Spotify Plays

What counts as a play on Spotify?

Thirty seconds of playback. A track has to run for at least thirty seconds before Spotify records a stream against it, and anything shorter is logged as a skip instead. That single threshold shapes this entire product, because it is the line every delivery has to cross before it appears on the counter at all.

It also explains a price pattern worth understanding before you compare tiers. Playback costs a supplier real time, so the cheapest deliveries cluster at exactly thirty or thirty-one seconds — the shortest playback that registers. Uniformity at that boundary is one of the most legible shapes in listening data, because a room full of real listeners does not stop at the same second.

Deliveries that vary the play length, and let a share of plays run through to the end of the track, take longer to produce and cost more per thousand. Asking a supplier which of the two they run is the most useful question available to you, and the answer describes most of what you are buying. The tiers priced above are built on varied durations rather than a flat thirty-second pass.

What does Spotify do about streams it identifies as artificial?

Spotify operates a published artificial streaming policy and states the consequences under it openly. Streams identified as artificial earn no royalties, are removed from public stream totals and from chart eligibility, and are excluded from the recommendation system. On the platform's own side of the transaction, that is the extent of it: the number is corrected and the stream stops counting.

There is a second party involved, and this is the part most pages selling this product leave out. Spotify reports detected activity to the distributor that delivered the release, and distributors publish their own policies covering what follows. DistroKid, TuneCore, CD Baby and their peers describe a range that runs from a warning notice to a per-track administrative fee charged to the artist, and in repeated cases to withdrawal of the release. That is a contract between you and your distributor rather than between you and Spotify, and reading yours takes about five minutes.

This is the one product in the catalogue where a third party you already have an agreement with holds a stated position on the purchase, so naming it is the job of this section. The delivery here is built around the patterns that assessment reads — pacing, duration variance, account diversity, market spread — and everything else is information you are entitled to have in front of you while you choose a tier.

How does the delivery actually work?

You paste the Spotify link and the order runs against it. No password, no Spotify for Artists access, no app to install and nothing to authorise. The delivery reaches the public side of a public track, which is the same surface any listener already reaches from a search box.

Streams are then added across a delivery window rather than all at once. Start is usually within a few hours of the order clearing, and the window scales with the quantity: a small order finishes inside a day or two, a larger one is deliberately paced over a week or more. That pacing is a property of the product rather than a queue backlog, for the reasons the delivery-time section below sets out.

The one requirement on your side is that the track stays live and public for the length of the window. A release pulled or switched to a different distributor mid-order leaves the delivery with no target, so the order stops rather than redirecting itself somewhere else, and anything undelivered is refunded.

Do bought streams earn royalties?

Royalties are a separate system from the play counter, and this section exists because the two get described as one thing constantly. Spotify pays out of a pooled pot on a per-stream basis, and the figure per stream lands in fractions of a cent — the commonly reported range is roughly a third to half a cent, moving with the listener's market and with whether they are on a paid or an ad-supported plan.

There is also a floor underneath the whole arrangement. A track has to reach at least 1,000 streams within the preceding twelve months before it generates any recorded royalty, and the policy carries a unique-listener condition alongside the raw stream count. Both conditions exist to separate listening from volume, and they apply to every track on the platform regardless of how its streams arrived.

Run the arithmetic before you settle on a tier. At a third of a cent, the streams needed to cover even a modest order sit far above what that order delivers, and streams identified as artificial pay nothing at all. The accurate description of this product is that it moves a public number; a page describing it as an outlay that returns itself in royalties is describing an arrangement the payout rules do not contain.

How do algorithmic playlists decide what to include?

Discover Weekly, Release Radar, Radio and Autoplay all read the same family of signals, and the raw play count is the lightest of them. What the recommendation system measures is what a listener did with a track: whether they saved it, whether they added it to a playlist of their own, whether they came back to it later, whether they reached the end, and how quickly they skipped when they did not.

