www.music-tomorrow.com Open in urlscan Pro
52.17.119.105  Public Scan

Form analysis
3 forms found in the DOM

/search

<form action="/search" class="search-nav w-form"><input type="search" class="search-input w-input" maxlength="256" name="query" placeholder="Search our archives" id="search" required=""><input type="submit" value="Search"
    class="search-button-2 w-button"></form>

Name: wf-form-Newsletter-Form — POST https://music-tomorrow.us18.list-manage.com/subscribe/post?u=20898f433ad075354712e41a6&id=dced1dd226

<form id="wf-form-Newsletter-Form" name="wf-form-Newsletter-Form" data-name="Newsletter Form" action="https://music-tomorrow.us18.list-manage.com/subscribe/post?u=20898f433ad075354712e41a6&amp;amp;id=dced1dd226" method="post" class="newsletter-form"
  data-wf-page-id="623cd19ebee9324841d82041" data-wf-element-id="9d571e54-15c3-15b7-f534-9ca86a539bbd" aria-label="Newsletter Form"><input type="email" class="text-field-normal w-input" maxlength="256" name="Email" data-name="Email"
    placeholder="Your email address" id="Email-3" required=""><input type="submit" value="Subscribe" data-wait="Please wait..." class="button-black w-button"></form>

Name: wf-form- — GET

<form id="wf-form" name="wf-form-" data-name="Report Form" method="get" class="report-form" aria-label="Report Form" data-dashlane-rid="5beaca23bc709f73" data-form-type="login,step">
  <input type="hidden" name="report-link" data-name="Report Link" value="">
  <input type="hidden" name="report-slug" data-name="Report Slug" value="">
  <input type="hidden" name="report-name" data-name="Report Name" value="">
  <input type="hidden" name="report-cover" data-name="Report Cover" value="">
  <input type="email" class="text-field-normal w-input" maxlength="256" name="email" data-name="Email" placeholder="Your email address" id="email" required="" data-dashlane-rid="c29105b4f1590a8b" data-form-type="email" data-kwimpalastatus="alive"
    data-kwimpalaid="1646890753297-1">
  <label class="w-checkbox" data-dashlane-label="true">
    <input type="checkbox" name="Opt-In-Mailing-List" id="Opt-In-Mailing-List" data-name="Opt-In Mailing List" required="" class="w-checkbox-input" data-dashlane-rid="3e0cb2a92b24a89d" data-form-type="consent">
    <span class="tiny-text-black w-form-label" for="Opt-In-Mailing-List">I agree to receive emails from Music Tomorrow (you can opt out at any time)</span>
  </label>
  <input type="submit" value="Download the report" data-wait="Please wait..." class="button-black w-button" data-dashlane-rid="78f3594c4fc93c65" data-form-type="action,next">
  <span id="1646890753297-1" data-dashlanecreated="true"
    style="color: initial; font: initial; font-feature-settings: initial; font-kerning: initial; font-optical-sizing: initial; font-synthesis: initial; font-variation-settings: initial; forced-color-adjust: initial; text-orientation: initial; text-rendering: initial; -webkit-font-smoothing: initial; -webkit-locale: initial; -webkit-text-orientation: initial; -webkit-writing-mode: initial; writing-mode: initial; zoom: initial; accent-color: initial; place-content: initial; place-items: initial; place-self: initial; alignment-baseline: initial; animation: initial; app-region: initial; appearance: initial; aspect-ratio: initial; backdrop-filter: initial; backface-visibility: initial; background: url(&quot;chrome-extension://fdjamakpfbbddfjaooikfcpapjohcfmg/content/injected/logo-autofill-known.svg&quot;) center center / contain no-repeat; background-blend-mode: initial; baseline-shift: initial; block-size: initial; border-block: initial; border: none; border-radius: initial; border-collapse: initial; border-end-end-radius: initial; border-end-start-radius: initial; border-inline: initial; border-start-end-radius: initial; border-start-start-radius: initial; bottom: initial; box-shadow: initial; box-sizing: initial; break-after: initial; break-before: initial; break-inside: initial; buffered-rendering: initial; caption-side: initial; caret-color: initial; clear: initial; clip: initial; clip-path: initial; clip-rule: initial; color-interpolation: initial; color-interpolation-filters: initial; color-rendering: initial; color-scheme: initial; columns: initial; column-fill: initial; gap: initial; column-rule: initial; column-span: initial; contain: initial; contain-intrinsic-block-size: initial; contain-intrinsic-size: initial; contain-intrinsic-inline-size: initial; content: