It always starts the same way. A campaign wraps, and the team pulls up the performance report. The click-through rate? Above benchmark. CPM? 30% lower than the platform average. There’s a general sense of relief, maybe even celebration. Everyone’s nodding. Job well done. But then someone asks the one question that doesn’t show up on the dashboard: “Did it actually drive anything?”
That’s when things get murky.
Benchmarks: The Rental Tux That Sort of Fits
Benchmarks have become a go-to comfort metric in media analytics. They’re quick to reference, look clean in a deck, and offer a sense of relative performance. But relying on them too heavily is like judging how well a tuxedo fits based on how it looks on a mannequin. Sure, it might look good at first glance, but it wasn’t made for you. It wasn’t designed for your body, your movement, or your context.
Platform benchmarks from Meta, TikTok, YouTube, and others are built on data from thousands of campaigns—across every imaginable industry, audience, budget, and creative format. They’re directionally helpful, but ultimately impersonal. A benchmark might tell you your CPM is in the “top quartile,” but it won’t tell you if the impressions actually made a difference. It won’t tell you if your creative landed, or if you moved anyone meaningfully through the funnel. That’s why they’re often misleading when used in isolation.
Baselines: The Hoodie That Learns You
Now compare that to baselines. A baseline is your favorite hoodie. You’ve worn it a hundred times. It fits better every time you put it on—not because it was expensive, but because it knows you. In the same way, a baseline is built from your past performance data. It reflects how your campaigns have actually performed in the real world—with your brand, your audiences, your creative, and your spend levels.
Where benchmarks are generic and one-size-fits-most, baselines are personal. They allow you to say, “When we run $75K on Meta using UGC content in Q4, we typically see a 1.2% CTR and a $7.50 CPL.” That’s not just a reference point—it’s a performance memory that helps you plan more effectively and realistically. It turns your historical data into a predictive tool. And when used consistently, it becomes a much more powerful decision-making engine than any external benchmark.
The Deceptive CPM
Here’s a real-world example of where benchmarks can lead you astray. A consumer brand recently ran a Connected TV campaign. On paper, it looked like a win—the CPM came in 30% below the industry benchmark. The team called it a success and moved on. But when the placements were reviewed more closely, the truth came out: most of the impressions ran in overnight slots. Low viewability, no engagement, and no brand recall.
Yes, it was “efficient.” But it was also forgettable. The campaign delivered impressions, not impact. This is the danger of evaluating success based purely on a cost metric. When you use benchmarks as a substitute for real effectiveness, you can easily mistake cheap delivery for meaningful performance.
So How Should You Use Benchmarks?
Benchmarks aren’t inherently bad. They’re useful for orientation—especially when you’re entering a new platform, testing a new format, or setting initial expectations. The problem is when they become the only frame of reference. The best use of benchmarks is as a complement to your own data, not a replacement for it.
If you’re comparing campaign performance, make sure the benchmark is even relevant. Was it measured on the same objective? Was the creative type similar? Was the spend in the same ballpark? If not, you’re comparing apples to something else entirely. Always layer your benchmarks with baselines. Let your internal data guide the narrative. If your own performance was down compared to last quarter—even if you beat the platform average—that’s a red flag, not a reason to celebrate.
And above all, don’t confuse “above average” with “successful.” A high CTR that leads to no action, or a low CPM that reaches the wrong audience, doesn’t help your brand grow.
Benchmarks Are One-Dimensional. Baselines Are Multi-Layered.
Benchmarks give you a snapshot. Baselines give you a timeline. Benchmarks tell you how you stack up against others. Baselines tell you how you’re progressing over time. One is outward-facing; the other is rooted in your own evolution. Both have their place, but only one helps you get better in a way that matters.
Want to Go Deeper?
If this resonates, check out this excellent episode from The Analytics Power Hour:
🎧 #254: Is Your Use of Benchmarks Above Average? featuring Eric Sandosham.
It’s a thoughtful, clear-headed take on how analytics teams can stop chasing “average” and start building measurement frameworks that actually reflect reality.
Takeaway? Benchmarks can help you start the conversation. But baselines will help you finish it. A rental tux might look good in a photo, but a hoodie that fits is what you’ll reach for when things get real.
Next time someone says, “We beat the benchmark,” ask a better question:
Did we beat ourselves at our best?