Completion rate carries the most weight, and skip rate is its shadow. A play that stops at thirty-one seconds reads, in the system's terms, as a track someone left — the counter recorded a stream, and the behavioural read of it is a listener walking away. That is the honest relationship between this product and algorithmic reach: the public number moves, and the signals feeding Radio and Discover Weekly stay exactly where they were.

If algorithmic placement is the outcome you want, the signals that carry it are saves and followers, because each is a deliberate act by a listener rather than a by-product of playback. Even those work in proportion to genuine listening, which is why the order of operations that holds up is content first, deliberate signals second, and play count as the presentation layer over the top.

What is the difference between monthly listeners and total plays?

They are two different measurements and they behave differently. Total plays is cumulative, sits under each track, and only ever climbs. Monthly listeners counts unique accounts that played you at least once inside a rolling 28-day window, sits at the top of your artist profile, and moves in both directions as that window slides forward.

The distinction matters because different readers look at different numbers. A curator opening your profile sees monthly listeners first, since that is the headline figure Spotify puts there. A promoter or a label scout usually reads both, and the relationship between them tells its own story: a large cumulative total against a small monthly figure reads as a catalogue that had a moment, while the two moving together reads as an artist being played right now.

An order here moves the per-track total. It moves the profile headline only to the degree that the delivery arrives from distinct accounts inside the current window, which is why monthly listeners is priced as its own product instead of being folded into this one. If the number you are actually judged on is the one at the top of the profile, that is the page to weigh against this one.

How many plays should I buy for one track?

Fewer than instinct suggests, and sized against your monthly listener count rather than against a round number. Both figures are public and sit a scroll apart, so they get read together: a track showing 400,000 plays on a profile showing 60 monthly listeners is the most legible mismatch on the platform — obvious to a curator, obvious to a promoter, and the exact shape automated review is built around.

A workable band is a few thousand plays against a few hundred monthly listeners, which reads as a track that caught something. As a rough guide, keeping a single track's cumulative total inside roughly fifty to a hundred times your monthly listener figure stays within what happens on its own; past that point the two numbers start explaining each other to anyone who looks.

Spread over three or four releases rather than concentrated on one, the same budget reads better still. A catalogue where several tracks sit in a normal band looks like an artist with an audience, while one track carrying everything looks like exactly that — and the second impression is the one a curator forms in about four seconds.

How long does delivery take, and why is it slower than the rest of the site?

Longer than any other product here, and that is a design decision rather than a capacity limit. Streams are spread over days instead of hours because velocity is one of the heaviest inputs into how listening gets assessed. Twenty thousand plays landing in an afternoon and nothing afterwards is a shape that does not occur outside a paid campaign, and pacing is the cheapest available way to stay clear of it.

The second reason is that Spotify's public counters refresh on their own schedule, generally around once a day. The number you see lags the delivery, sometimes by more than a day, and that gap is the platform's refresh cycle rather than an order sitting still. Checking the track page twenty minutes after ordering will show you the number you started with.

Spotify for Artists runs on a different rhythm again, with its own daily consolidation and a separate handling of same-day figures, so the number there and the number on the public track page can disagree for a day at a time. Both settle to the same place once the cycle catches up, and the public page is the one your visitors are reading.

Is buying plays the same as playlist promotion?

No, and separating the two is worth doing before any money moves. Buying plays adds streams to a track you already control. Playlist promotion means paying someone to place your track on a playlist their followers listen to — a different transaction with a different failure mode, and one Spotify treats differently again depending on whether the placement was disclosed.

Both can end in the same place. A playlist stuffed with paid placements and no genuine listeners produces exactly the stream pattern the artificial streaming policy is written around, which is how artists occasionally end up penalised for a service they understood to be legitimate promotion. The variable that decides it is whether the playlist has an audience that chose to follow it.

If that route interests you, playlist followers and plays are separate purchases with separate mechanics, and neither one substitutes for the other. Plays sit on a track; playlist followers sit on a playlist you own and change how it reads to someone deciding whether to subscribe to it.

