What are Benchmarks?
Your metrics only tell half the story. A 3.2x ROAS might be excellent in one industry and below average in another. Benchmarks answer the question every marketer eventually asks: "Is this number actually good?"
Benchmarks compare your store's performance against a peer group of anonymized brands like yours (matched to what your store sells and the scale you operate at) so you can see whether you're leading the pack or have room to grow on the metrics that matter most.
The 6 metrics you can benchmark
Benchmarks cover the six metrics most marketers use to gauge performance:
Metric | What it tells you | Direction |
Blended ROAS | Total return on ad spend across all channels | Higher is better |
Blended CPA | Blended cost to acquire a customer | Lower is better |
CPM | Cost per 1,000 impressions — how expensive your reach is | Lower is better |
Avg CTR | Average click-through rate on your ads | Higher is better |
New Customer % | Share of orders coming from first-time customers | Higher is better |
AOV | Average order value | Higher is better |
For each metric you'll see your value, the median for your peer group, and a rank badge:
🟢 Top 25% — you're outperforming most brands like you
⚪ Average — you're in the middle of the pack
🔴 Bottom 25% — there's clear room to improve
Your peer group ("Brands Like You")
Benchmarks are only useful if you're being compared to the right brands, so Triple Whale builds your peer group for you. When you enable Benchmarks, we automatically find the brands most similar to yours and use them as your comparison cohort. There's nothing to configure: you're always compared against your closest peers based on your store's profile.
How we find brands like yours. Rather than bucketing you into a broad category, we match on two things together:
What your store sells: your products, the audience you serve, and how you position your brand (premium vs. value, niche vs. broad, and so on).
The scale you operate at: signals like your revenue, advertising investment, and average order value.
Combining both means your peer group is made up of brands that genuinely resemble yours — not just anyone who happens to share an industry label — so the comparison is relevant and fair.
Note: We need a minimum number of similar brands in a peer group before we'll show benchmarks (this also keeps every brand's data anonymous). If there aren't yet enough comparable brands, you simply won't see a benchmark for that metric.
How to turn on Benchmarks
Benchmarks are available on the Attribution page while configured for the Triple Attribution model with a Lifetime attribution window. Once you're in that view:
Go to the Attribution page.
Set the attribution model to Triple Attribution and the attribution window to Lifetime, with the source set to All.
Open the date picker comparison and toggle on Benchmarks ("Compare against peer cohort medians").
Your benchmark metric cards appear below the attribution chart, and an inline comparison is added to your attribution table. To turn the comparison off, toggle Benchmarks off in the date picker.
Note: Benchmarks only appear on Triple Attribution / Lifetime / All channels — this is the exact basis the peer comparison is computed on, so the numbers always line up. If you switch to a different model, window, or a single channel, the benchmark comparison is hidden automatically. Time-over-time date comparison is also disabled while Benchmarks are active, since the two views can't be shown together.
Reading the distribution view
Click the chevron on any metric card to open the distribution view that provides a deeper look at where you land relative to your peers.
You'll see:
A distribution curve showing how all the brands in your peer group are spread across the metric.
A "Median" line marking the middle of the pack.
A "You" marker showing exactly where your store falls on the curve.
Your value, the Peer cohort median, and Your rank (e.g. Top 25%) summarized above the chart.
A note on whether higher or lower is better for that metric, so the rank is never ambiguous.
The further the "You" marker sits toward the better-performing tail, the stronger your performance relative to similar brands. Seeing yourself in the Bottom 25% on, say, CPM is a clear signal that your cost of reach is high for your category, highlighting a concrete place to focus optimization.
How benchmarks are calculated
Where the data comes from
Benchmarks are computed daily from aggregated, anonymized performance data across Triple Whale brands. Pixel-based metrics (Blended ROAS, CPA, New Customer %, AOV) come from attributed order data; ad-platform metrics (CPM, CTR) come from connected ad accounts.
Attribution model
To keep comparisons apples-to-apples, every brand's benchmark figures are computed on a single, consistent basis using the Triple Attribution model with a Lifetime attribution window. This is exactly why Benchmarks are only shown when you're viewing that same model and window: you're always compared on a like-for-like footing, so there's never a mismatch between how your numbers and your peers' numbers were measured.
How your rank is determined
For each metric we fit a distribution across your peer group and use it to place your value on a percentile scale:
Most metrics (ROAS, CPA, CPM, CTR, AOV) follow a log-normal distribution, which fits the right-skewed shape typical of marketing metrics.
New Customer % uses a normal distribution, since it's a bounded 0–100% rate.
Your value is mapped to a percentile via the distribution's cumulative function, then translated into a rank: Top 25%, Average, or Bottom 25%. For "lower is better" metrics (CPA, CPM) the scale is inverted so that being in the Top 25% always means strong performance.
Outlier handling
Extreme values that would distort the peer comparison are excluded before percentiles are computed (for example, implausibly high ROAS values are filtered out). All monetary metrics are normalized to a common currency (USD) before comparison and converted back for display.
Your data is private
Benchmarks are built to be useful without ever exposing any individual brand's data:
Peers are anonymized at the data layer. Individual brands in a cohort are referenced only by irreversible hashed identifiers. Peer brand identities are never sent to your browser. You can never see which specific brands are in your cohort, and they can never see yours.
Only aggregated statistics leave the system. Data includes medians and percentile ranges across the whole peer group. No individual store's metric value is ever exposed.
A minimum peer-group size is enforced. A benchmark is only returned once your peer group meets a minimum size threshold (at least 5 similar brands). Smaller cohorts return no benchmark, so no individual brand's data can be inferred.
Data-quality gating. Peer ad-platform data is included only when it passes validity checks (valid currency, real spend, properly mapped ad accounts), so benchmarks aren't skewed by misconfigured or partial data.
FAQ
Why don't I see Benchmarks?
Benchmarks appear on the Attribution page only when you're on the Triple Attribution model, the Lifetime window, and viewing all channels (not a single source or a campaign-group breakdown). Benchmarksw are only available in the Automate or Enterprise packages.
Why doesn't a metric show a benchmark?
We only show a benchmark once there are enough similar brands in your peer group to produce a reliable, anonymous comparison. If there aren't yet enough comparable brands for a given metric, that benchmark won't appear.
Can other brands see my numbers? No. All comparisons are anonymized and aggregated. No brand (including you) can identify another brand's individual metrics.