initial; content-visibility: initial; counter-increment: initial; counter-reset: initial; counter-set: initial; cursor: initial; cx: initial; cy: initial; d: initial; display: inline; dominant-baseline: initial; empty-cells: initial; fill: initial; fill-opacity: initial; fill-rule: initial; filter: initial; flex: initial; flex-flow: initial; float: initial; flood-color: initial; flood-opacity: initial; grid: initial; grid-area: initial; height: 16px; hyphens: initial; image-orientation: initial; image-rendering: initial; inline-size: initial; inset-block: initial; inset-inline: initial; isolation: initial; left: 409.102px; letter-spacing: initial; lighting-color: initial; line-break: initial; list-style: initial; margin-block: initial; margin: initial; margin-inline: initial; marker: initial; mask: initial; mask-type: initial; max-block-size: initial; max-height: initial; max-inline-size: initial; max-width: initial; min-block-size: initial; min-height: initial; min-inline-size: initial; min-width: 16px; mix-blend-mode: initial; object-fit: initial; object-position: initial; offset: initial; opacity: 1; order: initial; origin-trial-test-property: initial; orphans: initial; outline: initial; outline-offset: initial; overflow-anchor: initial; overflow-clip-margin: initial; overflow-wrap: initial; overflow: initial; overscroll-behavior-block: initial; overscroll-behavior-inline: initial; overscroll-behavior: initial; padding-block: initial; padding: initial; padding-inline: initial; page: initial; page-orientation: initial; paint-order: initial; perspective: initial; perspective-origin: initial; pointer-events: initial; position: absolute; quotes: initial; r: initial; resize: initial; right: initial; ruby-position: initial; rx: initial; ry: initial; scroll-behavior: initial; scroll-margin-block: initial; scroll-margin: initial; scroll-margin-inline: initial; scroll-padding-block: initial; scroll-padding: initial; scroll-padding-inline: initial; scroll-snap-align: initial; scroll-snap-stop: initial; scroll-snap-type: initial; scrollbar-gutter: initial; shape-image-threshold: initial; shape-margin: initial; shape-outside: initial; shape-rendering: initial; size: initial; speak: initial; stop-color: initial; stop-opacity: initial; stroke: initial; stroke-dasharray: initial; stroke-dashoffset: initial; stroke-linecap: initial; stroke-linejoin: initial; stroke-miterlimit: initial; stroke-opacity: initial; stroke-width: initial; tab-size: initial; table-layout: initial; text-align: initial; text-align-last: initial; text-anchor: initial; text-combine-upright: initial; text-decoration: initial; text-decoration-skip-ink: initial; text-indent: initial; text-overflow: initial; text-shadow: initial; text-size-adjust: initial; text-transform: initial; text-underline-offset: initial; text-underline-position: initial; top: 7625.25px; touch-action: initial; transform: initial; transform-box: initial; transform-origin: initial; transform-style: initial; transition: initial; user-select: initial; vector-effect: initial; vertical-align: initial; visibility: visible; border-spacing: initial; -webkit-border-image: initial; -webkit-box-align: initial; -webkit-box-decoration-break: initial; -webkit-box-direction: initial; -webkit-box-flex: initial; -webkit-box-ordinal-group: initial; -webkit-box-orient: initial; -webkit-box-pack: initial; -webkit-box-reflect: initial; -webkit-highlight: initial; -webkit-hyphenate-character: initial; -webkit-line-break: initial; -webkit-line-clamp: initial; -webkit-mask-box-image: initial; -webkit-mask: initial; -webkit-mask-composite: initial; -webkit-perspective-origin-x: initial; -webkit-perspective-origin-y: initial; -webkit-print-color-adjust: initial; -webkit-rtl-ordering: initial; -webkit-ruby-position: initial; -webkit-tap-highlight-color: initial; -webkit-text-combine: initial; -webkit-text-decorations-in-effect: initial; -webkit-text-emphasis: initial; -webkit-text-emphasis-position: initial; -webkit-text-fill-color: initial; -webkit-text-security: initial; -webkit-text-stroke: initial; -webkit-transform-origin-x: initial; -webkit-transform-origin-y: initial; -webkit-transform-origin-z: initial; -webkit-user-drag: initial; -webkit-user-modify: initial; white-space: initial; widows: initial; width: 16px; will-change: initial; word-break: initial; word-spacing: initial; x: initial; y: initial; z-index: auto;"></span>
</form>

