A tweet is finished in a few hours. An Instagram post has a couple of days. A YouTube Short can be sitting in a search result two years from now, still being watched by people who have never heard of you. YouTube is the only major social platform that is also a search engine with a long memory, and that changes what your content is worth there.
The good news is that you probably do not need new ideas to start. The posts that already worked for you are proven scripts. Repurposing them into Shorts is mostly a production and formatting job, not a creative one.
What a Short actually is, technically
A Short is a vertical video, 9:16, that YouTube surfaces in the dedicated Shorts feed as well as in normal search and suggested results. The maximum length was raised to three minutes in late 2024, up from sixty seconds, which quietly changed the format. You are no longer forced to compress an idea into a minute, and a two-minute explainer with a real answer in it now qualifies.
You no longer need to put #Shorts in the title. YouTube detects the format from the aspect ratio and duration. What you do still need is a proper title and description, because unlike other short-form platforms, a Short is a video on your channel and is indexed like one.
The link rule that catches everyone out
This is the detail most repurposing advice gets wrong. Since 2023 YouTube has made links non-clickable in Shorts descriptions, Shorts comments and the vertical live feed, as an anti-spam measure. You can still type a URL. It will render as plain text that nobody can tap.
The workaround YouTube pushed people toward is the links section on your channel profile, which is prominent and clickable. So the funnel for a Short is: video, then channel page, then link. Practically that means your channel banner and profile links have to be set up properly before Shorts are worth making, and your call to action inside the video should say “link on my channel” rather than “link in the description”. Plenty of people are still telling viewers to check a description link that cannot be clicked.
It also means the tracked link belongs on the channel profile and in any long-form video descriptions, where links do still work. Set those up once, and every Short you publish feeds a link you can actually measure.
Turning a written post into a Short
Start with the posts that already earned attention. A thread that got saved, a LinkedIn post that got long comments, a Reddit answer that got upvoted. Those have been market-tested. Do not start with your favourite idea, start with the audience’s favourite idea.
- Take one point per Short. A five-point thread becomes five Shorts, not one crammed video.
- Say the payoff in the first two seconds. Not a greeting, not your name, not “in this video”. The answer, immediately.
- Shoot vertical and frame yourself off-centre if you plan to add text, so nothing collides with the UI overlays.
- Burn captions into the video. Most viewing is silent.
- End on a specific next step, and make it match where the clickable link actually is.
The production bar is lower than people fear. A phone, a window for light, and a lapel mic will beat a beautifully lit video with bad audio every time. Viewers forgive rough video and abandon bad sound within seconds.
Write titles and descriptions like search queries
This is where Shorts differ most from TikTok. The title of a Short is indexed and shows in search results. Write it as the question a person would type, not as a tease. “How to fix a 426 error on the LinkedIn API” will keep earning views. “You will not believe this bug” will not, because nobody searches for it.
The description should contain a couple of sentences of real context using the same phrasing, because it feeds both search and the suggested sidebar. Treat it like a meta description. It costs a minute per video and it is the difference between a Short that dies with its first push and one that keeps being found.
Publishing consistently matters too, and this is where a scheduler stops the effort from collapsing after two weeks. Batch a month of Shorts in one afternoon, then release them on a schedule alongside the written versions on your other networks. seenpaid handles the cross-network scheduling side of that across 21 platforms, so the same idea ships as a Short, a post and a thread without you doing it three times by hand.
Batch the production or it will not survive a busy month
The reason most repurposing plans fail is not the idea, it is the context switching. Setting up a camera, checking the light, finding a quiet room and getting into the right headspace costs about the same whether you record one Short or eight. Doing it once per video means the real cost per Short is triple what you estimated, and the habit collapses the first week something goes wrong.
So batch. Pick a morning, set up once, and record eight to twelve Shorts in a row from a list of proven posts you prepared the day before. Edit them in one session too. Two half-days a month is a realistic commitment that produces a video every weekday, and it survives a bad week because the work is already done and queued.
The queue is what makes it durable. Once a batch exists, publishing is a scheduling decision rather than a daily act of will. Release them evenly rather than dumping them, and pair each Short with the written version of the same idea going out on your other networks the same day, so one piece of thinking covers a week of presence everywhere instead of one video on one platform.
Do not treat Shorts as a separate channel from your long-form
Shorts and long-form videos live on the same channel and feed each other. A viewer who finds a Short and subscribes then sees your longer videos, and YouTube surfaces both to the same audience. If your channel has nothing but Shorts, a new subscriber has nowhere to go next and the relationship stalls at the moment it started.
A workable structure is to keep the Shorts as the discovery layer and have two or three substantial long-form videos as the destination. Those longer videos are where clickable links still work, where you can explain something properly, and where a viewer decides whether to become a customer. The Shorts do the reach; the long-form does the convincing.
Measuring content that pays you six months later
YouTube Analytics is genuinely good at what it measures: views, average view duration, swipe-away rate, and where viewers came from. It has one hard limit, the same limit every social analytics dashboard has. It stops at the click. It cannot tell you that the person who watched a Short in March bought from you in September.
That gap is bigger on YouTube than anywhere else, precisely because the payoff is so delayed. The video you forgot about is the one making today’s sales, and if you judge Shorts on the first week of views you will kill the wrong ones and keep the wrong ones.
Fix it by attributing revenue rather than traffic. Keep tracked links on the channel profile and long-form descriptions, ask new customers how they found you, and look at which videos are associated with actual payments over a long window rather than a week. seenpaid ties Stripe revenue back to individual posts read-only, which is the only way a long-tail channel like YouTube gets judged honestly.