What Rick Rubin Would Cut Before You Send That Beat

How do you know which idea is worth sending?

Rubin’s test is the house on the mountain. Imagine you moved somewhere no visitor could ever reach and curated the place entirely for yourself. Would you make the same choices? The work that passes is the work you’d want to exist whether or not anyone showed up.

Point that at your ideas folder. Not which loop is most likely to get a reply — which one do you actually want to hear finished. That’s the one to send, because it’s the only one you’ll still care about in three weeks when a verse comes back and it needs work.

The Red Hot Chili Peppers version is more mechanical. For one album they might record 40 or 50 songs, then everyone — the band and Rubin together — votes each one A, B, or C. A unanimous A goes on the record; anything the vote is really divided on probably doesn’t. You don’t have a band to vote, but a folder of short sketches is already a shortlist waiting to be sorted.

What should you cut before you send a track?

Start with the one-tenth problem. Stack ten things on top of each other and each one becomes a tenth as important. One guitarist playing one take carries more personality than eight overdubs of the same part, because you can hear a person in it.

So the move is subtraction, not addition. Mute one layer and listen to what gets louder. It’s usually the part that was carrying the idea the whole time, and now it has room. Keep going until you’re near 40 percent, then add back only what the song can’t do without.

Then cut once more for the send. What a collaborator needs is the shortest version that still carries the idea — the hook, the loop, the eight bars you would defend, somewhere between fifteen and thirty seconds. That isn’t a watered-down version of your track. It’s the same edit, done one more time.

This is why Muselink caps every snippet at 10–15 seconds, and the reason isn’t file size. A full arrangement lets you dodge the decision: send everything, hope they find the good part. Fifteen seconds makes you name it. The side effect is that it’s too little material for anyone to release as their own.

Does it matter how your collaborator works?

Rubin has worked with both extremes — he produced Jay-Z’s 99 Problems and executive-produced Eminem’s The Marshall Mathers LP 2 — and describes them as opposites. Jay-Z writes nothing down: it happens in his head, then he gets up and says it once or twice, and that’s it. Eminem writes the lyrics out, works studiously, and records take after take, trying different things. Neither one is doing it wrong.

In a collab, a process mismatch gets read as a character flaw. The fast one decides the slow one is flaky. The slow one decides the fast one doesn’t care. Both are wrong, and the song dies of a misunderstanding neither person said out loud.

So say it in the first conversation. “I record fast and rough, you’ll have something tonight” and “I need a week and three passes before I send anything” are both good answers. Finding out on day nine is what kills it — and how someone answers is the same signal that separates serious collaborators from curious ones.

Who is the song for?

Picking covers to pitch Johnny Cash for the American series, Rubin worked from a single image: what suits the Man in Black. He has said he saw Cash as a mythological figure more than as the funny, flesh-and-blood guy. Not a genre, not a reference playlist — a character.

Note where the authority actually sat, because it’s the part worth copying. Rubin chose what to offer; Cash chose what to sing. By Rubin’s own account more than half the songs were Cash’s picks and all of them were his call in the end — Cash killed Rubin’s pitch of Addicted to Love, and refused Hurt until Rubin talked him round. The character narrowed the suggestions. It never overruled the artist.

Two people making one song usually hold two different pictures of the finished thing and never compare them. Then they argue about a snare for three weeks and think the argument is about the snare.

Name the character before the first take. Who is this for, where does it get heard, what would it never do. One message settles a hundred small decisions later — including which producer or vocalist you should be looking for in the first place.

What if you and your collaborator want different things?

Rubin borrows a distinction from Jimmy Iovine, who told him: “I’m in the banking business. You’re in the church business.” Iovine works from the bottom line — he heard a Tom Petty song once and said flatly that it was going to buy Petty a house. Rubin starts from belief instead.

Neither is the wrong business to be in. The damage comes from mixing them halfway through a project, because a structural disagreement then shows up disguised as a personal one. Someone gets called a sellout, someone gets called precious, and the track never comes out.

Ask early. Is this a demo or a release. Is anyone pitching it. What happens if it does well. It’s the same conversation as credits and splits, and it costs far less before the song exists than after.

What if you don’t actually enjoy the work?

Rubin calls himself a lazy workaholic. In his words: “I have to force myself to do it, but I do force myself.” His natural state is doing nothing. He spent most of the first 25 years of his career in a dark room for 16 hours a day regardless.

The distinction is that he loves the finished thing, not the grind that produces it. Those are two separate motivations, and only one of them shows up reliably. If you’ve been waiting to feel inspired before you open your DAW, you’ve been waiting on the wrong signal.

This is the one place collaboration does something discipline can’t. Someone waiting on a hook is an external deadline, and an external deadline still works on the days your own motivation doesn’t.

Does any of this help if nobody’s listening yet?

Worth saying plainly: Rubin asks these questions from inside a career where the audience is guaranteed. Cutting to 40 percent is a different exercise when 100 percent is a finished record with a label behind it than when it’s four loops on a laptop and nobody waiting.

His rules are decision rules. They tell you what to keep. They don’t tell you how to get heard, and no amount of taste finds you a vocalist. That half is still logistics — rooms to post in, people to ask, replies that never come.

Muselink.app takes the second half and runs it on the first half’s logic. Every card is a 10–15 second snippet tagged with the role its owner is looking for, so what gets judged is the sound. Chat opens only on a mutual match, so there’s no inbox to sort.

And until you match, you show up as a number rather than a name. That’s the house on the mountain turned around to face the listener — strip out the follower count, the bio and the credits, and the only thing left to judge is whether the fifteen seconds is any good.

Some honest limits. It’s iOS only for now, and it’s a discovery layer rather than a workspace, so you keep your own DAW. If what you want is feedback rather than a collaborator, the communities do that better than any matching app. And if you want a finished mix rather than a creative partner, hiring an engineer outright is the honest answer — that is a different job from the one this solves.

The Rubin material here comes from his conversation on David Senra’s podcast and his book The Creative Act — the ideas are his, the collab translation is ours. Get early access — free during early access.