Team members: Shijin,Felix, Esra, Devika

Title: “Decoding the Hits: Analyzing the Spotify 2019 Top 50 ”

Task:

Our team is tasked with analyzing the Spotify Top 50 Songs of 2019 to extract a data-driven profile of a hit song. We need to test key hypotheses to help the director confidently greenlight projects and marketing budgets.

High danceability and energy make a song more successful.

Big-name artists have a higher chance of charting.

Hypothesis

Hypothesis 2

Hypothesis 3

Hypothesis 1

The most popular songs tend to be energetic and evoke feel-good emotions in listeners.

Action:

You acquired the Spotify Top 50 2019 dataset from Kaggle. It includes audio features and metadata of the most-streamed tracks.

Challenges:

• Small sample size (only 50 hits).
• Genre definitions were inconsistent.
• No data on unsuccessful songs for contrast.

Enrichment Plan (For Future):

Include average-performing songs from Spotify Charts API to contrast and validate hypotheses.

Response to Director:

“We’ve pulled the top global data — let’s start by seeing what works in the winners' circle first.”

​

DATA ACQUISITION & EXAMINATION

​

Objective

To ensure accurate SQL import and analysis, we cleaned and standardized the dataset using Python.

1. Column Standardization:

• Renamed all columns to snake_case (e.g., Track.Name → track_name) for readability and SQL compatibility.

2. Genre Cleaning & Grouping:

• Converted all genres to lowercase and stripped whitespace:�Grouped detailed sub-genres (e.g., “dance pop”, “reggaeton flow”) into broader parent categories like pop, rap, latin, or edm.

3. Track Name Cleaning:

• Problem: Some track names had accented/non-ASCII characters that could break SQL.
• Solution: We used regex to remove them and saved the files with UTF-8 encoding.

​

Database Schema: ERD Overview 

Three core tables:

• Artist: Stores artist metadata
• Genre: Represents cleaned and grouped genres
• Track: Contains track-level features and metadata

​

​

Design enables analysis of:

• How audio features like danceability, valence, or energy vary across genres
• Artist recurrence in Top 50
• Trends between musical features and artist popularity

PK = Principal Key

FK = Foreign Key

Transformation Hurdles

​

LIKE A BRIDGE OVER TROUBLED WATER...

We encountered several challenges during data transformation:

• Inconsistent data types: Some features had incorrect or mismatched types, which could cause SQL import errors or limit analysis.

​

• Genre inconsistencies: Many genre labels were duplicated or too granular — for example, both “pop” and “latin pop” were present.

​

• Normalization: We standardized genre entries by converting them to lowercase, and grouping similar subgenres.�

​

Visual 1 

Bar chard: Happiest songs are energetic and high valance

Most high-valence songs are pop and latin genres.

These tracks sound happier, with valence > 75 and varied BPMs.

Supports the hypothesis that pop/latin fit party/workout vibes.

​

​

Bar Chart of Songs by Genre

→ Pop, Rap, and Boy band top the list.

Pop appears most often among songs with both high valence and danceability.

Suggests pop is a strong fit for upbeat, danceable playlists.

Genres like boy band, R&B, and Latin show the highest avg valence & danceability — great for upbeat playlists.

The highest tempo tracks are pop, latin, and rap.

High BPM often aligns with high energy, ideal for workout playlists.

Confirms that these genres contribute to energetic vibes.

Insight: Streaming hits live in the zone of “moveable & memorable” — high beat, emotional charge. 

​

Top 10 high tempo tracks by genre & Valence

​

Strong positive correlation between energy, tempo, and danceability. 

Director's Insight: 

“Perfect.— the most streamed songs are both high-energy and danceable. That’s our target sound.”

Scatter Plot: Popular Tracks vs. Energy and Danceability diagram of trek popularity 

Conclusions & Business Implications

• Resonance Studios can greenlight songs fitting the hit profile.
• Allocate more marketing to shorter, danceable tracks with trending genres.
• Groom new artists into the pop or spaces using this data blueprint

RESULT

Business Implications:

Results (RESULT): 

✅ Confirmed all three hypotheses: 

High danceability/energy = higher ranking 

Known artists dominate the charts 

The most popular songs tend to be energetic and evoke feel-good emotions in listeners.

Director's Decision: 

“This is the blueprint we needed. I’m approving a pilot EP with three new artists modeled after these findings.” 

Major Obstacles

Challenge:

We lacked data on lower-performing songs to contrast — hard to definitively say what doesn’t work.

Learning:

To give deeper business advice, we must balance analysis of hits with flops. Future versions of this project will include that data.

Reflection for the Client:

“Next time, we compare both sides — hits and misses. That’s how we truly predict success

​

Insights & Visual Analysis

Our analysis highlights Pop, Latin, and Rap as the dominant genres among high-performing tracks in 2019.�

These genres consistently score high in tempo, valence, and danceability — key traits for songs that are energetic and emotionally engaging.�

High BPM and high valence correlate strongly with success, especially in the streaming era where short, catchy, and upbeat tracks are favored.�

Genres like Boy Band and Pop also shine, combining danceability and happiness — a winning combo for chart-topping songs.�

Conclusion: The top hits of 2019 followed a clear pattern — they were fast-paced, feel-good, and danceable.�

This supports our hypothesis: upbeat and energetic music drives commercial success, particularly on streaming platforms.

​

🎤 Slide 10: Closing Slide 

Title: Decoding the Hits: Analyzing the Spotify Top 50 (2019) 

Team:  

Final Message to Director Chris 

“Music might be an art — but the charts speak in data. Let’s make the next big hit together.” 

Thank You! 🎶 

Click Play Now to Listen

A Poetic Playlist

Metaphors

in Music