The single goes into a big playlist on a Thursday. By Monday the monthly listener count has tripled and somebody has already screenshotted it. The recap deck gets a slide titled "Audience Growth."
Six weeks later the number has drifted back almost to where it started. Saves were modest. The next release opens to roughly the same room as the last one.
Nothing dishonest happened. The first report just treated two different things — being heard and being chosen — as one metric, because for years the platform only really gave you one number to put in a slide.
Spotify's public monthly listener count covers unique people who heard the artist in the previous 28 days, whether they went looking for the music or it arrived in a Daily Mix while they were cooking. That makes it a reasonable measure of reach. It says nothing about how the listening started, whether anyone came back on purpose, or whether the artist now has a more durable audience than they did in August.
Spotify for Artists gives you better instruments than that now. The work is reading them without turning them into another scoreboard.
Start with where the stream came from
Spotify splits sources into two categories, and the split is behavioural rather than moral.
Active sources are the ones that required a decision. Spotify lists three: the artist profile and catalog — playback from the profile, album pages or the artist's "This is" playlist; the listener's own playlists and library — their saved songs, saved albums, their own playlists; and the listener's queue.
Programmed sources are where Spotify or another person put the music in front of someone. Spotify groups these as editorial and personalized editorial playlists including charts; personalized playlists, autoplay and mixes — Discover Weekly, Release Radar, Radio, Autoplay, Daily Mix, daylist and AI DJ; and other listeners' playlists.
Programmed doesn't mean worthless. It's how most people meet most artists, and a listener who finds a song in an editorial playlist on Tuesday can be on the artist profile by Friday. The useful question isn't which kind of listening is better. It's how many people move from one to the other, and what seems to make that happen.
One practical warning from Spotify's own documentation, because it catches people building spreadsheets: when you look at sources by listeners rather than streams, the subtotals can add up to more than your total listener count. One person using three sources is counted in three places. Any percentage you build on that denominator will be wrong, and wrong in a flattering direction.
Monthly listeners is a window, not a fan count
The public number is a rolling 28-day window. It rises on editorial placement, algorithmic pickup, a collaboration, an ad campaign, a video that travelled, or somebody rediscovering a four-year-old album. When whatever caused the plays stops causing plays, those people fall out of the window.
Which is why a peak monthly listener figure is a poor standalone measure of anything retained. It's a record of how wide the door was open, not of who walked in.
Inside Spotify for Artists, the total audience is a different view: unique listeners from the previous two years, divided into three groups.
- Monthly active listeners intentionally streamed the artist from an active source in the past 28 days.
- Previously active listeners were in that group once but haven't intentionally streamed for at least 28 days. They may still be hearing the music through programmed sources.
- Programmed listeners have only ever streamed through programmed sources, with no active source for this artist in at least two years.
Monthly listeners and total audience answer different questions. One is recent reach. The other is a longer behavioural record of everyone who has encountered the artist.
Neither is a mailing list. Neither is a customer file. Neither gives the artist any way to contact those people off the platform, which is worth saying out loud in rooms where somebody is about to describe 400,000 monthly listeners as "our audience."
Read the active audience in layers
Monthly active listeners are split again, by intentional streams in the last 28 days:
- Light — one or two
- Moderate — three to fourteen
- Super — fifteen or more
These are descriptions of behaviour inside a rolling window, not permanent labels for human beings. A super listener can go quiet because they changed jobs. A programmed listener can become active the week a tour is announced. Someone who genuinely loves the artist might do most of their listening on Apple Music and show up here as light, or not at all.
Spotify does publish two aggregate figures about the top tier, and they're worth knowing precisely because they get misquoted. Averaged across artists of all sizes, super listeners are about 2% of an artist's monthly listeners but drive over 18% of monthly streams, and one super listener streams as much as 20 programmed listeners.
Those are Spotify's cross-artist averages. They're a good argument for paying attention to the segment. They are not a coefficient for your own catalog, and they do not belong in a revenue projection — multiplying your super listener count by anything and presenting the result as forecast income is how a measurement exercise turns back into a vanity exercise with extra decimal places.
