You play a good set. Somebody films it on a camera that’s better than a phone, and about three weeks later the whole night goes up on YouTube as one forty minute video. It gets 214 views. That footage cost you a rehearsal, a venue night and somebody’s entire weekend in an editing timeline, and now it just sits there, because one long upload really only gets a single shot at being found.

The numbers aren’t on your side here. Luminate’s year-end report, which came out in January 2026, counted something like 106,000 new tracks landing on streaming services every single day, and found that 88 percent of tracks never get past a thousand plays. Meanwhile Neal Mohan, who runs YouTube, wrote in his 2026 letter that Shorts were averaging around 200 billion views a day. So there’s attention out there. It just doesn’t wander over to a cold upload by itself. Bands deal with that in all sorts of ways, from working the mailing list and leaning on the other acts on the bill to paying to give a new video its first push, and the ones who actually get somewhere tend to have picked a route before the show, not three weeks after it.

One shoot, four edits

The decisions that matter happen at the venue, not in the timeline. Two cameras is the practical minimum: one locked off wide on a tripod covering the whole stage, one handheld moving between players. The wide angle is your safety net and the source of the full-set video. The handheld gives you faces, hands on frets, and the drummer counting in, which is what the vertical clips need.

Get the desk feed if the venue will let you. Phone audio recorded off a PA sounds exactly like phone audio recorded off a PA, and there’s no plugin that fixes it afterwards. No board feed available? Put a small recorder on a stand somewhere near front of house, not at the stage lip, where you’ll mostly capture whichever amp happens to be closest.

Then grab thirty seconds of the room before the set, and thirty after: doors opening, the merch table, somebody tuning up, the crowd shuffling out afterwards. It is filler between the music and you will be glad you have it.

Pick the premiere time carefully

Premieres create a landing page that exists before the video does, with a countdown and a live chat. That page can be shared for a few days in advance, which turns a passive upload into a small event people put in their calendar.

Bands throw this away on the scheduling more than on anything else. If the edit happens to finish at 2am and you set the premiere for 2am, nobody is coming. Work out roughly when the people who already listen to you are awake and not at work, then tell the mailing list about it twice, and turn up in the chat yourself once it starts.

Picking the Shorts

The instinct is to cut a Short per song. That produces four clips that all start with a band walking on stage and nothing happening for eleven seconds.

Cut for moments instead. The key change everyone in the room reacted to. The eight bars where the guitarist and bassist are locked in and grinning at each other. The thing the singer said between songs that got a laugh. The crowd noise after the last chord, which almost always sounds better than you remember.

Three or four clips of fifteen to forty seconds each, posted across a fortnight, will out-perform the full video every time, and every one of them can point back to it.

Chapters make the full set searchable

The long video still deserves to exist, it just needs to be navigable. Add chapter markers using the actual song titles and one forty minute file turns into eight findable performances. Someone searching a song name can drop into the middle of your set instead of scrubbing along a blank timeline hoping to hit it.

Write a description that says where the show was, when, who played and who filmed it. Yes, it reads like housekeeping. It’s also most of what YouTube has to go on when it decides what your video is even about.

The first few hundred views

Here is the uncomfortable bit. All of the above assumes somebody watches the first upload, and for a band without an audience yet, that assumption is doing a lot of work. YouTube decides how far to carry a video partly on how the earliest viewers behave, so a video that reaches nobody in week one tends to stay reached by nobody.

The free routes are the obvious ones and they still work: send it to the mailing list, ask the venue and the other bands on the bill to share it, post the Shorts where your listeners already are, and put the link in every reply to anyone who asks what you sound like.

Paying for that early traction is just marketing, assuming the traffic is real. And the usual criticism of that corner of the industry is fair enough, so it’s worth knowing what you’re looking at before you spend anything. Plenty of sellers hand you bot views that get scrubbed off within a fortnight, or dump ten thousand views on you in one afternoon in a pattern no genuine video has ever produced. What separates the ones worth using is fairly dull, and you can check all of it. Views should arrive gradually rather than in a single spike. The accounts behind them should be real people with watch histories. And nobody should ever ask you for a channel password, only a link. If a service wants account access, it isn’t a promotion service.

If you do spend, put it behind the video you’ve already committed to repurposing. Push a one-off upload and what you’ve bought is a number on a page. Put the same money behind the video that anchors a month of Shorts, chapters and a follow-up premiere, and the attention carries into all of them.

Film one show, upload one video, and you’ve made a document of the night. Film the same show with six uploads already mapped out and you’ve made yourself a month of work, in the good sense. Most of that difference gets decided before anyone plugs in.