Your best reference is probably not in your category
The instinct when researching short-form video is to look at what similar brands are doing. It is the most natural move available and one of the least productive.
Why competitor research underdelivers
Your competitors share your constraints. The same legal caution, the same claims they can and cannot make, the same assumptions about what a serious brand in this category posts. Whatever is wrong with your category’s content is usually wrong with theirs for the same reasons.
Their numbers are also hard to read. A competitor with a larger following gets more reach on worse creative, and you have no visibility into what they tried that failed. You end up studying a filtered, flattering sample of an approach shaped by the same limits as yours.
What a viewer is actually comparing you to
Nobody watches your video next to a competitor’s. They watch it between a friend’s post and something genuinely entertaining, on a feed with no category boundaries at all. The competition for attention is everything, not the eight other brands selling what you sell.
That reframes the research question. You are not trying to be the best video in your category. You are trying to be worth not scrolling past, which is a standard set by people who are not competing with you commercially at all.
Structures travel, subjects do not
A format that makes a complicated process watchable in fifteen seconds is useful regardless of whether the original was about software, cooking or plumbing. The mechanic — show the result first, then rewind to the method — has nothing to do with the subject it was first used on.
This is why a kitchen gadget video can be the right reference for a skincare brand, and why a plumber explaining a repair can be the right reference for a developer tool. The subject is the part you replace. The structure is the part you keep.
Searching further out on purpose
The practical version is to define the problem structurally before searching. Not “skincare videos” but “videos that make a slow result feel immediate.” Not “SaaS demos” but “videos that explain something abstract without a whiteboard.”
Those searches return references from everywhere, which is the point. The ones from outside your category are the ones your competitors are not already copying, which is the only way to end up with something that does not look like everything else in the feed.
It also changes what you notice. Searching your category trains you to see topics, because that is what varies between the results. Searching a structure trains you to see mechanics, because the topics are all different and the structure is the only thing the results have in common.
Where this advice stops
Distance is not automatically good. A structure that depends on something your brand cannot supply — a charismatic on-camera presenter, a dramatic physical transformation, a genuinely funny writer — does not become available just because you found it somewhere unexpected. The constraint is what you can actually film, not where the idea came from.
Category knowledge is also real. Your audience has expectations about what a brand in your space sounds like, and violating them costs something. The argument here is not to ignore your category; it is that your category tells you the register and your references should tell you the structure.
A reasonable split is to look outward for the mechanic and inward for the boundaries. What has held attention somewhere real, and what can we honestly say and show. Most teams do the opposite: they take the register from outside and the structure from their competitors, and produce something that is both unfamiliar and unoriginal.