Text Content

BlogState of Data ReportAcademyProductCompanyLogin



AI & Machine Learning
Data & Marketing
Data & Streaming


INSIDE SPOTIFY’S RECOMMENDER SYSTEM: A COMPLETE GUIDE TO SPOTIFY RECOMMENDATION
ALGORITHMS

Dmitry Pastukhov
February 9, 2022



THE STATE OF DATA REPORT 2023 IS NOW LIVE

Answer questions about your favourite tools and platforms, make an impact on the
future trends, and get a discount code to purchase the report.

Take me to the survey

This article gets into great detail on how Spotify Recommender System works,
outlining the process Spotify recommender follows to understand assets and users
on the platform. If you're curious to find out more about how artists and their
team can influence the recommender system to optimize algorithmic traffic, check
out our ongoing series on Recommender System Optimization.

As we move ahead into the 2020s, an ever-increasing share of music consumption
and discovery is going to be mediated by AI-driven recommendation systems. Back
in 2020, as much as 62% of consumers rated across platforms like Spotify and
YouTube among their top sources of music discovery — and be sure that a healthy
chunk of that discovery is going to be mediated by recommender systems. On
Spotify, for instance, over one third of all new artist discoveries happen
through "Made for You" recommendation sessions according to the recently
released Made to be Found report.

Yet, as algorithmic recommendations take center stage in the music discovery
landscape, the professional community at large still perceives these recommender
algorithms as black boxes. Music professionals rely on recommender systems
across platforms like Spotify and YouTube to amplify the ad budgets, connect
with the new audiences, and all-around execute successful release campaigns —
while often having no clear vision of how these systems operate and how to
leverage them to amplify artist discovery.

The topic of unveiling AI-driven recommender systems and providing music
professionals with the resources and tools they need to understand and manage
these algorithms will be a big focus for Music Tomorrow throughout 2022. A few
weeks back, we kicked off the year with an article covering the ins and outs of
the famed TikTok "For You" algorithm. This time around, we'll dive deeper into
the topic with a breakdown of Spotify's recommender system (which can be, to an
extent, extrapolated to other DSP recommendation engines).


HOW RECOMMENDATION AND MUSIC DISCOVERY WORKS ON SPOTIFY?

In a lot of ways, Spotify's recommendation engine is dealing with a similar flow
as TikTok's "For You" algorithm, playing the matchmaker between the creators (or
artists) and users (or fans) on a two-sided marketplace. However, as opposed to
TikTok, in this case, we don't have the courtesy of recently leaked internal
documentation to uncover the makeup of the system. What we do have is the
company's extensive public R&D records, its API, and some common sense. That is
not to say that we don't know anything definitive on how the system works — in
fact, a healthy chunk of Spotify's recommendation approach has been widely
publicized — but we would have to descend into the area of educated guesses when
it comes to some of the more granular details. Don't worry, though: we'll make
it clear once we depart the land of facts.


BEHIND THE ALGORITHM: UNDERSTANDING MUSIC AND USER TASTES

In broad strokes, at the core of any AI recommender system, there's an ML model
optimized for the key business goals: user retention, time spent on the
platform, and, ultimately, generated revenue. For this recommendation system to
work, it needs to understand the content it recommends and the users it
recommends it to. On each side of that proposition, Spotify employs several
independent ML models and algorithms to generate item representations and user
representations. Let's break down exactly how this process works — starting with
the track/artist representations:


GENERATING TRACK REPRESENTATIONS: CONTENT-BASED AND COLLABORATIVE FILTERING

Spotify's approach to track representation is made up of two primary components:

 1. Content-based filtering, aiming to describe the track by examining the
    content itself
 2. Collaborative filtering, aiming to describe the track in its connection with
    other tracks on the platform by studying user-generated assets

The recommendation engine needs data generated by both methods to get a holistic
view of the content on the platform and solve the cold start problems when
dealing with newly uploaded tracks. First, let's take a look at the
content-based filtering algorithms:

