Background of company

Spotify is a digital music streaming service established in 2006 in Stockholm, Sweden, by Daniel
Ek and Martin Lorentzon. The platform started in October 2008 and has since grown to become
one of the world's most famous music streaming platforms.
Spotify has revolutionized the way we listen to music. The streaming service has emerged as the
undisputed tastemaker, from mood tracks to machine learning algorithms. Spotify is the present
and future of music after the period of cassette tapes, CDs, mp3s, and iTunes. You've probably
heard of it even if you do not use it. Spotify has upended the music business by allowing you to
listen to your favorite tracks for free whenever and wherever you want.
Spotify's innovative approach to music streaming, including its customized playlists and
recommendations and its social features, enabling users to share and find new music with friends
and followers, has contributed to its success.
Spotify made its first acquisition in 2013 when it bought Tunigo, which helped users create, find,
and share new music and playlists on Spotify.
Spotify has established connections with other businesses in the music industry, such as record
labels and streaming platforms, and extended its offerings to include original content, such as
exclusive music releases and podcasts. Early in 2023, Spotify will have more than 400 million
active monthly users across 178 different countries.

Competitive advantage and practical implications

Spotify, one of the largest on-demand music service in the world, has a track record of breaking
technological barriers and utilizing big data, artificial intelligence, and machine learning to its
advantage. Tens of millions of people listening to music every minute of the day gives users
access to a staggering amount of information, such as which songs get the most plays, where
listeners are tuning in from, and even what devices they are using to access the service. The
service continues to gather data points and use that data to train algorithms and machines to
listen to music and draw conclusions that impact both its operations and users' listening
experience.
Its robust recommendation system is one of Spotify's key competitive advantages. Spotify can
create customized playlists and song suggestions using previous listening data by combining
machine learning (ML) techniques, natural language processing (NLP), and convolutional neural
networks (CNN).
Hyper-personalized recommendations
Using ml, Spotify groups users based on shared musical tastes after analyzing user activity. With
the aid of this data, it can suggest music to users based on "similar" users' tastes (collaborative
recommendation).
Also, they scan tens of thousands of articles, blogs, and message boards using natural language
processing (NLP) to examine the wording used to describe particular musicians and songs. With
these insights, they can then classify musical genres (such as smooth jazz, melodic rap, and
classic rock), making it more straightforward for Spotify to locate similar artists and create
custom playlists (content-based recommendation).
Using CNN, Spotify analyses the song's tempo, musical key, volume, and other raw audio data.
With the aid of this data, the recommendation engine can be improved by classifying songs
according to the sort of music they are.
Spotify, in particular, extensively uses reinforcement learning, a machine learning model that
employs contextual cues to optimize toward a long-term reward. Spotify's machine learning
algorithm will eventually receive long-term user pleasure as a reward.
In order to keep users coming back to listen more frequently, Spotify's recommendations are
therefore made with the user's overall satisfaction in mind at every turn.

Fig 1: How Spotify uses Machine Learning and AI [1]

Playlists Created and Curated by AI

By using ai to construct whole playlists, Spotify goes beyond simple recommendations. There
are a variety of formats for these AI-powered playlists.
The discover weekly playlist compiles new and classic songs users will likely enjoy by using
information about users' listening preferences.
The release radar also automatically compiles a playlist of the most recent songs from artists the
users follow each week and offers tailored recommendations based on those following.
Along with mixtapes depending on moods, activities, and environmental elements like the day of
the week or time of day, the home page also offers various suggested lists of podcast episodes
users might enjoy.
Reinforcement learning is used by the company in this additional field as well. The model tries
to steer users towards audio choices that will leave them feeling more fulfilled by creating
playlists and suggesting content.

Fig 2: Specialized playlists curated by Spotify [1]

Natural Language Search

Natural language search on Spotify is powered by artificial intelligence. Before now, Spotify
matched material to searches by matching specific phrases entered into a search window. This
strategy was better than nothing, but it had several drawbacks. For example, it could only match
terms that were quite similar to those found in a song, album, or podcast title, so it couldn't
always produce results of good quality.
It differs from natural language search. Natural language search uses AI techniques like natural
language processing (NLP) and deep learning to comprehend the semantic relationships between
words, so it does not necessarily need to match your search terms exactly to the words in the title
to locate what you are looking for. This is so that natural language search can comprehend word
synonyms, paraphrases, and any information with the same meaning as what is being searched.
Even if your search does not exactly match what is specified in the podcast metadata, it is now
much simpler to find podcasts on pertinent topics thanks to a feature that Spotify has just made
available for podcast search.

