You get the add on a Tuesday. The playlist has forty thousand followers, the counter starts moving, and for three days it feels like the thing finally happened.
Then it flattens. No Discover Weekly. No Release Radar. The track never reaches the larger audience you thought the placement might unlock.
Nothing broke. Nobody scammed you. The placement was real and the streams were real, and that is exactly why this one is hard to see: the damage does not arrive as a problem. It arrives as a win that goes nowhere.
A bad record is not the only explanation for a campaign that goes nowhere. Sometimes the record was simply put in front of the wrong audience. This analysis separates what Spotify has published from what promotion companies infer.
Four words you need before the rest of this makes sense
No gatekeeping here. If you already know these, skip the list.
- A save is a listener adding your track to their own library or their own playlist.
- A skip is a listener leaving before the track ends. Spotify counts a play as a stream at 30 seconds.
- A playlist add, here, means a normal listener putting your song into a playlist they made. That is not the same as a curator adding you to theirs.
- An algorithmic playlist is one built by software rather than by a person: Discover Weekly, Release Radar, Radio, Autoplay.
What Spotify actually says it watches
On its own recommendations page, Spotify writes: "As you engage with Spotify, actions such as searching, listening, skipping, or saving to Your Library influence our interpretation of your taste." It calls the result your taste profile, and says "your taste profile is the most important input for creating the best overall user experience."
One more line from the same page matters more than the other two: "When users take similar actions on certain content, our algorithms have more information about how people might want to interact with it."
Source: Spotify, "Understanding recommendations", spotify.com/us/safetyandprivacy/understanding-recommendations (checked 10 August 2026)
Read that last line as an artist instead of as a listener. When a crowd of people all do the same thing to your song, the system learns from it. That works in both directions.
How a wrong-crowd placement can stall a house track
Say you make organic house. Congas, a filtered pad, a vocal that arrives at 1:40. You land on a list called something like "EDM Bangers 2026".
The people on that list are there for a drop at forty seconds. They get ninety seconds of drums. They skip, and they skip early. Nobody saves it, because they did not come looking for a record to live with. They came for a mood.
The placement can leave the system with a large, consistent block of listener actions on your track, and all of them point the same way. Skip, skip, no save, skip. That resembles the "users take similar actions" case Spotify describes. Spotify does not publish the exact consequence for a track.
The version that catches more house producers is smaller and much harder to spot, because the genre tag is right.
A 3am deep roller lands on a poolside sunset playlist. Same genre. Near enough the same tempo. Completely different job. The sunset crowd wants something that sits behind conversation, and yours was built for a dark floor at hour four. They do not hate it. They just reach for the next one.
Then there is the third case: the technically-correct house playlist that is pure background listening. Thousands of plays, almost no saves. The behaviour reads as nobody minded and nobody cared.
The wrong playlist does not always do nothing. It can do damage.
None of those three placements broke a single rule. That is what makes them expensive.
Punished versus ignored, the line nobody draws
Almost everything written about this subject blurs two completely different situations together. They deserve separate names.
Punished is a policy problem. Spotify defines an artificial stream as "a stream that doesn't reflect genuine user listening intent, including any instance of attempting to manipulate streaming services like Spotify by using automated processes (like bots or scripts)." When it detects them, Spotify says those streams "do not earn royalties", "do not count toward public stream numbers or charts", and "do not positively influence recommendation algorithms." It also lists what can follow: removal from playlists, your distributor suspending your account, and in repeated or severe cases the track coming off the platform.
Spotify is blunt about who sells this. On its page about services that guarantee streams it writes: "Any service that claims to offer guaranteed placement on playlists on Spotify in exchange for money are in violation of our terms & conditions." On the artificial-streaming page it adds: "They may claim to be able to get a song streamed, playlisted, or prioritized in Spotify's algorithmic recommendations. None of this is true."
Sources: Spotify for Artists, "Artificial Streaming", artists.spotify.com/artificial-streaming · Spotify Support, "Artificial streaming and paid 3rd-party services that guarantee streams" (both checked 10 August 2026)
Ignored is the quiet one. No bot. No rule broken. No email, no flag, no penalty, nothing to appeal. Real people heard your track in the wrong playlist lane and behaved honestly. The track may stop expanding through recommendations, but Spotify does not expose enough of the system to prove why.
The numbers you will see quoted, and where they come from
Search this topic and you get thresholds delivered with decimal-point confidence. Here is where each one actually comes from, checked on 10 August 2026.
| The claim | Who publishes it | What it rests on |
|---|---|---|
| A 20% save rate from cold (unfamiliar) listeners means a much higher chance of Discover Weekly expansion | Chartlex | Its own campaign set, described as patterns across 2,400+ Chartlex campaigns. No Spotify source given. |
| A skip rate above 35% in the first 30 seconds stalls distribution | Chartlex | The same internal campaign data. |
| A good save rate is 8% to 15% or higher | Artistrack | Not attributed to Spotify. |
| Skips inside the first 30 seconds count more heavily than a skip at two minutes | Emitha | Matches the direction of what Spotify publishes. The weighting itself is Emitha's read, not a Spotify statement. |
| Low saves and high skip rates stop algorithmic growth | GPM Music Group | Correct in direction, and they do not claim a number. |
| "Actions such as searching, listening, skipping, or saving to Your Library influence our interpretation of your taste" | Spotify | Spotify. |
Only the last row is a Spotify statement.
Why we do not publish our own numbers yet
To be fair to all four of them: Emitha and GPM have the mechanism right, and Chartlex's numbers come from campaigns they actually ran, which is a very different thing from making them up. But a number drawn from one company's client base is that company's average, not a Spotify rule. Nobody has published a threshold, because nobody outside Spotify can see one.
We do not publish our own numbers yet either. When we have run enough campaigns to say something honest, it will live on a dated Selectors Club campaign-data page with the sample size printed next to it, and you will be able to check the maths yourself.
What a right placement looks like
Turn the mechanism around and the buying question changes. Not how many followers. Will these listeners save it.
Four checks before you accept any placement, paid or free:
- Play the last twenty tracks on the list. Would a DJ play your record inside that run, at that hour of the night? If you have to argue for it, that is a no.
- Check that the listeners are people. Real playlists have followers whose behaviour makes sense: the artists on it have believable monthly listener counts, and the adds are spread out over time rather than dumped in one day.
- Ask who added you, and whether they heard the whole record. A person who listened all the way through and said yes is a completely different signal source than a genre filter that matched a tag.
- Ask what happens if the answer is no. Anyone who cannot say no to you is not choosing. They are placing.
That last check is the idea behind the service Selectors Club is launching. Working house DJs will hear the record in full, say no when it does not fit, and make only earned playlist adds. Selectors Club will not guarantee placement, and by Spotify's own terms, nobody should.
Reading a dance floor is the whole job of a DJ. Choosing where a record goes uses the same skill.
Are you a house producer looking for a Spotify playlist network that actually understands house music?
What nobody outside Spotify can prove
Here is the honest edge of all of this.
We can see what Spotify says goes in. Artists and campaign operators can see what comes out on their dashboards. The middle is closed, and no outside promotion company has seen the model.
So when you read a specific trigger number, treat it as somebody's average dressed up as a law. And when a service promises you a specific algorithmic outcome, remember that the outcome was never theirs to promise.
The inputs are documented even if the exact consequence is not. Spotify publishes that skipping and saving shape what it thinks a listener wants, and that similar actions from many people teach it more. Put a record in front of people who were never going to save it and you may be answering a question about your song with the wrong crowd holding the microphone.
Streams count plays. Saves show who wanted the track again.
Before you pay, compare what the main promotion services actually sell.