Explainer

The System1 Star Rating, Explained

If you've seen a Super Bowl ad ranking or a Christmas ad league table in the trade press, you've seen star ratings. Here's what the number actually measures, what it deliberately doesn't, and how it relates to the other number ad testers argue about — attention.


What the stars measure

System1 shows a finished ad to a panel of real viewers and measures emotional response as it plays. The star rating compresses that into one scale, and its claim to fame is validation: System1's research ties higher stars to long-term brand growth, which is why brand teams treat a strong star score as evidence an ad will compound over years, not just perform this quarter. Their published work has long argued most ads earn low stars — emotionally flat advertising is the norm, not the exception.

What the stars don't measure

A panel is a captive audience: respondents watch the whole ad because that's the job. A feed viewer owes you nothing and decides in seconds. So a star rating tells you how people feel about an ad they watched — it can't tell you what fraction of a real feed audience would still be present at second five. That's not a flaw; it's a scope boundary, the same kind we publish about our own score (attention, not persuasion or brand growth).

Hook (0–3s)CTA window (last 20%)0s3s15s24s30spredicted engagementAttention dropwhere viewers bailPeak momentPurchase Signalavg. engagement hereHook Strengthavg. of first 3s
The attention side of the picture: one engagement signal, read four ways — hook window, first meaningful drop, peak frame, CTA window.

Stars vs attention scores, side by side

Source: stars come from human panel reactions; attention scores from a model — ours trained on fMRI brain-response data — reading the creative itself. Shape: a star rating is one summary number; an attention score comes with a second-by-second curve showing exactly where the grip fails. Speed and cost: fieldwork pace and brand-budget pricing versus minutes and about $2.45 per test. Question answered: "will this build the brand?" versus "will anyone still be watching when it tries?" The deeper comparison: PreTestAds vs System1, and how attention relates to performance metrics.

Using both without paying twice for the same answer

The two metrics stack cleanly because they own different failure modes. Screen every cut for attention first — it's the cheap gate, and it catches the ads that die before emotion matters. Send the survivors with brand-level stakes to panel testing for the emotional read. What you shouldn't do is buy fieldwork to discover a weak hook: that's a $2.45 question. Start with a live sample report to see what the attention side looks like.

Get the attention half in minutes

Score your ad's hook and drop-off second — up to 8 free credits, no credit card.

Test Your Ad Free

Frequently asked questions

What does a System1 star rating measure?

It summarizes the emotional response a panel of real viewers had to an ad, on a star scale System1 has validated against long-term brand growth. More positive emotional response earns more stars; System1's published work argues most ads land at the low end of the scale.

Is a star rating the same as an attention score?

No. Stars summarize how a captive panel felt about an ad they were shown in full. An attention score predicts whether a feed viewer — who can leave at any second — keeps watching, and where they bail. Emotion presumes the watch; attention tests it.

Can I get a star rating without buying System1?

The star rating is System1's proprietary metric, so no. What you can get self-serve is the attention side: a 0–100 percentile score, hook strength, and drop-off second for about $2.45 per test, in minutes, with up to 8 free credits — measuring the gate an ad has to clear before emotion gets its chance.

By Nina Krecicki · Published