Due to how artificial activities can compromise their charts, recommendations and royalty systems, Spotify has begun aggressively filtering out any fake streams from Spotify. The terrifying thing for an artist is that you will get flagged without ever intending on doing anything wrong; a lot of times this is due to some promo vendor using these types of bots or through a sketchy playlist network.
When it comes to finding out if someone is trying to get around the Spotify system by having manufactured play counts (aka fake streams), the short answer is that Spotify tracks usage behaviour from one representing account to the next (across devices, locations & sources) for data associated with non-realistic characteristics which would indicate possible fraudulent activity. When Spotify's systems determine that certain streams or plays were created through surrogate means, these streams may be invalidated, withdrawn from the respective playlist's position(s), and/or trigger distributor alerts/warning notifications to the account representing the streams or plays.
Many musicians turn to PromosoundGroup to help achieve a transparent, accountable approach to their marketing strategy through authentic engagement by their fans. Instead of simply seeking evidence of more plays as proof of success, we also strive for total view counts to continue appearing as natural or "live" engagement, using analytics reports as verification for both types of interactions.
1) Clear Meaning (No Fluff)
Generated or incentivized, false streams have falsified the true demand for users. Different types of streams can come from bots, click farms, account farms or "always-on" playbacks set up to hack a track. It is not unusual for an occasional spike in streams (like from an influencer posting to a new country) to occur. However, a huge spike with no corresponding growth in genuine engagement (like followers or saves), is a major issue.
2) Where / When the Problem Appears
Most artists notice the issue in one of three places:
● Artists on Spotify: unpredictable increases in streams, a new number one country that doesn't completely add up, and one unsupported playlist driving all of your streams.
● Distributor emails: warnings about artificial activity or royalty holds.
● You've had many streams since your release, but it hasn't resulted in growth in monthly listeners, number of followers, the total amount of songs saved by your fans or received any DMS on social media.
Context is important. A legitimate editorial playlist could see an initial spike, as could the viral nature of a Tik Tok video or being featured in a highly regarded blog post. A fast spike in the numbers of someone or something that we know is not legitimate, e.g., anonymous playlist with no type of brand presence or valid saves; having questionable geo information would be where to begin diagnosing what might be a legitimate play through fraud.
3) Causes - Grouped by Category (VERY IMPORTANT)
A) Listening-pattern anomalies (behavioral signals)
When people inquire about how Spotify pinpoints phony streams, this "bucket" (for example) will often be the base. Genuine users do not act like automated bots. Spotify also recognizes metaphors of listening patterns which suggest clusters of sessions that appear to be the same mechanically.
● Looping one track repeatedly with little variation (especially 24/7 behavior).
● Unnatural completion or skip patterns (for example, huge volume with identical listen durations).
● Streams that have an increase in views, but there are no downstream actions such as saves, compression to playlists, follows, and visits to profiles.
B) Device, IP, and account fingerprinting (infrastructure signals)
With Spotify's data, activity can be tracked to different device types, IP addresses, history of accounts, and login characteristics. Bot farms typically don't change their set up; therefore, they usually have many of the same devices, networks, and user behaviours as an audience does.
● Many accounts streaming from the same IP blocks or suspicious hosting providers.
● Emulated devices or abnormal client behavior (automation footprints).
● Accounts that lack a decent amount of regular activity away from their target (no actual searching, library building, or variety).
C) Playlist and source anomalies (traffic provenance)
Many fake Spotify streams are generated through questionable playlists where scams can be concealed from the service's monitoring. Spotify analyzes what type of traffic each playlist generates, as well as how the playlist behaves like legitimate playlists.
● No-name playlists that suddenly deliver the majority of your streams.
● Playlists with thousands of tracks, random naming, and no clear curation logic.
● Really large "playlist" streams for many of my songs but they were not helped by any significant algorithmic downstream from either radio or the discover weekly playlist after that.