Spotify for Artists App

Spotify recently released the Spotify for musicians app, which offers mobile access to analytics
—everything from which playlists bring in new listeners to how many streams they receive
overall. Spotify is attempting to make its mountains of data accessible to musicians and their
managers. Similar to google analytics for music. Although it was first released earlier this year as
an online version, the mobile app now enables musicians to access information from the tour
bus. The geographic streaming data can be helpful to musicians and their teams in better
organizing tours. Also, artists have more influence over their appearance on Spotify, including
the ability to choose the "artist's pick," edit their profiles and add playlists.
The corporation has made a determined effort to empower artists and reduce their skepticism of
it, and this is just the latest example of that. Another Spotify product, fans first, searches through
data to identify an artist's fervent followers and then targets them with exclusive promotions.

Bayesian Additive Regression Trees

Spotify's "home page" employs a machine learning (ML) system called BART to offer listeners
real-time music suggestions, claims a 2019 data science central article. Bayesian additive
regression trees, often known as BART, are distinctive in that they:

· The streams longer than 30 seconds are best for it. It implies that it dismisses any song
users hear for shorter than 30 seconds.
· The technique was created to remove positional bias, and they retrain the model once
daily, depending on the interaction data they collect. The weighting of bottom-page clicks
is greater than top-page clicks
As for the following steps, in January 2021,Sspotify received a new patent for an AI program
that analyses voice data to identify users' "emotional state, gender, age, or accent" and then
makes better music recommendations based on this knowledge.
Spotify’s Artificial Intelligence Investments and Acquisitions
Since approximately ten years ago, Spotify has steadily acquired AI firms or firms with ties to
AI.

· To enhance its music recommendation systems, Spotify purchased Tunigo in 2013.

· Echo nest, a provider of music intelligence that Spotify utilized to enhance suggestions,
was purchased by the firm in 2014.
· The data science firm seed scientific was purchased by Spotify in 2015.
· Spotify purchased Sonalytic in 2017. To identify sounds and suggest music, Sonalytic
uses machine learning.
· Moreover, to enhance music searches and recommendations, Spotify purchased the AI
startup Niland in 2017.
· Spotify just purchased Sonantic, an AI-driven text-to-speech engine, as part of its most
recent AI-related acquisition.
Also, Spotify launched a regular gathering called machine learning day in 2018 that brings
researchers from the firm together to discuss fundamental AI concepts.

Business Model and Strategic Framework

Spotify uses the freemium business model and generates revenue through paid subscriptions and
advertisements. Customers can select from a range of premium paid plans that cater to the
overall user base and provide discounts. To accommodate their size and purpose, spotify
additionally offers its advertisers customizable advertising options.
Users get a memorable, personalized experience thanks to the current music library, excellent
playback, diverse catalogue, personalization options, and sharing capabilities.

Spotify’s Target Audience:

Based on the purpose of usage, Spotify targets listeners, advertisers, and content creators.
Users: they are the consumers of the labelled and original content available on the platform.
Advertisers might be paid or free users, companies, organizations, brands, or organizations who
use the platform to reach audiences and accomplish their goals.
Podcasters and artists who produce unique content and broadcast it online via the platform are
considered content creators.
Depending on the user base, Spotify majorly targets people between the age group of 18-30 years
as they have the capacity to spend some extra stuff and also, they are more enthusiast for the
domain. For the people of age above them they also have content specialized for them. 56% male
members and 44% female members is the data as per the gender graphics.

Value Creation: the software creates value for listeners, creators, and advertisers.

Fig 3: Value creation for listeners, creators and advertisers [2]

Spotify makes money from 2 sources

1. Premium Subscription
Users of Spotify's ad-free premium edition can play any kind of song on demand, find and hear
playlists, share playlists and music, listen to songs selected by Spotify's radio function, locate
and hear new music, and build playlists.
The premium version includes no ad interruptions, play music anywhere across a range of
devices, offline playback for saving music offline, and pay your way through flexible payment
modes and unlimited skips.
Fig 3: Premium features and plan on Spotify [2]
Spotify premium has 4 major plans named individual, mini, duo, family apart from them there
are various offers also which are given to students and different people depending on the
category they belong and also the Spotify’s collaboration with the company example people with
a specific credit card will have an offer of 10% off, etc.

2. From Advertisements
In addition to using cutting-edge technologies to target its customers with audio, video, podcast,
and multi-format ads, Spotify makes money from native banner advertisements.
Spotify supports different kinds of advertisements that vary in type, size and user engagement.
Sponsored playlists: here, companies can sponsor the top Spotify playlists for a week in order to
tie their brands with the service. It increases brand recognition and advertises the product. Both
laptops and smartphones can run them.
Branded moments: vertical video adverts that appear at the start of playlists and, if viewed all
the way through, grant users 30 minutes of ad-free streaming. These may be used with laptops,
smartphones, and tablets.
Fig 4: Brand adverts and Sponsored playlists on Spotify [2]
Sponsored sessions: these also come before a 30-minute listening session that is undisturbed.
These do, however, include a banner to show the advertisement and a horizontal video style.
Both smartphones and tablets can run them.
Video takeovers: these are delivered during a listening session between songs when users
actively browse through the catalogue to enjoy and discover music and podcasts. They run only
on laptops.
Display: clickable banner ads displayed at the bottom for 30 seconds, visible throughout the
timespan.
Homepage takeovers: showcase the brand message on the front of Spotify’s desktop homepage
for 24 hours. Works only on a laptop.