Build a four-part baseline before the campaign moves anything
Capture it before release week changes the shape of everything. Same weekday, same reporting window, every time.
- Reach — monthly listeners, new listeners, programmed listeners
- Intent — monthly active listeners, active-source streams
- Depth — light, moderate and super listeners, plus streams per listener where it's meaningful
- Retention — previously active listeners, reactivated listeners
Next to the numbers, write down the conditions: release date, playlist adds, what was pitched and when, paid campaign start and stop dates, social posts that did something, collaborations, tour announcements, any press worth mentioning.
One detail that causes quiet confusion in international teams: Spotify records analytics in UTC. A label in Los Angeles reconciling platform days against a local ad calendar will produce mismatches it can't explain three months later, usually in the direction of making one day look better than it was. Keep the timestamps and write down which timezone the report is in.
CREWPORT's guided release workflow helps organize release timing, assets and delivery information so the campaign record is consistent and the dates in your baseline are the dates things actually happened. The audience interpretation lives in Spotify for Artists — we don't see or supply those segments.
Diagnose the shape, not just the direction
Five patterns, each pointing somewhere different.
Monthly listeners up, programmed listeners up, active listeners flat. The recording got exposure. So far there's no comparable growth in intentional listening. Look at which programmed sources moved, and check again in two or three windows before deciding — active movement often lags.
New listeners and new active listeners rise together. More people met the artist and some of them chose to come back. The follow-up question is whether they settle into light, moderate or super over the next few windows, or drift out.
Previously active down, reactivated up. Lapsed listeners are coming back. The timing might point at a campaign touchpoint; it doesn't prove one, and Spotify's own recommendation systems were running the whole time.
Super listeners rise while total reach stays modest. The release is deepening a small audience rather than widening it. That can be worth more commercially — tickets, direct sales, the next release opening better — but Spotify behaviour on its own doesn't demonstrate any of those outcomes.
Monthly listeners rise then fall after a playlist ends. Expected. Temporary reach leaving a rolling window is arithmetic, not failure. What matters is what's left behind in the active and previously active segments once it's gone.
Measure movement, not just segment size
Spotify reports listener-conversion measures — new listeners, new active listeners, reactivated listeners. Read them as flows rather than totals:
- New listener rate — how much previously unreached audience entered during the period
- Active conversion — how many listeners joined the monthly active segment, without assuming one campaign did it
- Reactivation — how many previously active listeners returned to intentional listening
Resist inventing a conversion percentage when the interface doesn't give you a compatible denominator. The source-subtotal problem above is the main trap, and it's an easy one to fall into because the resulting number always looks encouraging.
Export the reported measures under their original names. A home-built ratio can be genuinely useful internally — just label it as your calculation and write down the denominator somewhere the next person will find it.
Run a controlled test after the playlist ends
The interesting window usually starts when the spike stops.
Pick one release with real programmed reach. Over the following four to eight weeks, give those listeners a reason to make another deliberate choice: update the artist profile so it isn't eighteen months stale, connect the track to a catalog song that actually suits it, put out a performance asset, talk directly to the audience the artist already has, make the next release feel like it belongs to the same story.
Change as few things at once as the schedule allows. Log every intervention with a date.
Then compare:
- programmed listeners entering the period
- new active listeners
- movement from light into moderate or super
- reactivated listeners
- active-source share
- what's retained once the original programmed source declines
This is observation, not a randomized experiment. Spotify's recommender, unrelated user playlists, a support slot, a sync placement and a piece of press can all move the same listeners during the same eight weeks. The goal is to find out whether the artist can hold on to some portion of exposure, not to credit one Instagram post with a segment shift.
A hypothetical two-single comparison
A label puts out two singles from the same developing artist, three months apart.
The first gets a substantial playlist lift. Monthly listeners triple; most of the increase is programmed. The team celebrates the peak, posts the screenshot, and does little once the placement ends. Active listener growth is limited and the public number contracts over the following windows.
For the second single they record the baseline first and prepare a follow-through sequence — a profile refresh, a live performance linked from social, a catalog playlist built around the new track — and log each date as it happens.
