What measured studies say about titles and thumbnails: rewriting titles for each community lifts engagement; outrage framing lifts engagement too - at a social cost
Two studies (abstract level). (1) Hu et al.: on a large dataset of Reddit posts sharing YouTube videos, 21% of post titles barely change the YouTube title; statistically, rewritten titles get more engagement, and a controlled multi-phase experiment that neutralises video popularity, timing and community norms finds effective rewrites show emotional resonance, lexical richness and fit with community-specific norms; a fine-tuned BERT predicted the better title in pairs with 74% accuracy, beating near-random baselines including GPT-4o. Implication: when distributing a video to another platform/community, write a native title rather than reusing the YouTube one - and do not assume a general LLM can pick the winner without tuning. (2) Park et al.: a fine-tuned vision-language classifier of moral-emotional framing in thumbnails+titles across ~400,000 news-channel YouTube videos (Korea and US) finds "other-condemning" moral outrage consistently increases views, likes AND comments across bo...
Structured fields
{
"citations": [
{
"url": "https://arxiv.org/abs/2505.03769v1",
"note": "abstract; abstract"
},
{
"url": "https://arxiv.org/abs/2601.21815v2",
"note": "abstract; abstract"
}
]
}Evidence claims
- [sourced_fact] Two studies (abstract level). (1) Hu et al.: on a large dataset of Reddit posts sharing YouTube videos, 21% of post titles barely change the YouTube title; statistically, rewritten titles get more engagement, and a controlled multi-phase experiment that neutralises video popularity, timing and community norms finds effective rewrites show emotional resonance, lexical richness and fit with community-specific norms; a fine-tuned BERT predicted the better title in pairs with 74% accuracy, beating near-random baselines including GPT-4o. Implication: when distributing a video to another platform/community, write a native title rather than reusing the YouTube one - and do not assume a general LLM can pick the winner without tuning. (2) Park et al.: a fine-tuned vision-language classifier of moral-emotional framing in thumbnails+titles across ~400,000 news-channel YouTube videos (Korea and US) finds "other-condemning" moral outrage consistently increases views, likes AND comments across both cultures; the authors warn this can deepen polarisation and release their classifiers. Implication for creators: outrage reliably buys engagement in news contexts, but it conflicts with honest packaging guidance (misleading thumbnails are common and detectable, C3-B425-02) and with community trust; prefer emotional resonance that does not misrepresent content. · locator: abstract; abstract source
Rights
| component | license | hosting | redistribution | modification | training | Evidence and notes |
|---|---|---|---|---|---|---|
| documentation | unknown | permitted | permitted | unknown | unknown | Applies to this original summary only, not to the cited sources. |