Why Did My Spotify Monthly Listeners Drop?

Usually because a release is aging out of the 28-day window, and a drop of 20% or more is common enough that 7.1% of artists in our data had one in a single month. What matters is what it looks like next month, because 88% of those big drops stayed big.

After 20 years, 19 Billboard Top 20 albums and catalog that's crossed 3.3B+ streams, I've watched a lot of artists open Spotify for Artists, see the line fall, and decide something is wrong with them. Most of the time nothing is wrong. Sometimes it's the most useful number they'll see all year. I wanted to know which is which, so I measured it.

What a Normal Month Looks Like

I took every pair of readings in our artist database that sat 26 to 34 days apart, starting from at least 100 monthly listeners. That's 176,770 month-to-month changes across 109,248 artists. The median change was 0.0%. Here is how it spreads out by size at the start of the month.

Read your own row. If you're at 3,000 listeners and you're down 12%, that's a Tuesday. One artist in five at your size had a month like it. The size effect is just averaging: a few thousand people can be moved by one playlist coming and going, and a few hundred thousand can't.

When a Drop Is Real

A drop of 10% to 20% is inside the ordinary spread. A drop of 20% or more at 5,000 listeners or above is not rare, but it is the part of the table that's worth a second look.

So I looked at what happened after. For the artists who had a 20%-plus drop and another reading about a month later (686 of them):

That bottom row is the one I keep rereading. A 20%-plus drop does not bounce. Nearly three in ten kept falling by another 10% or more. If I narrow it to 2026 alone (162 artists) it says the same thing: 5.6% back to within 10%, 88.9% still 20% below. That sample is small, which is why I'm not leaning on it, but it agrees.

So waiting is not a plan. A big drop means the thing that was feeding you stopped. It might be a playlist you were on, a post that ran its course, or a release cycle ending. The listeners come back when something new gives them a reason.

Days 31 to 60 After a Release

This is the pattern that surprised me most. I matched 15,841 monthly changes in 2026 to the date of each artist's newest release, but only where that release was already on the record by the end of the month, so I knew whether a new song landed inside it.

The cliff is the second month. The month after a release, 27.5% of artists lost 10% or more and 14.8% lost 20% or more, close to triple the 4.9% of artists whose newest release was over a year old. By days 61 to 120 it's back to a normal rate. The 28-day window is doing exactly what it says: the people who showed up for the song leave the count about a month after they last played it.

It's also the answer to the question in the headline. If you released something about six weeks ago and your number is falling, that isn't a verdict on the song or on you. It's the shape of a release. I wrote about the same cycle from the other side in how often you should release music: artists who stay quiet decay, artists who release don't, and the gap is a month of attention, not a miracle.

What I'd Actually Do

I'd stop checking it daily. A daily number makes every dip feel like an emergency, and the table says one month in five is a dip at small sizes.

Then I'd ask a harder question than why the number fell. Of the people who came, did anyone stay? Followers hold when listeners wander off, and I laid that ratio out in why your song gets streams but no fans. And if the dip is real, the fix lives in the record, not the dashboard. A song people replay is a production question, which is why the cost of mixing and mastering belongs in the plan next to the release date. A producer can tell you in an afternoon whether the next release has what it takes to keep a listener past day 30.

When you're ready to price that honestly, the rate calculator is calibrated to where you actually are as an artist, not to a number built for somebody else's catalog.

I don't promise a drop reverses, or that a release fixes it. I just don't like watching people panic over a line that does this to everyone.

Method

Source: the Freshly Baked Studios artist database, populated from Chartmetric. Spotify monthly-listener readings were cleaned from text fields and cast to numbers. A monthly change is the percent difference between two consecutive readings 26 to 34 days apart for the same artist, starting from at least 100 listeners (176,770 pairs, 109,248 artists). The follow-up table uses pairs that also had a next reading 26 to 34 days later (686 after 20%+ drops, 979 after 10 to 20% drops, 8,248 in total); the 2026-only check used readings dated January 1, 2026 onward with a next reading 20 to 45 days later (162 after 20%+ drops). The release table uses 2026 pairs (15,841) joined to each artist's newest-release date, kept only where the record was refreshed at or after the end of the month and the release date was on or before it, so a release inside the month counts as inside. Only the newest release is stored, so earlier releases inside a month are not visible. Chartmetric skews toward artists with a public contact route, so the very bottom of the market is under-represented.