Suppose the second campaign reaches fewer monthly listeners but produces more new active and reactivated listeners, followed by growth in moderate listeners.
That wouldn't prove the follow-through caused the movement. It would show the second release built a differently shaped audience, which is a result worth testing again rather than a conclusion.
No figures or commercial outcomes are assumed in that example. It's a shape, not a forecast.
What it costs
The segment analytics come with Spotify for Artists; Spotify hasn't announced a separate analytics subscription for them. The expense is attention — keeping baselines, exporting consistently, explaining it when a definition changes, and joining platform data to what the team actually did.
Spotify's paid campaign tools can target particular audience groups where the artist, release and market are eligible. That's a separate budget conversation. Don't buy Marquee or Showcase because a segment exists on a screen; decide first whether the objective is growth, reactivation or depth, and then judge the result against that objective rather than against monthly listeners.
The whole workflow fails quietly when reporting becomes decorative. If nobody will change a campaign, a content plan or a release sequence because of the audience shape, more screenshots aren't worth the hours.
Who should hold off on conclusions
Wait before interpreting when the artist has very few listeners and the percentages swing wildly on single-digit movements; when the window contains a major collaboration; when paid and organic activity are tangled together; when the source breakdown is incomplete; or when someone is comparing two 28-day periods that don't line up.
Be careful with cross-artist benchmarks, too. Genre, catalog size, release frequency, audience geography and playlist exposure all change the segment mix, and a comparison across two of those is mostly noise.
Two things not to say, because both go past what the data supports: programmed listening isn't "fake," and a super listener isn't a confirmed paying customer.
FAQ
What exactly are monthly listeners?
Unique people who streamed the artist in the previous 28 days, from any source — active or programmed. It's a rolling window, so it contracts when a source of plays stops.
How is total audience different?
It covers unique listeners from the last two years and splits them into monthly active, previously active and programmed. Longer view, behavioural rather than promotional.
What counts as an active source?
The artist profile and catalog including the "This is" playlist, the listener's own library and playlists, and the queue.
What counts as programmed?
Editorial and personalized editorial playlists including charts; Discover Weekly, Release Radar, Radio, Autoplay, Daily Mix, daylist and AI DJ; and playlists made by other listeners.
How many streams makes a super listener?
Fifteen or more intentional streams in the last 28 days. Moderate is three to fourteen, light is one or two.
Is 2% / 18% a number I can use for my own artist?
It's Spotify's average across artists of all sizes, published as context. Use it to justify the attention, not as a multiplier in your own model.
Why don't my source percentages add to 100?
When viewed by listeners rather than streams, one person can appear under several sources. Spotify flags this explicitly. It isn't a usable denominator.
Are these segments a fan list?
No. They describe listening behaviour on Spotify. They don't identify anyone, don't travel across platforms, and give the artist no way to contact anybody.
Report what changed in behaviour
A good release recap is shorter than most of them are. Where did the reach come from. How much intentional listening followed it. Which listeners came back. Whether anything deepened after the programmed exposure faded. What the team wants to test next.
Monthly listeners still belong in there. The number honestly describes how many people heard the music in a four-week window, which is a thing worth knowing.
It just shouldn't be asked to carry the rest of the story on its own. The audience gets easier to read once the report separates being played from being chosen — and then checks whether the choosing happened twice.
CREWPORT keeps release timing, assets, identifiers and delivery information consistent so the campaign record your analysis sits on is accurate. We don't supply Spotify audience data or interpret it for you — that reading happens in Spotify for Artists, with your own campaign log beside it.
Note: this is general operational guidance, not a performance forecast. Spotify defines and can revise its own segment and source definitions, and the 2% / 18% and 20-to-1 figures are company-published cross-artist averages rather than predictions for any individual artist. Audience segments describe behaviour on Spotify only; they establish no cross-platform identity and no causal attribution between a campaign action and a listener's movement. Confirm current definitions in Spotify for Artists before building a process on any threshold here.
Sources
- Spotify for Artists — Audience segments on Spotify
- Spotify for Artists — Source of streams
- Spotify for Artists — The complete guide to audience segments
- Spotify for Artists — Super Listeners: your guide for developing fans who go deeper