Spotify Top 100 Case Study
Spotify Top 100 Case Study
Embedded Files
Introduction
Spotify is a proprietary Swedish audio streaming and media services provider founded on 23 April 2006 by Daniel Ek and Martin Lorentzon. It is one of the largest music streaming service providers, with over 422 million monthly active users, including 182 million paying subscribers, as of March 2022.
Spotify offers digital copyright restricted recorded music and podcasts, including more than 82 million songs, from record labels and media companies. As a freemium service, basic features are free with advertisements and limited control, while additional features, such as offline listening and commercial-free listening, are offered via paid subscriptions.
Business task
Analyze the top 100 songs of each year from 2010-2019 to gain insight into how these songs have made it to the top 100.
Questions for analysis
What are some trends in the songs on the top 100?
How could these trends help Spotify artists?
How could these trends help influence marketing strategy?
Key Stakeholders:
Daniel Ek – Spotify’s co founder and Chief Executive Officer
Spotify marketing analytics team – A team of data analysts responsible for collecting, analyzing, and reporting data that helps guide Spotify’s marketing strategy.
Prepare:
Spotify top 100 songs from 2010-2019 data on Kaggle will be used for this analysis. Because of the small sample set and lack of participating years variables such as quantity of streams and number of plays, this data has limitations.
Installing and Loading Packages:
The CSV file was opened In Google sheets and formatted the "time/date" column from "text" to "time". The files was saved locally to the computer and imported Into R Studio.
Process:
Inspecting data frames:
To make sure the data frames were imported properly, the head() function was used. The functions colnames() and glimpse() functions were used additionally to explore the data frames and find similarities.
I ran into issues when trying to view column details because of naming. Spaces are not always used in every coding syntax. R wasn't able to process the request so they were renamed so that I would not run into issues later on during the analysis.
The time frame for the songs are supposed to be 2009-2019. There were songs which had year release dates which were outliers. These were corrected to their appropriate year released date.
The summary() function was used to extract key statistics from each data frame. This function works well with quantitative data. The columns with character values don't benefit from the summary function as can be seen with TopGenre and ArtistType.
As I was culling through the data I realized that there were 3 rows at the end of the data which contained nothing. To ensure the integrity of the work, they were deleted to provide an accurate representation.
Analyze:
Summarize the data:
The n_distinct() and nrow() functions are used to determine the number of unique values and the number of rows in a data frame, respectively.
The average duration for the top 100 songs was 3.60 minutes.
The average BPM (beat per minutes) is 122.0
The top year which has the most top 100 chart songs is 2018.
Taylor Swift appeared the most In the top 100 chart songs with 21 appearances.
Dance Pop was the Top Genre among the all the songs.
The songs with the highest BPM of 206 is Honey Bee by Blake Shelton and FourFiveSeconds by Rihanna, Kanye, and Paul McCartney.
The songs with the highest Energy are Riverside by Sidney Samson and Get Up (Rattle) by Bingo Players.
Most Energetic songs are Anaconda by Nicki Minaj and Yes Indeed by Lil Baby.
The song with the greatest duration was Not a Bad Thing by Justin Timberlake.
The song with the shortest duration was Old Town Road by Lil Nas X.
The top artist type was Solo.
The columns Duration, BPM, Speech, Acoustic, and Val were put into a table to view all the averages side by side. (The Mean is the average)
The average song duration was 220.4 seconds (3 minutes 67 seconds). The average speech was 10.06. Speech; The higher the value the more spoken word the song contains. The average acoustic was 14.37. The average val was 50.9. Valence (val); the higher the value, the more positive mood for the song.
Share:
In order to export the data, I used write.csv2 to export the data to document form to better import into Tableau.
Figure 1: The Relationship between Top Genre and Danceability
Figure 1 shows the visible representation of the Top Genres and their Danceability. It is simple to see which genre categories perform the best in terms of danceability. The majority of pop and hip hop songs tend to outperform the rest of the genres.
Figure 2: The Relationship between BPM and Top Genre
Figure 2 shows the average BPM per genre. The average BPM Is clustered around 125 bpm.
Figure 3: The Relationship between BPM Correlation and Artist Type
Figure 3 indicates the correlation between Artist type and BPM. The BPM Is used as an indicator for the commonality between top songs. Duos and Band/ Groups lead the BPM every year in the decade.
Act:
Although hit song means any widely played or big-selling song, the specific term hit record usually refers to a single that has appeared in an official music chart through repeated radio airplay audience impressions, or significant streaming data and commercial sales.
When determining what goes Into getting a song on the top 100 charts there are many factors which can be observed. There is a strong correlation between the Top Genre (Dance Pop) and Danceability. Danceability: the higher the value, the easier it is to dance to this song.
If the hit song were to be measured on BPM, then bands and groups would have an advantage since they account for a large portion of BPM and Artist type correlation.When It comes to BPM, the average for the Top 100 songs Is 125 beats per minute. This could be a possible threshold for emerging songs to base a single criteria upon.
Such trends could help artists understand the dynamics which go In making a hit song. There are other factors which are at play which the data does not account. These Include but are not limited to; featured artists, market timing, and target audience. These outlying factors should be made available to artists to help them better tailor their music the way they see fit. Having Insight Into these types of data offer a greater advantage for a more competitive market and help push Spotify further.
Page updated
Google Sites
Report abuse