ANALYZING ARTIST-SOURCED METADATA

As soon as Spotify ingests the new track, an algorithm will analyze all the
general song metadata provided by the distributor and metadata specific to
Spotify (sourced through the Spotify for Artist pitch form). In the ideal
scenario, where all the metadata is filled correctly and makes its way to the
Spotify database, this list should include:

 * Track title
 * Release title
 * Artist name
 * Featured artists
 * Songwriter credits
 * Producers credits
 * Label
 * Release Date
 * Genre & sub-genre tags* 
 * Music culture tags*
 * Mood tags*
 * Style tags*
 * Primary language*
 * Instruments used throughout recording*
 * Track typology (Is it a cover? Is it a remix? Is it an instrumental?)
 * Artist hometown/local market*

*Sourced through S4A

The artist-sourced metadata is then passed downstream, as input into other
content-based models and the recommender system itself.

ANALYZING RAW AUDIO SIGNALS

The second step of the content-based filtering is the raw audio analysis, which
runs as soon as the audio files, accompanied by the artist-soured metadata, are
ingested into Spotify's database. The precise way in which that analysis is
carried out remains one of the secret sauces of the Spotify recommender system.
Yet, here's what we know for sure — and what we can reasonably assume:

Let's begin with the concrete facts. The audio features data available through
Spotify API consists of 12 metrics describing the sonic characteristics of the
track. Most of these features have to do with objective sonic descriptions. For
example, the metric of "instrumentalness" reflects the algorithm's confidence
that the track has no vocals, scored on a scale from 0 to 1. However, on top of
these "objective" audio attributes, Spotify generates at least three perceptual,
high-level features designed to reflect how the track sounds like in a more
holistic way:

 1. Danceability, describing how suitable a track is for dancing based on a
    combination of musical elements, including tempo, rhythm stability, beat
    strength, and overall regularity.
 2. Energy, representing "a perceptual measure of intensity and activity", based
    on the track's dynamic range, perceived loudness, timbre, onset rate, and
    general entropy.
 3. Valence, describing "the musical positiveness of the track". Generally
    speaking, tracks with high valence sound more positive (e.g., happy,
    cheerful, euphoric), while songs with low valence sound more negative (e.g.,
    sad, depressed, angry)

Yet, these audio features are just the first component of Spotify's audio
analysis system. In addition to the audio feature extraction, a separate
algorithm will also analyze the track's temporal structure and split the audio
into different segments of varying granularity: from sections (defined by
significant shifts in the song timbre or rhythm, that highlight transitions
between key parts of the track such as verse, chorus, bridge, solo, etc.) down
to tatums (representing the smallest cognitively meaningful subdivision of the
main beat).


Temporal Audio Analysis for Lil Nas X – Industry Baby (feat. Jack Harlow)
(Visualization by Spotify Audio Analysis)

Educated assumption: The combination of data generated by the audio analysis
methods should allow Spotify to discern the audio characteristics of the song
and follow their development throughout time and between different sections of
the track. Furthermore, the 12 audio features date back to 2013, when it was
just a part of The Echo Nest audio analysis output (an audio intelligence
company acquired by Spotify back in 2014). 

The chances are that in almost ten years, Spotify audio analysis algorithms have
gone a long way — which would mean that the audio feature data fed into the
recommendation system today is much more detailed and granular than what's
available through the company's public API. For instance, one of the research
papers published by Spotify in 2021 states that the audio features are passed
into the model as a 42-dimensional vector — which could mean that Spotify's
audio analysis produces 42 different audio features (although that's
circumstantial at best). 

In addition, the company's extensive research record details Spotify's
experiments with ML-based source separation and pitch tracking & melody
estimation. If these projects were to make it into production, that would mean
that the audio analysis system would be able to slice the track down to isolated
instrumental parts, process them separately and define all the melodies and
chord progressions used throughout the composition.

In practice, all of the above would mean that Spotify audio analysis can define
the recordings uploaded to the platform in great detail. The final output of the
system might define the track along the lines of "this song follows a
V-C-V-C-B-V-C structure, builds up in energy towards the bridge and features an
aggressive, dissonant guitar solo that resolves into a more melancholic and calm
outro". Or even something much more detailed — the point is that in all
likelihood, Spotify's audio analysis algorithm can reverse-engineer the track
almost entirely and extract the broadest range of characteristics from the raw
audio files. From the artist's perspective, it might be wise to assume that
Spotify has access to the DAW project files.

