Helpful?

What is a good website RPM in 2026?

Q: Which option pays the most in the website rpm benchmarks by niche (us traffic, display ads, 2026) table?

Finance / insurance / legal / B2B, at $20–35 ($50+). That row represents the strongest case in this dataset, so use it as an upper bound rather than an expectation.

Q: What is a realistic low-end figure?

Entertainment / memes at $3–6 ($8). Plan your costs so the low end still works, then treat anything above it as upside.

Q: Why do the numbers vary so much?

The spread between the highest and lowest row is about 12×. Revenue concentration and audience size versus audience intent explain most of that gap — see the drivers section above for the full list.

Q: Where do these numbers come from?

Benchmarks aggregate publisher-reported RPMs across major ad networks for 2025–2026, US-weighted session geography. RPM here means total ad revenue per 1,000 pageviews across all units — the network-agnostic definition.

Q: How can I estimate my own number instead of using a benchmark?

Use the Website Ad Revenue Calculator on RevenueLab — it takes your own inputs and returns a figure specific to your setup, which is always more accurate than a published range.

Short answer

A good website RPM in 2026 is $10–15 for general content with mostly US traffic. Anything above $20 is strong; finance, insurance, and B2B sites on premium ad networks regularly clear $30–50 RPM.

Website RPM benchmarks by niche (US traffic, display ads, 2026)

Niche	Typical RPM	Top quartile
Entertainment / memes	$3–6	$8
News / general lifestyle	$6–10	$14
Food / recipes	$10–16	$22
Tech / software / education	$12–18	$25
Travel	$12–20	$28
Finance / insurance / legal / B2B	$20–35	$50+

How to read this table

Finance / insurance / legal / B2B sits at the top of the table ($20–35) — $50+. If your situation looks like this row, plan against the upper half of the range rather than the midpoint.
Entertainment / memes anchors the bottom ($3–6) — $8. Treat this as the conservative case you should still be profitable at.
The gap between the top and bottom row is roughly 12×. That spread is why a single blended average is close to useless here — pick the row that matches your setup instead of averaging the column.
Most rows are ranges, not single figures. The low end usually reflects a weaker month, a softer audience geography, or an unoptimised setup; the high end reflects a well-run, well-targeted operation of the same size.
With 6 reference points in the "website rpm benchmarks by niche (us traffic, display ads, 2026)" table, the fastest way to use this page is to find the closest row, take its typical rpm, then stress-test it ±30% before you build a plan on it.

Context

Three levers move RPM more than anything else: geography (US traffic pays 3–10× tier-3), ad network (AdSense < Ezoic < Mediavine/Raptive), and content intent (a 'best business credit card' reader is worth 10× a 'funny cat pictures' reader to advertisers). Seasonality is real too — Q4 RPMs run 30–60% above Q1 as advertiser budgets flood in and then reset.

What moves this number

Revenue concentration

Most creator income is concentrated in one or two streams. The healthy target is no single source above roughly half of total revenue.

Audience size versus audience intent

Buying intent beats raw reach. Small, specific audiences convert to paid products at rates large general audiences never reach.

Platform take rates

Platform cuts, payment processing, and taxes typically remove 30–50% of gross before anything reaches your account.

Time cost

The right comparison is revenue per hour of production, not revenue per post — many high-revenue formats lose on that basis.

Methodology

Benchmarks aggregate publisher-reported RPMs across major ad networks for 2025–2026, US-weighted session geography. RPM here means total ad revenue per 1,000 pageviews across all units — the network-agnostic definition.

Assumptions and caveats

Figures are gross before platform fees, processing, and tax.
Medians hide a long tail — the distribution is heavily skewed toward the top decile.
This page was last reviewed on 2026-08-29. Ranges are updated as new data lands, so re-check before using them in a contract or a plan.
Use these numbers as a starting range, not a guarantee — your own historical data always beats a benchmark.

Frequently asked questions