The Cheap CPM Assumption Fails At Scale

Maybe it’s time we have a chat. Shall we?

Let’s talk about the efficiency of reach (aka CPM) and where this efficiency actually happens across 2,668 YouTube creator brand sponsorships.

And most importantly, what does that mean for your 2027 creator budgets?

EXECUTIVE SUMMARY

Efficient Reach Doesn’t Hide At The Small End.

It Scales.

The default logic of creator brand partnership buying says smaller channels deliver cheaper reach: lower rates, hungrier creators, more efficient CPMs. Our data says the opposite.

Of the 120 creator-brand partnerships that have crossed 1 million views, 60 (that’s half for all you math majors) sit in the top 10% of CPM efficiency (score ≥ 84.7 on our 0–100 index). To hold this down for the general studies majors, if scale and efficiency were unrelated, roughly 10% of the brand partnerships with 1mm+ views would be in the top decile (top 10%) of performance, but we see that 50% of the 1mm+ views channels land there. The concentration is 5× chance, which means “it ain’t chance”.

The pattern holds true across the entirety of the analysis as well: the correlation between a creator-brand partnership video views (log scale) and its CPM efficiency score is 0.70, whereas the median top-decile placement delivered 399,737 views against 19,139 for the median bottom-decile placement, a 21× gap.

Reach efficiency doesn't hide in the "cheap" end of the market. It concentrates at the top, which means that larger creators are more efficient. This is why you buy the jumbo-pack of toilet paper rolls from Costco: it’s more efficient. This paper (not to be confused with the tp you buy at Costco) is for CMOs and influencer marketing managers planning 2027.

50%

of million-view placements sit in the top decile of cpm efficiency (60 of 120)

r = 0.70

correlation between views (log) and cpm efficiency across 2,668 placements

5x

the concentration chance would produce -- a random cohort would put ~10% in the top decile

21x

median views, top cpm decile vs bottom decile (399,737 vs 19,139)

Analysis set: 2,668 YouTube sponsored placements live March 2024 – July 2026 from The Outloud Group's sponsorships, each with views-to-date and a 0–100 CPM efficiency score. All placements anonymized.

SECTION 01

The Cheap CPM Assumption

Every media planner has heard the argument: big creators are a premium play, so if you want efficiency, buy small. And the rate card supports this: smaller channels quote lower absolute prices, and a lower sticker price feels like a lower CPM waiting to happen.

But an effective CPM isn't made of price alone. It's price divided by delivered views and delivered views are where scaled placements run away from the field. A cheap buy that underdelivers is an expensive buy wearing a small invoice. Yes, it’s kinda like buying a 4-pack of toilet paper: low cost but a bad ratio of wipes to dollar paid.

We tested the assumption across 2,668 YouTube sponsored placements from March 2024 through July 2026, each scored 0–100 on CPM efficiency. In other words, how efficiently a piece of content was bought in comparison to our full book of YouTube creator brand partnerships.

the finding, in one sentence

The cheap CPM assumption fails at scale. Reach efficiency concentrates with the biggest creators, not with the smallest.


SECTION 02

Scaled Placements Own The Frontier

Put every placement on two axes, views to date (log scale) and CPM efficiency score, and the mass collects in the upper right. The million-view cohort posts a median CPM score of 84.9 against a book-wide median of 49.5: the median million-view placement sits at the doorstep of the top decile. Sub-million placements? Just 8.1% of them crack it.

Horizontal: views to date (log scale) · Vertical: CPM efficiency score

Under 1M views (N=2,548) 1M+ views (N=120) Top-decile CPM (N=60)

If reach efficiency lived at the small end of the market, the yellow would pool on the left. Instead, half of all placements that has crossed a million views bought its reach in the top decile of the entire book -- and the correlation between scale and efficiency runs 0.70 across all 2,668 placements.


SECTION 03 / 04

The Pattern Holds WITH LARGE & SMALL CREATORS

Read from the efficiency side: it repeats. Of the 30 most reach-efficient placements (read: low CPM), 11 have already crossed 1 million views. This shows that scaled content is overrepresented when it comes to low CPMs.

And the other end makes the point louder. The 30 placements with the highest CPM scores (high CPM) have a median of 9,105 views to date, a floor of 1,707, and not one within sight of a million. When a placement buys reach expensively, it is almost always a small placement.

11/30

top-30 reach-efficient placements already past 1M views

8.1%

sub-million placements that reach the top CPM decile

9,105

median views of the 30 least reach-efficient placements

1,707

views floor in the bottom-30 CPM cohort -- none near 1M

Inefficiency doesn't concentrate from partnering with large creators; it comes from partnering with smaller creators, exactly where the CPM thesis says to shop.


Sections 05 / 06

Why the Correlation Holds -- and Where to Point 2026 Spend

There are 3 reasons that support small creators being less efficient than large creators, but none establish causation on their own.

1) YouTube's algorithm compounds winners: content that earns watch time gets distributed, so views scale non-linearly while price is set upfront.

2) Audience fit is self-selecting: channels get big because audiences reliably show up and that reliability is precisely what a sponsor buys.

3) Sponsorship craft matures with scale, meaning creators who have run hundreds of integrations deliver more predictably.

The honest caveat: crossing a million views is partly an outcome of being efficient, not only a cause. But for a buyer the correlation is the planning-relevant fact: betting that YouTube channels that average fewer views per video will deliver cheap CPMs is betting against the cohort where efficiency actually concentrates.

01

Stop Treating Scale as a Premium Tax

Half the million-view cohort bought top-decile reach. Scale belongs in the efficiency case, not just the awareness case.

02

Screen on Efficiency, Not Raw Price

Raw price is a trap. Evaluate each placement against delivered CPM relative to all the spots you have run to determine efficiency.

03

Rebalance Test Budgets

If your test-and-learn pool defaults to smaller channels on a cheaper CPM thesis, your control group is holding the inefficiency.

04

Re-Score Quality

Views compound. A placement's efficiency ranking is not fixed at day 30 -- or day 90.

Big high-view ad placements are the most cost-effective way to reach people. Cheap ads might look like a deal, but they are usually inefficient on a per-dollar-spent basis.


Methods

2,689 YouTube sponsored placements live between March 1, 2024 and July 14, 2026 from The Outloud Group's sponsorships; 2,668 carry both views-to-date and a CPM efficiency score and form the analysis set. Placements are anonymized; no creator or channel is identified.

CPM efficiency is indexed 0–100 against the book (higher = cheaper delivered reach). Top-decile threshold: score ≥ 84.7 (n=267). Book-wide median score: 49.5.

Million-view cohort: n=120; 60 (50%) at or above the top-decile threshold; cohort median score 84.9. Sub-million placements reaching the top decile: 8.1%. Median views: top decile 399,737, bottom decile 19,139. Bottom-30 by CPM score: median 9,105 views, minimum 1,707. Top-30 by CPM score: 11 of 30 above 1M views.

Pearson r between log10(views-to-date) and CPM efficiency score across the analysis set = 0.695.

The Outloud Group has been matching brands with creators for over 18 years -- and we don’t stop at surface metrics. If you want your influencer program screened on both axes -- reach efficiency and conversion evidence -- let’s talk.

Next
Next

Cheap Reach, Expensive Conversions