You cannot game the algorithm

Every few months someone sells a way to "beat" Spotify. A playlist scheme. A save-to-library ritual. A bot farm dressed up as a growth agency. The pitch is always the same: the algorithm is a puzzle, and they have the answer.

Glenn McDonald spent years as a Spotify data scientist. In a Billboard India interview with Mohini Chaudhuri (16 September 2026), he is blunt. You almost certainly cannot game it. If you think you did, that is often worse than failing, because you will spend the next few years repeating a coincidence.

His book, You Have Not Yet Heard Your Favourite Song, is about how people actually find music. This post is the version of that conversation we would give an artist, a manager, or a label running work in Aden. It is not a recap of Billboard's article. Read theirs too. Link at the end.

Can you game the Spotify algorithm?

No, not in any way you can trust.

Platforms change ranking constantly. A tactic that "worked" last quarter is usually a story you told yourself after a spike. McDonald's warning, as he told Billboard India, is about false attribution. You pitch a mega-playlist, streams jump, you assume the pitch caused it. Maybe a radio add landed the same week. Maybe a support slot. Maybe one real fan shared the track in a group chat you will never see.

If you treat that jump as a formula, you will copy the wrong move for years.

Time is better spent on work you can inspect: writing, playing shows, talking to the people who already come back. Those leave a trail. Algorithm folklore does not.

Why unique artist names matter for search

Discovery is not only recommendation. A lot of it is someone typing a name they half-remember.

Generic names collide. If you are "Blue Sky" or "The Sessions", search has to guess which of twenty acts the listener meant. McDonald's practical advice, via that Billboard India piece, is to spend a few minutes on uniqueness. Check the name on the platforms you care about. Check whether a domain is free. If you can still rename, do it before the first proper release, not after the merch is printed.

You do not need a joke name. You need a name that points at you.

Playlists are not magic

A playlist is a room. Fit the room.

Broad new-release dumps and mega-genre lists look impressive on a one-sheet. They also attract skippers. Skips teach the system that your track is not for the people who just heard it. McDonald's point, as reported by Billboard India, is that the wrong playlist can feed the wrong Release Radar neighbourhood. You get a week of numbers and a longer stretch of listeners who will never come back.

Prefer playlists that already serve your listeners. A 2,000-follower list run by someone in your scene is worth more than a 500,000-follower "New Indie" hopper that chews through 80 tracks a Friday. Pitch less. Pitch closer.

If you run a label, stop treating playlist adds as the campaign. They are one distribution surface. The campaign is still the shows, the mailing list, the radio, the community that will play the next song without being paid to.

Why buying bots is a waste

Do not buy bots. Platforms look for this. They punish it.

McDonald's split, as told to Billboard India, is useful:

• Fake streams / bot farms. Fraud. Detection exists. Artists and labels get the accounts and the numbers taken down. The people who sell you the "growth" keep the fee. They are selling shovels.
• Fan looping. Existing fans repeating a track can nudge the stats a bit. That is not the same as buying a stranger's script to inflate a new artist. Do not confuse loyalty with a loophole.

If a service promises chart positions, "algorithmic boosts", or a guaranteed playlist network in exchange for a wire, walk away. The worst case is not wasted money. It is a strike on the catalog you spent years making.

Never buy bots. Never let a distributor, plugger, or intern do it "quietly" on your behalf. Quiet does not exist on a log file.

Should multilingual releases use separate artist profiles?

If you release the same songs in more than one language, separate artist profiles per language can help people find the right version.

McDonald's tip, in that Billboard India conversation, was aimed at Indian artists working across languages. It travels. A Hindi profile and an English profile keep search, playlists, and recommendation neighbourhoods from mixing two audiences that do not share a catalog. The listener who searched in Tamil should land on the Tamil recordings, not a stack that starts with the English mix.

This is metadata work, not a hack. Same songs, clear identities, fewer wrong clicks.

What to spend time on instead

The algorithm is not a person you can charm. Listeners are.

Write the next song. Play the rooms that already know you. Keep a list of who actually showed up, who playlisted you because they mean it, which radio person replied, which support slot converted. That is slow. It also compounds.

Aden is for that trail, not for phantom ranking tricks. Use the calendar for shows and release dates you can keep. Use campaigns and outreach for the people you are actually talking to. Keep versions, splits, and the release itself in one place so a spike can be checked against something true: a date, a city, a mailing, a playlist that fits.

If a week looks magical and you cannot name the humans behind it, do not build a religion around it.

Source

Mohini Chaudhuri, "Can Artists Game The Algorithm?", Billboard India, 16 September 2026. Interview with Glenn McDonald, former Spotify data scientist, around his book You Have Not Yet Heard Your Favourite Song.

Further reading: the Billboard India article above. This post is Aden's take for independent artists, managers, and labels. It does not reproduce Billboard's reporting.

FAQ

Can artists game the Spotify algorithm?

No. Ranking changes constantly, and a spike you cannot explain is usually coincidence. Treating it as a formula wastes years. Write, play shows, and learn who actually listens.

Do unique artist names help discovery?

Yes. Search has to tell you apart from every other act with a similar name. Check platforms and a domain before the first proper release if you can still rename.

Should I pitch every big playlist I can find?

No. Broad new-release and mega-genre lists can train the wrong audience, including a worse Release Radar mix. Prefer playlists that already serve your listeners.

Does buying bots or looping streams work?

Buying bots is fraud. Platforms detect it and punish catalogs. Fan looping by real listeners is not the same thing, and it is not a growth plan for a new artist. Never buy bots.

Should multilingual releases use separate artist profiles?

If you release the same songs in more than one language, separate artist profiles per language help search and recommendation keep those audiences apart.

What should I spend time on instead of algorithm hacks?

Songs, shows, and the people who already come back. Track real relationships, dates, and outreach. Do not build a process around a spike you cannot name.