ANALYZING TEXT WITH NATURAL LANGUAGE PROCESSING MODELS

The final component of the content-based track representation is the Natural
Language Processing models, employed to extract semantic information describing
the track/artist from music-related text content. These models are applied in
three primary contexts:

 1. Lyrics analysis. The primary goal here is to establish the prominent themes
    and the general meaning of the song's lyrics while also scanning for
    potential "clues" that might be useful down the road, such as locations,
    brands, or people mentioned throughout the text. 
 2. Web-crawled data (focusing primarily on music blogs and online media
    outlets). Running NLP models against web-crawled data allows Spotify to
    uncover how people (and gatekeepers) describe music online by analyzing the
    terms and adjectives that have the most co-occurrence with the song's title
    or the artist's name.
 3. User-generated playlists. The NLP algorithms run against the user-generated
    playlists featuring the track on Spotify to uncover additional insights into
    the song's mood, style, and genre. "If the song appears on a lot of
    playlists with "sad" in the title, it is a sad song."

The NLP models allow Spotify to tap into the track's cultural context and expand
on the sonic analysis of how the song sounds with a social dimension of how the
song is perceived. 

The three components outlined above — artist-sourced metadata, audio analysis,
and NLP models — make up the content-based approach of the track representation
within Spotify's recommender system. Yet, there's one more key ingredient to
Spotify's recipe for track representation: 


COLLABORATIVE FILTERING

In many ways, collaborative filtering has become synonymous with Spotify's
recommender system. The DSP giant has pioneered the application of this
so-called "Netflix approach" in context of music recommendation — and widely
publicized collaborative filtering as the driving power behind its
recommendation engine. So the chances are, you've heard the process laid out
time and again. At least the following version of it:

"We can understand songs to recommend to a user by looking at what other users
with similar tastes are listening to." The algorithm simply compares users'
listening history: if user A has enjoyed songs X, Y and Z, and user B has
enjoyed songs X and Y (but haven't heard Z yet), we should recommend song Z to
them. By maintaining a massive user-item interaction matrix covering all users
and tracks on the platform, Spotify can tell if two songs are similar (if
similar users listen to them) and if two users are similar (if they listen to
the same songs).

Sounds like a silver bullet for music recommendation, doesn't it? In reality,
however, this item-user matrix approach comes with a host of issues that have to
do with accuracy, scalability, speed, and cold start problems. So, in recent
years, Spotify has moved away from consumption-based filtering — or at least
drastically downplayed its role in track representation. Instead, the current
version of collaborative filtering focuses on the track's organizational
similarity: i.e., "two songs are similar if a user puts them on the same
playlist".

By studying playlist and listening session co-occurrence, collaborative
filtering algorithms access a deeper level of detail and capture well-defined
user signals. Simply put, streaming users often have pretty broad and diverse
listening profiles — in fact, building listening diversity is one of Spotify's
priorities, as we've covered in our recent article on fairness in music
recommender systems — and so the fact that a lot of users listen to song A and
song B doesn't automatically that these two artists are similar. After all,
artists like Metallica and ABBA probably have quite a bit of shared listeners. 

If, on the other hand, a lot of users put song A and song B on the same
playlist, that is a much more conclusive sign that these two songs have
something in common. On top of that, the playlist-centric approach also offers
insight into the context in which these two songs are similar — and with
playlist creation being one of the most widespread practices on the platform,
Spotify has no shortage of collaborative filtering data to work through.

Today, the Spotify collaborative filtering model is trained on a sample of ~700
million user-generated playlists selected out of the much broader set of all
user-generated playlists on the platform. The main principle for choosing the
playlists that make it into that sample? "Passion, care, love, and time users
put into creating those playlists."

Now, we finally arrived at the point where the combination of collaborative and
content-based approaches allows Spotify's recommender system to develop a
holistic representation of the track. At this point, the track profile is
further enriched by combining the outputs of several independent algorithms to
generate higher-level vectors (think of these as mood, genre, style tags, etc.).
In addition, to deal with the cold start problem when processing freshly
uploaded releases that don't have enough NLP/playlist data behind them, some of
these properties are also extrapolated to develop overarching artist algorithmic
profiles.

However, to turn this track- and artist-level data into relevant
recommendations, the engine needs to marry it with the data describing the users
— which brings me to the next section.