Spotify’s Cost Structure

It includes
1. Royalties or licensing fees
One of the major costs for the streaming service is the royalties that Spotify pays to get the
license to stream particular content on its platform.
Recording royalty: large music labels like Sony music entertainment group, warner music
group, and universal music group have recording license arrangements with Spotify. The sum
paid as compensation for Spotify's ability to stream the actual recordings of these record labels
on its platform is known as a recording royalty.
Composition royalty: Spotify has composition license agreements with the original songwriters
and composers as well.
Therefore, when a premium or ad-supported user plays a song on the platform, spotify pays
recording royalty to the music label and the artist and pays composition royalty to the owners
and composers of that song.
2. Company operations.
Cost of revenue: it includes all the costs incurred to deliver the service and value to customers.
E.g., maintenance of website and app, salaries, payment to partners, transaction fee, other
overhead costs etc.
Research and development: costs incurred on personnel and facilities associated with the
research and development of the service — for example – salaries to engineers etc.
Sales and marketing: it includes all costs incurred on personnel and facilities related to sales,
support, marketing, business development and customer service functions.
General and administrative expenses: costs incurred on personnel and allocated facilities for
general functions, administration, legal and financial purposes.

Competitors
1. Apple music
An audio streaming service called apple audio was introduced in 2015, and membership is
available for both monthly and yearly fees. Over 75 million songs are included in its library,
accessible in more than 100 nations. (as of 2021).
Its primary competitive advantage is its extensive product portfolio, which includes several
devices that support apple music. However, Spotifys retains the advantage because it stays
committed to its core platform and is not diversifying beyond music. While apple attempts to
promote brands synergistically to increase user growth, Spotify will concentrate on its core
advantage of being the most significant player in the digital music space.

2. Amazon music
Amazon music is a streaming service that charges a monthly fee to listen to tracks on demand. It
is one of three primary music streaming services. (the other two being Spotify and apple music).
Amazon owns the streaming program, which debuted in 2007. It is available on all leading
platforms, including android and IOS devices and desktop web browsers. Amazon music has a
catalog of over 75 million songs, directly rivaling apple music and Spotify.
Despite being a well-known retailer, apple music, and Spotify pose a severe threat to amazon's
music streaming service, primarily due to the latter's poor standing in digital music. Outside the
amazon ecosystem, it has yet to experience much development, which will hurt its ability to
draw in new clients.

3. YouTube music
The company introduced a music streaming program called YouTube music in 2015. It enables
users to enjoy free music, album, or playlist access. (with ads). Also, free users can make their
own playlists and download the desired music to their mobile devices. Paid YouTube red
subscribers have access to an audio-only option, which allows them to listen to music without
being interrupted by a video.
With its subscriber base lingering around 30 million as of 2020, it has yet to make real strides in
the digital sphere. In 2019, Spotify held a 35 percent share of the market for online music
streaming, which the music streaming service provider managed at 9%.

4. Pandora
Pandora is a music streaming service that enables users to build personalized radio stations by
selecting their favorite songs or artists. The free version allows users to choose from a list,
whereas the paid version allows limitless skips and an ad-free experience.
Spotify’s extensive content library over pandora allows it to grow and acquire market share.
Furthermore, the company's most significant advantage is its aggressive approach to growing its
user base through price cuts and tying local music stores to a free subscription to their streaming
services.

Conclusion
Some people believe Spotify to be superior to its rivals since it offers one of the largest music
catalogs of any music streaming service, with over 70 million songs. The user interface of
Spotify is straightforward, easy to use, and highly adaptable. Also, it provides several features,
including social sharing, podcast support, and customized playlists. The highly sophisticated
algorithms used by Spotify use machine learning to suggest customized music playlists to
consumers based on their listening habits and behavior. Spotify provides a free, ad-supported tier
and a premium membership with extra features, including ad-free streaming, high-quality audio,
and offline playing, making it accessible to all users.

References
 https://www.forbes.com/sites/bernardmarr/2017/10/30/the-amazing-ways-spotify-uses-
big-data-ai-and-machine-learning-to-drive-business-success/?sh=282337184bd2
 https://www.marketingaiinstitute.com/blog/spotify-artificial-intelligence
 https://www.aidataanalytics.network/data-monetization/articles/data-visualization-
monetization-and-personalization-spotify