What does a stream count change for the person reading it?

It changes a judgement made in a few seconds, in the places where a stream count is read as a credential. A promoter deciding whether to book you, a playlist curator glancing at a track before opening it, a sync agent shortlisting for a brief, a listener who found your name and is checking whether anyone else has heard of you — all of them see the number, and none of them see how it got there.

That effect is real and it is narrow, which makes it easy to size correctly. It works at the moment of a first look and stops working the second someone plays the track, because from that point the music is doing the talking. Measuring the order against how much that specific first impression is worth to you is the calculation that keeps this purchase proportionate.

It is also worth knowing which surfaces show the number. The track page shows a play count; the artist profile shows monthly listeners and a popular-tracks list ordered by recent activity; a playlist shows neither. A booking email with a link, on the other hand, shows whatever the recipient clicks — which is usually the profile.

Does the track have to be public and available in every market?

It has to be public, and it has to be available in the markets the delivery plays from. A release that is geo-restricted, still on a pre-save page, or withdrawn by a distributor cannot receive streams, so an order against one fails at the first attempt rather than sitting in a queue. Failed orders are refunded rather than held.

Territory settings are worth a glance before ordering, because they are set at distribution time and then forgotten. A release limited to a handful of countries narrows the pool of accounts that can deliver, which slows the window and concentrates the delivery in a way that shows in your own dashboard afterwards.

Podcast episodes, local files and unreleased mixes all sit outside this product for the same reason: the delivery needs a public catalogue track it can open by URL. If you are working with a podcast catalogue, that is a separate product with a separate delivery method, priced separately across the Spotify services.

Which link should I paste — track, album or playlist?

The track link, in almost every case. Copy it from Spotify's own share menu, which gives you a clean open.spotify.com URL with the track ID in it. That ID is what the delivery queue reads, so a link copied from a browser address bar with tracking parameters attached still works, and a third-party shortener wrapped around it may not.

An album link spreads the delivery across the tracks on that album, which is occasionally what someone wants for an EP and rarely what someone wants for a single. A playlist link spreads it across everything on the playlist, including tracks that are not yours, so it is almost never the right choice for an order aimed at one release.

If the goal is to lift one song, paste that song. If the goal is to make an EP read as listened-to throughout, an album link distributes the same quantity evenly, which produces a more believable shape than one track carrying a total the rest of the record does not support.

Do plays do more for a new release or for a back catalogue track?

They do different work in each case. On a new release, the number is being read by people who came to check whether the release landed — the first week is when press contacts, curators and your own audience look, and a track that reads as heard gets more of those looks converted into plays.

On a back catalogue track, the number is doing a slower job. An older track with a healthy count is what a new listener finds when they go looking through your profile after hearing one song, and it is what a sync agent or a booker sees when they scroll your popular list. Nothing is time-critical there, so the delivery can be paced as slowly as you like.

There is a practical argument for the back catalogue that most guides skip. Your popular-tracks list is ordered by recent activity, so streams delivered to an older track can change which five songs a visitor is offered first. Deciding which song greets a new visitor is a real editorial choice, and this is one of the few levers you have on it.

How do plays sit next to saves and followers?

They occupy different positions in the same picture. Plays are the volume number; saves are the intent number; followers are the return number. A profile where all three move together reads as an audience, and a profile where one moves alone reads as one number moving alone, which is a distinction any experienced curator makes without thinking about it.

In practical terms, that argues for proportion rather than for buying everything. A track with several thousand plays and a handful of saves is a normal ratio for a song people liked but did not keep; the same track with more saves than plays is a shape that does not occur. Saves generally land somewhere between two and ten per hundred plays on tracks that are working.

Followers move on the slowest clock of the three, because following an artist is a commitment a listener makes after more than one song. That is why a follower count that jumps while the catalogue stays quiet is the most conspicuous of the three, and why followers are usually the last thing to add rather than the first.

