If you score a thumbnail and see a low number on one factor, it helps to know whether that is unusual or whether almost everyone scores low there. We scored 642 real videos for a pre-registered test of our own score, so we can show how each of the twelve factors is actually distributed. The pattern is not the one you might expect: the lowest factors are not the visual basics but the ones about urgency and curiosity.
Key takeaways
- Among 642 real videos from 60 established channels, urgency scored lowest: a median of 40 out of 100, with 70% of videos below 50.
- Curiosity was next, with a median of 55 and 28% of videos below 50. Cliffhanger, rarity and fear of missing out followed.
- The visual and clarity factors scored high almost everywhere. Thumbnail readability never fell below 85, and coherence between thumbnail and title had a median of 90.
- A low score is not a failure. A tutorial does not need urgency, and a calm title can be the right one. Low simply means the pair does not signal that quality.
Where does this data come from?
It comes from the videos scored for our published test of the score. We took 60 established channels across six categories, tech, education, cooking, gaming, fitness and lifestyle, and scored 642 videos, which form 321 matched pairs from the same channel. The scores were produced on 4 September 2026 by the same pipeline that runs in the analyzer. Each video receives a 0-100 score on each of twelve factors, plus an overall score.
Be clear about what this is. It is a distribution of scores across a particular sample, not a measurement of how creators in general perform. The channels are established, which skews the visual factors upward, and a model produced the scores. The method and evidence page describes the sample in full.
What counts as "low" here?
We use two simple readings: the median score for each factor, and the share of videos scoring below 50. Fifty is a round threshold we chose for this article, not a standard, and it is labelled as such wherever it appears. The medians do not depend on it.
Which of the twelve factors score lowest?
The lowest are the ones that describe how a video holds attention, not how the image is drawn. The table shows the six lowest by median, with the share of videos scoring below 50 on each.
| Factor | Median score | Videos below 50 |
|---|---|---|
| Urgency | 40 | 70% |
| Curiosity | 55 | 28% |
| Cliffhanger | 65 | 19% |
| Fear of missing out | 65 | 12% |
| Rarity | 70 | 13% |
| Withheld answer (Zeigarnik effect) | 70 | 6% |
Urgency stands apart: seven in ten videos in the sample do not signal that watching now matters. That is expected, because most videos are not time-sensitive, and a title that invents urgency invites the exaggeration that raises clickbait risk. Curiosity is the other low one, which suggests that titles and thumbnails in this sample mostly state their subject and leave little unsaid.
Which factors score high almost everywhere?
The visual and clarity factors. Thumbnail readability ran from 85 to 95 across all 642 videos, with a median of 92. Coherence between thumbnail and title had a median of 90, and emotion a median of 80. Promise clarity had a median of 75 and title strength 74. Clickbait risk had a median of 6, where a high number is the bad one, so almost no video in the sample looked exaggerated.
These are the factors an established channel has already learned to get right. They are also the ones that separated the better-performing videos from the weaker ones least, with the exception of promise clarity. The article on readability as table stakes goes into why a factor that everyone clears stops being a differentiator.
Do the low factors matter more because they are low?
A little, though not much. In the 321 pairs, the better-performing videos scored slightly higher than the weaker ones on the low factors: urgency 44.6 against 41.1, curiosity 53.5 against 51.9, and fear of missing out 64.8 against 60.7. Those are gaps of a few points. They suggest that where a factor varies, it can carry some signal, which is consistent with the finding that the factors about the message separated videos more than the visual ones did. They do not show that raising urgency raises views.
What should you do with a low score on one factor?
Ask first whether the quality belongs in your video. Urgency does not suit an evergreen tutorial, and a withheld answer does not suit a straightforward review. A low score there is not a defect to fix.
- Score the thumbnail and title together with the free thumbnail and title analyzer.
- Find the lowest factor and ask whether this video should have that quality at all.
- If it should, change the thumbnail or title in one way that raises it, and nothing else.
- Score again and compare, keeping the change only if the pair still says something true about the video.
- Check the result against your audience with an A/B test before you commit.
Step four matters. Raising urgency or curiosity by exaggerating raises clickbait risk, which the score counts against you. The guide to the twelve factors explains how each is read.
How much should you trust a distribution like this?
As a reference point, not a benchmark. Across the 321 pairs, the overall score ranked the better-performing video higher 59.5% of the time against a 50% coin flip, with a 95% confidence interval of 53.9% to 64.9%. That is a real effect and a modest one, so a single factor is a weaker signal still. The sample is 60 established channels, the scores come from a model, and a newer or smaller channel may sit elsewhere in the distribution. Use it to see whether a low number is common, then decide for your own video. The write-up of the 321-pair test has the method and the controls.
Correction, 6 October 2026. An earlier version of this article presented data that had not been collected. It has been rewritten from the published results of our 321-pair test. The list of corrections is on the method and evidence page.
Sources
- ThumbSignal. Method and evidence: how the ThumbSignal score was tested, for the design, the result, the controls and the limits.
- ThumbSignal. Pre-registration, pair scores and verification script, public since 7 October 2026. github.com/MonteroDev/thumbsignal-score-study
- ThumbSignal. We tested our thumbnail score on 321 pairs of real videos.
Frequently asked questions
Which thumbnail factor do creators miss most often?
We cannot say what creators in general miss. In our sample of 642 real videos from established channels, the lowest-scoring factor was urgency, with a median of 40 and 70% of videos below 50, followed by curiosity.
Does a low score on a factor mean my thumbnail will fail?
No. A low score means the thumbnail and title do not signal that quality. Some videos should not, such as a calm explainer without time pressure. Whether the video performs depends on the audience, the topic and the timing as well.
Where do these figures come from?
From the 642 videos scored for our published test of the score: 321 matched pairs from 60 established channels. Every number traces to that test, described on the method and evidence page.
Should I try to raise every low factor?
No. Raise a factor only when it fits the video and you can do it honestly. Pushing urgency or curiosity past what the video delivers raises clickbait risk and tends to cost trust.
How is the share below 50 calculated?
It is the proportion of the 642 videos whose score on that factor was under 50 out of 100. The threshold is our own choice for this article, so rely on the medians if you want a measure that does not depend on it.