Spotify is stating that this topic contains artificial activities according to the official guide such as different forms of manipulation and ways to enforce against them.
D) Promotion-market causes (vendor-driven risk)
It's common for an artist to have purchased "promo" services which may have used bots without the artist's knowledge...there are also many instances where an artist has been punished for using dubious playlist offers which they did not know were in place. An example of this can be found at the discussion thread Reddit for what this looks like in the real world.
4) Decision Tree / Diagnostic Block (MANDATORY)
● If you suddenly awoke to find that you had gained between 20 and 50,000 streams from a country of which you had never before targeted - it is highly probable that this was the result of either bot traffic OR (and especially) farm traffic using a playlist network.
● When there is an increase in streams while the number of saved albums, followers and profile visits remains unchanged, it indicates that there is a high likelihood that these streams came from low-quality or artificially generated (fake) listenings instead of true discovery.
● The main reason there would be an unknown playlist with the majority of spins through streaming services would have to do with placement on a risky playlist, rather than through organic algorithm growth.
● In the event that you receive a warning from your distributor, or your royalties are being withheld, Spotify may have already filtered any possibly deceptive or fraudulent traffic and your release has undergone an investigation.
● If there is an increase in search volume related to a real instance (press feature, creator post, live performance)...then you should check to see if the engagement rate corresponds with that real event to support that this is valid traffic.
5) Fix Checklist (Actionable, Ordered)
- Verification of source - To see how individual playlists or sources are contributing to overall streams within Spotify for Artists (playlist, algorithm, or listeners' own library), take screenshots to keep track and make notes.
- Look at the ratio of engagement: compare streams to saves, follows and playlist adds for the same period. A large number of streams and a small number of saves indicate warning signs.
- Geography of auditing: Review the leading nations and cities by size and scope. Determine whether or not you conduct business or advertise to those locations. If you don't, be cautious until you learn otherwise.
- To identify the unknown driving volume, you need to find it on Spotify. Then, review whether there are any identifying characteristics such as an identifiable curator, ownership of the playlist as a brand, social media accounts, and prior recorded history (e.g., previous playlists). You will also need to assess whether the playlist appears to be a "content farm"; if so, proceed with caution when partnering with this type of playlist.
- Stop All Paid Advertisements Immediately: If a campaign is currently running and an "uptick" occurs, stop it immediately. Continuing an active campaign will only continue to change the pattern and increase the length of the filtering process.
- 2020-09-30 Remember to email the distributor the vendor list with the traffic source(s) and method(s) used by each vendor. It is a big red flag to guarantee a specific number of streams. Save all of this correspondence for your distributor.
- Proactively notify your distributor: Contact customer support if you think you have had fake streams on Spotify, before they reach out to you. Provide all evidence including screenshots, timestamps, and vendor receipts.
- When attempting to lower the bad traffic rating from your ad campaign, adding new campaigns to cover the bad traffic may increase the amount of bad data created and can have a negative impact on your analytics reporting.
- Observe activity (via Spotify) for 7 to 14 days from Monday through Friday: Spotify monitors for suspicious activity on a continuous basis. Observe the following: does the spike in activity disappear? Have any new sources of activity emerged? Are your playlists normalized?
- Revamp your growth strategy by focusing on realistic (measurable) methods of engaging with your audiences, such as content-based exploration; targeted advertising with real audiences; collaborating with creators; and carefully selecting true curators when you pitch. If using a service, look for transparency in regards to the quality of their audience rather than just quantity.
6) What NOT to Do (Trust Section)
● Avoid purchasing packages that are marketed as "guaranteed" for streaming. The word guarantee usually refers to some kind of automated or controlled method of delivery, which is being filtered out on the Spotify network.
● You should not use your individual looping configurations on multiple devices to "experiment" with them. This could result in creating the same creating signals on Spotify that are used to indicate manipulation.
● Assume nothing, as the spreading out of bots internationally does not circumvent the detection of their behaviour. No engagement and random geographic activity is also a pattern that can be detected.