What does Spotify for Artists show about where streams came from?

It shows a source breakdown, and it is worth opening before and after any order so you know what you are looking at. Streams are grouped into buckets: Spotify's editorial playlists, algorithmic playlists like Discover Weekly and Radio, listeners' own playlists, a listener's own library, your artist profile, search, and an "other" catch-all for external links.

Delivered streams show up in the buckets that match how they were played, which in practice means the profile, search and other categories rather than the playlist and library ones. That is a useful thing to have seen with your own eyes, because it makes the difference between the counter and the recommendation signals concrete instead of theoretical.

The dashboard is also where a shortfall becomes visible. If a delivered figure is adjusted afterwards, the daily chart shows it as a correction on the day it happens rather than as a notice, and knowing to read it that way saves a support ticket. Nothing about it is hidden — it is simply reported as numbers rather than as messages.

How should a budget be spread across a catalogue?

Weight it toward the tracks that get opened. For most artists that means the three songs at the top of the popular list, the current single, and whichever older track keeps appearing in your listener sources — those five carry the majority of the first impressions your profile makes in any given month.

An even spread across twenty tracks produces twenty small numbers and changes nothing about how the profile reads, because a visitor scans the top of the list rather than the whole catalogue. A single large order on one track produces one number that explains itself. The shape that works sits between the two: three or four tracks in a believable band, with the current release slightly ahead of the others.

The same logic applies to timing across a release cycle. Money spent in the week of a release does presentation work while attention is on the profile; money spent six months later does maintenance work on a catalogue people arrive at from somewhere else. Both are legitimate; they are simply buying different things.

What happens to the number in the weeks after delivery?

It settles, and settling can go both ways. Public counters catch up with the delivery over the first day or two, and platform-side adjustments can revise a figure downward later, which appears as a quiet correction rather than as a notification. Planning around the delivered number rather than the peak number is the realistic way to think about it.

Beyond that, the count is static unless real listening continues. The delivery is a one-off event, so a track sits at whatever total it reached and then moves only at the rate its actual audience moves it. That is the honest shape of the outcome: a step, not a slope.

Monthly listeners behaves differently again, because its 28-day window slides. Whatever contribution a delivery makes to that figure ages out of the window within a month, which is worth knowing before reading a decline as something having gone wrong. The window emptying is the window working.

What separates one supplier from another?

Four questions, and the answers are more informative than any price comparison. What is the play duration distribution — a flat thirty-one seconds, or a spread with real completions in it? Over how many days is the quantity paced? From which markets do the accounts play? And what happens to the order if the track is unavailable when the delivery starts?

A supplier who answers all four is describing a production process. A supplier who answers none of them is reselling whatever is cheapest that week, which is also why the same order can behave differently a month apart. Price differences between tiers usually trace back to the first two answers, since both cost time.

The fifth question is about the order itself: what the refund position is when a delivery cannot start, and how you check what has been delivered. Both are stated above the picker on this page for exactly that reason, and the tier table shows the pacing window alongside the quantity rather than only the price.

What an order changes, and what the music carries

It changes the number a person reads before they press play. That is a genuine effect at a genuine moment — a curator with forty tracks to get through, a promoter with a shortlist, a listener who arrived from a link and is deciding whether to spend three minutes. The counter answers the question "has anyone else heard this?" and it answers it instantly.

What stays with the music is everything after that press of play: whether someone reaches the end, whether they save it, whether they add it to a playlist they made themselves, whether they come back next week. Those four behaviours are what the recommendation system reads and what compounds into an audience, and no counter reaches into any of them.

So the accurate way to size this purchase is as presentation rather than as growth. It puts a public number where a first impression is formed, on a track whose job starts the moment the number stops being read. Every tier above is priced on that basis, with the delivery mechanics, the pacing, and the platform's own position stated on this page rather than left for you to find out afterwards.