It's okay to call BS on media benchmarks.
A LinkedIn post claimed +40% CPM increases on Meta, citing a reputable report. I read the report and remain unconvinced. There were a few things the report never addressed.
Why You Should Question Media Benchmarks
Missing Context
How did CPMs vary by objective? Was the mix across awareness and sales goals consistent year over year? Did CPMs change within objectives?
Did the brands push harder into smaller but more expensive audiences? CPMs reflect spend vs. audience size. Deep funnel optimization shrinks audiences by focusing on a more valuable subset of people, raising CPMs.
How does the spend in the report reflect the Meta ad market at large? The sample is likely heavy spenders using an expensive measurement partner.
Beyond CPMs
The same logic applies to CTR, CPC, or CPA benchmarks. And especially cross-industry AOV or ROAS comparisons.
The Red Flag
The biggest red flag however was that if CPMs did really rise 40% year over year, that could only be explainable by either a correspondingly enormous collapse in ad inventory (and usage) or a correspondingly enormous surge in ad spend.
According to Emarketer, revenue per user is up a lot last year but it's up only half as much as the benchmark report said. Their sample was simply not representative.
The Takeaway
If an org wanted to write a real benchmark report with useful applicability that controls for all these variables, it would be very long.
Media benchmarks are only as good as their methodology. Before you let a headline change your advertising strategy, dig into the details. Question the sample size, the audience mix, and whether the data actually reflects your situation.
Next time you see a dramatic benchmark claim, read the full report before reacting.