● Do not disregard warning emails from those who distribute. By not responding or providing any feedback, the ability of the distributor to take action quickly and harshly increases.
How to Spot Fake Streams on Spotify in Your Own Stats
A real way to identify Spotify's fake streaming is by checking for discrepancies in the traffic to your profile - streams being generated by people who aren't listening to your music as expected.
● Warning sign: One playlist receives the majority of plays despite its lack of being an editorially driven or recognized by a curator or branded.
● Warning sign: A country appears from nowhere to take the top prize while no activity occurs on your social media or email list!
● Red flag: Streams jump but monthly listeners, followers, and saves do not rise proportionally.
● Warning sign: If you see that your first track has changed dramatically for one or two days only to be restored to its original state without cause you may want to proceed with caution.
Red Flags in "Spotify Promotion" Offers
● They promise "10,000 streams in 3 days" for a fixed price.
● The sources of listeners are not identifiable when they listen, such as via advertisements, creator content, owned audiences or curator lists.
● They brag that it is "100% safe" while implying automation or "our network" with no details.
● They show only stream screenshots, not saves, followers, listener retention, or playlist diversity.
How to Grow Without Triggering Spotify's Fraud Systems
● Pitch responsibly and as soon as possible - put your pitch through the Spotify for Artists pitching feature so that every genuine increase in traffic can be traced back to its source.
● Generate authentic intent with Short form content that drives people to look for you delivers better signals than random increases in playlist numbers (e.g., a 50% increase in playlist streams due to random playlist placements).
● Leverage advertising for targeted visits: Use other artist's audiences to target ads with click-through tracking (as opposed to click-through rates to measure success) towards saving and following on Spotify.
● Features and remix swaps can foster natural cross-pollination creating a more realistic geo pattern.
● Reach out to curators for information about their audience and any anticipated benefits from the playlist (e.g., saves/follows, future listeners).
FAQ: Short Answers About Spotify Fake Streams
Spotify has several methods to detect streaming from fake accounts. Listening patterns, device types, and anomalies with the account and content source are analyzed across playlists, traffic, and listening activity. This makes use of different types of analysis (i.e., pattern analysis) rather than one single metric.
If I purchase counterfeit Spotify streams by mistake, will I face a ban? This doesn't necessarily mean a user will be banned for such behaviour as the streaming platform filters, so your distributor might simply provide you with a warning. The correct thing to do is stop the source immediately, document the event, and speak with your distributor about what transpired.
Can other artists or businesses send fake streams to my song to hurt me? Theoretically, this could occur, however, as you're probably aware, Spotify measures monthly filtering and will likely disregard any abnormal activity as suspicious once they aggregate data across multiple users - so there's no guarantee that this would happen. Nonetheless, if you notice an inexplicable increase in streams, you'll want to document this occurrence and consider it a risk event.
An indication of an unsafe service is its promotion of guaranteed fixed numbers through "organic" growth. Fixed-number guarantees are very good indicators that delivery method is standard and that the delivery method is a factor in producing fake streams.
7) Summary (No New Info)
When monitoring for artificial activity on its platform, Spotify employs various methods to examine non-human patterns in listening behavior, in addition to examining device/account fingerprints, as well as identifying suspicious playlist sources. If you discover there is any sudden geo spike activity, large amounts of unknown playlists dominating your data, no legitimate saves or follows on your profile from users who listen to your music, you should: 1) Be cautious of risk associated with your policy on this type of use; 2) Stop your promotional activity until further notice; 3) Document (or retain) any data from where the download source originated; 4) Notify the person/functional area of your distributor who manages your account; 5) Consider attempting to remove unnatural growth patterns in your statistics. The safest, least risky type of growth is organic growth. Organic growth typically results in greater levels of engagement with listeners - which is what Spotify requires to justify paying for copies of your music that have been purchased by legitimate consumers.