GENERATING USER TASTE PROFILES

The approach to user profiling on Spotify is quite a bit simpler, at least once
we solve the track representations. Essentially, the recommender engine logs all
of the user's listening activity, split into separate context-rich listening
sessions. This context component is vital when interpreting user activity to
generate taste profiles. For instance, if the user engages with Spotify's
"What's New" tab, the primary goal of the listening session is often to quickly
explore music recently added to the platform. In that context, high skip rates
are to be expected, as the user's primary goal is to skim through the feed and
save some of the content served for later — which means that a track skip
shouldn't be interpreted as a definite negative signal. On the other hand, if
the user skips a track when listening to a "Deep Focus" playlist designed to be
consumed in the background, that skip is a much stronger sign of user
dissatisfaction.

Generally speaking, the user feedback can be split into two primary categories: 

 * Explicit, or active feedback: library saves, playlist adds, shares, skips,
   click-through to artist/album page, artist follows, "downstream" plays 
 * Implicit, or passive feedback: listening sessions length, track playthrough,
   and repeat listens

In the case of the Spotify recommender system, explicit feedback weighs in more
when developing user profiles. Music is often enjoyed as off-screen content,
meaning that uninterrupted consumption doesn't always relate to enjoyment. Then,
user feedback data is processed to develop the user profile, defined in terms
of:

 * Most-played and preferred songs and artists 
 * Saved songs and albums & followed artists
 * Genre, mood, style, and era preferences
 * Popularity and diversity preferences
 * Temporal patterns
 * Demographic & geolocation profile

Then the user taste profile is further subdivided based on the consumption
context: i.e., the same user might prefer mellow indie-pop on Sunday evenings
and high-energy motivational hip-hop on Monday mornings. In the end, Spotify
ends up with a context-aware user profile, that might look something like this:


Source: Spotify Research

This history-based user profile constantly develops and expands with fresh
consumption and interaction data. Recent activity is also prioritized over
historic profile: for instance, if the user gets into the new genre and it
scores well in terms of user feedback, the recommender system will try to serve
more adjacent music — even if the user's all-time favorite music is widely
different. 


RECOMMENDING MUSIC: INTEGRATING USER AND TRACK REPRESENTATIONS 

Woooh. You've made it. The intertwined constellation of algorithms behind the
Spotify recommendations has produced the two core components — track and user
representations — required to serve relevant music. Now, we just need the
algorithm to make the perfect match between the two and find the right track for
the right person (and the right moment).

However, the recommendation landscape on Spotify is much more diverse than on
some of the other consumption platforms. Just consider the range of Spotify
features that are generated with the help of the recommendation engine:

 1. Discover Weekly & Release Radar playlists
 2. Your Daily Mix playlists
 3. Artist / Decade / Mood / Genre Mix playlists 
 4. Special personalized playlists (Your Time Capsule, On Repeat, Repeat Rewind,
    etc.)
 5. Personalized editorial playlists
 6. Personalized browse section
 7. Personalized search results
 8. Playlist suggestions & enhance playlist feature
 9. Artist/song radio and autoplay features

In one way or another, all these diverse spaces are mediated by the recommender
engine — but each of them is running on a separate algorithm with its own inner
logic and reward system. The track and user representation form a sort of
universal foundation for these algorithms, providing a shared model layer
designed to answer the common questions of feature-specific algorithms, such as:

 * User-entity affinity: "How much does user X like artist A or track B? What
   are the favorite artists/tracks of user Y?"
 * Item similarity: "How similar are artist A & artist B? What are the 10 tracks
   most similar to track C?"
 * Item clustering: "How would we split these 50 tracks/artists into separate
   groups?"



The feature-specific algorithms can then tap into these unified models to
generate recommendations optimized for a given consumption space/context. For
instance, the algorithm behind Your Time Capsule playlists would primarily
engage with user-entity affinity data to try and find the tracks that users love
but haven't listened to in a while. On the other hand, Discover Weekly
algorithms would employ a mix of affinity and similarity data to find tracks
similar to the user's preferences, which they haven't heard yet. Finally,
generating Your Daily Mix playlists would involve all three methods — first,
clustering the user's preferences into several groups and then expanding these
lists with similar tracks. 


THE GOALS AND REWARDS OF SPOTIFY RECOMMENDATION ALGORITHMS

Now, as we mentioned in the beginning of this breakdown, the overarching goal of
the Spotify recommender system has to do primarily with retention, time spent on
the platform, and general user satisfaction. However, these top-level goals are
way too broad to devise a balanced reward system for ML algorithms serving
content recommendations across a variety of features and contexts — and so the
definition of success for the algorithms will largely depend on where and why
the user engages with the system.

