an average ROAS from mixed tracking would be a made-up number dressed as a benchmark, so we left it out. That gap is the whole point of the ROAS section below.

Meta CPM, CPC, and CTR Benchmarks by the Published Industry Data

Now set our accounts next to the widely cited industry figures. This is where benchmark posts get slippery, because the sources disagree, so we are showing you the disagreement instead of picking the number that flatters us. (For the search side of the ledger, see our companion ecommerce Google Ads benchmarks guide.)

Why is a published CPC (roughly $0.70 for traffic, $1.92 for leads) higher than the sub-$0.62 we see? Three reasons, and they matter for reading any benchmark:

• Industry mix: Platform-wide numbers blend expensive verticals like finance and insurance (LocaliQ puts their traffic CPC near $1.22) with cheaper ecommerce clicks (shopping and collectibles near $0.34). A DTC-only average sits toward the low end.

• Click definition: Some reports use "clicks (all)," which includes reactions and profile taps. Meta's link-click CPC, the one that maps to site traffic, is lower.

• Objective: Lead and traffic campaigns price differently than sales campaigns, and awareness cheapest of all.

By vertical, the published data shows the same shape we see in the field. In LocaliQ's 2025 traffic data, shopping, collectibles, and gifts posted the highest click-through rate at about 4.13 percent because the products are visual and impulse-friendly, apparel, fashion, and jewelry ran near 1.29 percent, and service categories like automotive repair sat lowest around 0.80 percent. Visual, impulse-friendly categories earn the click, while considered-purchase categories lean on trust and repeat buying instead.

What Is a Good ROAS on Meta Ads in 2026?

A good Meta ROAS in 2026 is the one that clears your break-even, and for most ecommerce brands that lands somewhere between 2x and 4x on purchase ROAS. Published averages back up that band: Trendtrack put the 2025 all-industry Facebook ROAS around 2.19x on average, while other 2025 datasets land the median closer to 1.93x (Visible Factors). The spread, again, comes from methodology and margin.

Here is the part that saves brands real money. Platform-reported ROAS is not the same as money in the bank. Meta counts view-through and click conversions inside its own attribution window, so its ROAS almost always reads higher than your blended, first-party reality. We have seen the gap run 30 to 50 percent on accounts where the attribution window was left at the default. That is why we manage to blended efficiency and incrementality, not the number on Meta's dashboard.

The benchmark that actually protects your margin is your own break-even ROAS, which you calculate from your contribution margin, not from any industry table. If your gross margin is 60 percent and you want to keep a slice for profit, your break-even ROAS might be 2.0x. A brand at 35 percent margin needs closer to 3.0x just to hold even. A 4x ROAS can still lose you money if your margins and returns are ugly, which is why we start every account with the math, not the benchmark.

How to Use Meta Benchmarks Without Getting Fooled

Benchmarks are a smoke alarm, not a thermostat. Use them to catch fires, then run your account off your own trend and margin. A few rules we hold to:

• Compare the same objective. Judge a sales campaign against sales benchmarks, not against a cheap awareness CPM you saw in a blog post.

• Fix tracking before you trust any number. After iOS and cookie loss, a clean Conversions API setup is the difference between a real ROAS and a fictional one. Bad tracking makes every benchmark comparison meaningless.

• Read metrics in pairs. CPM up with CTR up is fine. CPM flat with CPA up is a creative or landing-page problem. A single metric in isolation lies.

• Weight your own history over the industry. Your account last quarter is a better benchmark than a national average that has never seen your product, your margins, or your customer.

When a metric is off, the fix usually maps cleanly: a weak CTR is a creative problem, a high CPM with a healthy CTR is an auction or audience-saturation problem, and a strong CTR with a weak conversion rate is a landing-page or offer problem. Diagnose in that order and you will spend less time blaming the algorithm.

Frequently Asked Questions About Meta Ads Benchmarks

The Bottom Line on Meta Ads Benchmarks

Benchmarks are worth checking and dangerous to chase. The 2026 ranges are real: CPMs in the single-to-low-double digits, link CTRs around 1 to 2 percent, link CPCs under a dollar for most DTC brands, and purchase ROAS in the 2x to 4x band. But the account that wins is not the one that matches a national average. It is the one with clean tracking, creative that earns the click, and a team reading its own trend against its own margins every week.

If you want a read on whether your Meta account is actually healthy against your numbers, not a blog post's, we are happy to look.