YouTube RPM vs CPM — what each one actually means

Two different numbers, both per 1,000

CPM (cost per mille) is the advertiser-side price — what a brand pays YouTube for 1,000 ad impressions. RPM (revenue per mille) is the creator-side price — what the channel earns per 1,000 views, after YouTube's cut and across all monetized + non-monetized views.

The base relationship: YouTube takes 45% on standard pre-roll and mid-roll long-form ads. RPM ends up roughly 55% of CPM on long-form. But RPM also averages across non-monetized views, so the math isn't a clean 55% — it's 55% × (monetized share). Influencer Marketing Hub and ViDIQ both publish the same baseline formula in their 2026 breakdowns.

A worked example: 1,000 views on a finance video with a $25 CPM and 70% monetized share. Gross ad revenue = $25 × 70% = $17.50. YouTube's 45% cut leaves $9.63 — that's the long-form RPM the creator sees on the Analytics → Revenue tab.

Three things move RPM that have nothing to do with CPM: ad-fill (whether an ad actually serves), ad-block penetration in the audience, and the monetization status of the topic (firearms, current events, and certain health categories are systematically demonetized).

Open preview

Shorts pay differently — and it's not a small gap

Shorts revenue is pooled and split using a separate formula — not the standard ads-on-this-video model. Effective Shorts RPM runs $0.05–$0.09 in most niches in 2026, with strong niches occasionally reaching $0.15, per Influencer Marketing Hub benchmarks.

Mechanics: YouTube collects ad revenue from the Shorts feed as a whole, sets aside a slice for music licensing, then distributes the rest to creators based on share of total Shorts views. The creator then receives 45% of their allocated pool. That stacked deduction is what compresses Shorts RPM relative to long-form.

The gap is roughly 30–100x. A finance channel earning $15 RPM on long-form might earn $0.08 RPM on Shorts of the same content. A million Shorts views can be worth less than 100K long-form views in the same niche — a number creators don't internalize until they try to underwrite the channel.

Channels heavy on Shorts will see a much lower blended RPM than the long-form CPM their niche implies. Any honest valuation has to separate the two revenue streams — long-form RPM × long-form views, plus Shorts RPM × Shorts views — never a blended view multiplier.

Open preview

Geography flips the same channel sideways

RPM tracks CPM, and CPM tracks audience country composition more tightly than most creators expect. The same niche can have a 5–10x RPM spread between two creators based purely on where their viewers live.

Tier-1 country audiences (US, UK, Canada, Australia, Germany) clear $15–$40 CPM. Tier-2 (Brazil, Mexico, Japan, Turkey) clears $3–$15. Tier-3 (India, Indonesia, Bangladesh) clears under $3. The US specifically sits around $36 average CPM; India around $0.77 — a 47x spread, per upGrowth's 2026 country breakdown.

Practical read: when RPM drops and the niche didn't change, check the audience-country mix first. A wave of new Tier-3 viewers will drag blended RPM down even if the Tier-1 base is unchanged. That's not algorithmic punishment — it's the same CPM math, applied to a different audience mix.

Which one buyers underwrite

Buyers underwrite trailing twelve-month earnings × a niche-appropriate multiple. Trailing earnings come from RPM × monetized views, plus the rest of the revenue stack (memberships, Super Chat, brand deals). CPM is the input that produces RPM; it doesn't appear on a deal P&L.

Two channels with the same view count and the same CPM can have different RPMs because their monetized-view share differs. A finance channel running mid-roll ads on every video monetizes a higher share than a vlog with longer videos and only pre-roll inserts. Same CPM, different RPM.

When a broker asks for RPM, they're asking the right question. When a creator quotes CPM in negotiation, the buyer mentally divides by two and applies a discount for ad-fill — so leading with RPM and showing the Studio screenshot is faster and worth more.

Sponsorship CPM is the wild card. A creator can clear $40–$80 effective CPM on a direct brand deal, but buyers split that out because it's creator-dependent revenue (it walks with the creator if the channel sells). Ad RPM, by contrast, is asset-dependent — it stays with the channel under new ownership.

Before a listing or a private sale, pull twelve months of monthly RPM from Studio, sort it by long-form vs Shorts, and chart the audience-country split for each month. That three-tab pack is what a serious buyer's diligence asks for first — and it almost always moves the offer up because it removes guesswork.

Cross-checking the numbers Studio gives you

Studio's numbers are accurate but easy to misread. A short checklist before you quote your RPM to anyone.

Pick the right date range. Trailing 365 days is what buyers underwrite. Single-month snapshots can be 2–3x off the annual average — especially in finance (Q1 tax season) and gaming (Q4 launches).

Split long-form from Shorts explicitly. Studio's Revenue tab lets you filter. A blended top-line RPM that includes Shorts will under-represent your long-form rate and confuse a buyer.

Pull CPM as a separate column. If RPM is dropping but CPM is steady, the issue is ad-fill or monetized-view share — fixable. If both are falling together, it's an audience or content-mix shift — harder to reverse before a sale.

Screenshot the geo report. Buyers want to see the country mix that drives RPM. A clean Tier-1 majority is a discount-saver; a Tier-3 majority is something to disclose up front rather than discover during diligence.

Worked example — the same channel, two readings

Watching the two metrics diverge on a single channel is the fastest way to internalize why buyers only underwrite RPM.

The channel. A tech-review channel reports a $22 CPM in YouTube Studio and 1,000,000 monthly long-form views. A creator who confuses the two would multiply 1,000 × $22 = $22,000 and call that the monthly ad revenue. That number is wrong by roughly half.

The RPM reality. Monetized-view share is 65% (the rest are ad-blocked, sub-5-second, or in non-monetizable contexts). Gross ad revenue = 1,000,000 × 65% ÷ 1000 × $22 = $14,300. YouTube's 45% cut leaves the creator $7,865 — an effective long-form RPM of $7.87, not the $22 the CPM implied.

Add Shorts. The same channel also pulls 2,000,000 Shorts views at a $0.07 Shorts RPM = $140. So the blended top-line RPM across all 3M views is ($7,865 + $140) ÷ 3,000 = $2.67 — which is the number a careless calculator would surface, badly under-representing the long-form business and over-representing the Shorts contribution.

Why this matters at sale. A buyer who sees the $22 CPM and the $2.67 blended RPM needs the split to underwrite correctly: $7,865/mo of underwriteable long-form ad revenue is the real asset; the Shorts line is rounding error. The seller who presents the long-form RPM, Shorts RPM, and monetized-view share separately gets underwritten at the long-form rate. The one who hands over a blended number gets discounted to it.