For instance, the success of the autoplay queue features is defined mainly in
terms of user engagement — explicit/implicit feedback of listen-through and skip
rates, library and playlist saves, click-through to the artist profile and/or
album, shares, and so on. In the case of Release Radar playlists, however, the
set of rewards would be widely different, as users would often skim through the
playlist rather than listen to it from cover to cover. So, instead of studying
engagement with content, the algorithms would optimize for long-term feature
retention and feature-specific behavior. "Users are satisfied with the feature
if they keep coming back to it every week; users are satisfied with Release
Radar if they save tracks to their playlists or libraries." 

Finally, in some cases, Spotify would employ yet another set of algorithms just
to devise the reward functions for a specific feature. For example, Spotify has
trained a separate ML model to predict user satisfaction with Discover Weekly
(with the training set sourced by user surveys). This model would look at the
entire wealth of user interaction data, user past Discover Weekly behavior, and
user goal clusters (i.e., if the user engaged with Discover Weekly as a
background, to search for new music, save music or later, etc.) — and then
produce a unified satisfaction metric based on all that activity. 



The satisfaction prediction produced by the model is then, in turn, used as the
reward for the algorithm that would compose Discover Weekly playlists, thus
building a comprehensive reward system that doesn't rely on siloed, potentially
ambiguous user signals.

The Spotify recommender system is an extremely complex and intricate system,
with dozens (if not hundreds) of algorithms and ML models employed across
various levels, all working together to create one of the most advanced
recommendation experiences on the music streaming market. This system has been
developed and iterated on for close to 12 years now — growing in size,
capabilities, and complexity. Yet, as you can probably see, it is far from
unexplainable. Even without having direct documentation describing the
composition of the recommendation engine and all the secret ingredients, we can
get a pretty good understanding of its main parts and the governing principles
behind them.

From the music industry perspective, it is possible to use that knowledge to
optimize the artist's profile within that recommender system. A meaningful,
well-educated algorithmic strategy can maximize your chances of making it onto
algorithmic playlists and help ensure that the engine serves your music to the
right audiences, amplifying discovery and turning casual listeners into fans.

How would we go about it? We're currently hard at work building a way for
artists and their teams to understand and optimize their algorithmic profiles:
think of it as something of an SEO tool for streaming services. Or, should we
just say music's first-ever RSO tool, for Recommender System Optimization? Check
out the first part in our ongoing series on streaming RSO for a preview of our
model, and our vision for how music professionals can leverage such a tool to
optimize for algorithmic traffic and discovery on streaming platforms.


NEWSLETTER

Join our community of music data professionals and never miss out on new posts!

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.


DOWNLOAD THE REPORT

Leave your email below, and we will share the report with you!

I agree to receive emails from Music Tomorrow (you can opt out at any time)
Thank you! You will receive your report by email shortly
Oops! Something went wrong while submitting the form.


OPTIMIZE YOUR SEO ON STREAMING PLATFORMS

Understand how music discovery algorithms work and leverage this knowledge to
maximise algorithmic consumption for your upcoming and existing releases

Discover the course


CATEGORIES

Under the Hood
Data & Music Rights
VR & AR
Data & Live Industry
Data & Gender Equality
Data & Streaming
Data & Marketing
Blockchain & NFTs
AI & Machine Learning
General


RECENT POSTS

Getting it right: How our Data-Driven Approach Helped an Indie Label Drive 11X
Algorithmic Streams for Newly Released Music
The (Negative) Impact of Fake Streams on Artists' Algorithmic Performance
Towards Recommender System Optimization. Part 2: How Can Artists Influence
Recommendation Algorithms?
Towards Recommender System Optimization. Part 1: Our Vision and Our Data Tool
for Algorithmic Optimization on Streaming Platforms
GDPR & Music Data Ownership: Should We Treat Artist Data as Personal Data?


THE FIRST KNOWLEDGE HUB FOR DATA-DRIVEN MUSIC PROFESSIONALS


 * Blog
 * State of Data Report
 * Academy
 * Consulting Services
 * Careers

 * Company
 * Contact Us
 * Privacy Policy
 * Join Us on Socials
 * 

Test Mode