Menu
Spotify’s internal analytics team wanted to better understand user listening behavior — specifically, how engagement patterns varied by time, device, and content type (albums, artists, and tracks). The available raw data was from user activity logs was large and unstructured, the team needed to extract actionable insights about listening frequency, time preferences, and content performance. To support data-driven decisions on user engagement and product experience, a structured analytical solution was needed.
As a data analyst, my objective was to transform raw Spotify streaming history into an interactive Power BI dashboard that could uncover behavioral trends, highlight top content, and provide insight into when and how users interact with the platform. The goal was to enable Spotify’s marketing and product teams to make informed decisions about content promotion, playlist curation, and feature optimization based on real user behavior.
to make this happen I,
Data Preparation & Cleaning
Processed thousands of streaming records, converting timestamps and milliseconds into meaningful metrics such as total hours played, listening frequency, and average session duration.
Categorized listening sessions by device type (Android, iOS, Web, Windows, etc.), day of the week, and hour of playtime.
Identified and merged data on albums, artists, and tracks to ensure consistency and proper aggregation across years.
Dashboard Development
Designed a multi-page Power BI dashboard with dedicated sections for Albums, Artists, and Tracks.
Added KPIs such as total tracks played (13,665), total artists (4,112), and total albums (7,907) with year-over-year comparison indicators.
Built interactive visuals:
Trend charts for play counts over time.
Heatmaps for listening hours vs. days of the week.
Scatter plots showing the relationship between listening frequency and average playtime.
Top 5 rankings for albums, artists, and tracks.
Integrated slicers for filters such as device type, shuffle status, and skip rate to provide flexible exploration.
Visualization & Insight Optimization
Applied a dark Spotify-inspired theme for visual consistency and brand alignment.
Enhanced usability by adding tooltips, dynamic highlights, and KPI color codes for quick interpretation.
Created an interactive navigation system for seamless movement between summary pages and data-driven details.
The dashboard provided a comprehensive and dynamic view of Spotify listening behavior, delivering insights such as:
User engagement increased steadily from 2018–2023, with a peak in 2022 before a slight dip in 2024.
The Beatles emerged as the most played artist (13,621 plays), indicating enduring cross-generation popularity.
Listening activity peaks between 6 PM and 10 PM, especially on weekends — ideal timing for marketing pushes and playlist releases.
Weekday vs. weekend listening distribution showed strong weekend engagement, suggesting opportunities for event-based promotions.
Mobile (Android/iOS) dominated as the most used platforms, emphasizing mobile-first experience optimization.
These insights allowed the analytics team to refine personalized recommendations, improve release scheduling, and optimize user experience by focusing on engagement peaks and high-performing content.
Overall play activity has increased significantly from 2018 to 2023, indicating growing engagement and retention.
Action: Capitalize on this upward trend by introducing personalized year-end recaps or “rewind” campaigns to deepen user connection and encourage social sharing.
A slight decline in 2024 suggests emerging churn or behavior shifts.
Action: Conduct deeper cohort analysis to identify disengaged user groups and design re-engagement campaigns (e.g., “Rediscover Your Favorites”).
The heatmap revealed that the most active listening hours occur between 6 PM and 10 PM, especially on Fridays and weekends.
Action: Schedule new releases, curated playlists, or targeted notifications during these high-traffic hours to maximize impressions and stream counts.
Morning listening (6 AM–9 AM) showed modest but consistent engagement, likely tied to commuting habits.
Action: Promote short, mood-based playlists (e.g., “Morning Motivation”) to increase weekday retention.
Mobile platforms (Android and iOS) account for the majority of listening activity, while web player and desktop usage remain lower.
Action: Prioritize UI/UX improvements and new feature rollouts for the mobile app to strengthen user experience.
Action: Explore cross-platform incentives (e.g., “Continue listening on desktop for bonus skips”) to encourage multi-device engagement.
The Beatles, The Killers, and John Mayer are the top three artists by play count, collectively representing a large portion of total listening time.
Action: Leverage these high-performing artists in playlist collaborations or featured banners to boost discovery and engagement.
Top albums such as “Abbey Road” and “Past Masters” consistently outperform others.
Action: Promote similar genre playlists or “inspired by” recommendations to capture similar listener preferences.
Top tracks like “Ode to the…” and “In the Blood” demonstrate enduring replay value, suggesting strong emotional or nostalgic appeal.
Action: Use these tracks in algorithmic playlist seeding to enhance personalization accuracy.
Users listen more frequently in shorter sessions, with most tracks averaging under five minutes of continuous playtime.
Action: Optimize ad placement strategies for shorter listening sessions and test micro-podcasts or snackable content formats.
The scatter plot indicates high-frequency listeners engage repeatedly with a narrow selection of favorites.
Action: Encourage exploration through “Discover Weekly”–style recommendation nudges to diversify listening habits and increase session duration.
Weekend sessions show stronger intensity and longer play durations than weekdays.
Action: Launch weekend-exclusive promotions or “Friday Drop” campaigns to drive anticipation and boost weekend traffic.
Weekday engagement dips mid-day (12 PM–3 PM), indicating user downtime.
Action: Introduce background-friendly playlist features (e.g., “Focus Mode” or “Low-Fi Work Beats”) to re-engage users during working hours.
The number of unique artists (4,112) and albums (7,907) played indicates broad music diversity among users.
Action: Promote this variety through editorial curation — highlighting emerging artists alongside established names to sustain engagement.
The dashboard provides a holistic view of listening behavior across time, device, and content dimensions.
Action: Integrate this dashboard into Spotify’s analytics ecosystem for real-time tracking and campaign performance evaluation.
Action: Use insights to guide playlist algorithm tuning, advertising placements, and content licensing decisions.
| Area | Insight | Recommended Action |
|---|---|---|
| Engagement | 2022 peak, 2024 dip | Re-engagement campaigns & retention focus |
| Listening Hours | Evenings & weekends dominate | Schedule releases & notifications strategically |
| Devices | Mobile-first usage | Prioritize mobile experience optimization |
| Content | Beatles & classic rock dominate | Expand similar-genre marketing |
| User Behavior | Short frequent sessions | Introduce quick-play playlists & session-based ads |
| Regional Strategy | Global diversity evident | Localized recommendations per listening zone |
The video presents a walkthrough of a Power BI dashboard designed for interactive data analysis. It highlights several key user-friendly features that enhance the usability and flexibility of the report.
Users are encouraged to make full use of the embedded dashboard to explore and interact with its features. The dashboard has been designed for hands-on testing, allowing users to apply filters, use slicers. This interactive experience helps users better understand the data, test various scenarios, and evaluate the dashboard’s functionality in a real-time environment.
To support effective decision-making by the marketing, product, and analytics teams, there is a clear need to:
Establish a robust date-based relationship between the data table and a separate date dimension table in order to enable comprehensive time-series reporting and analysis.
This requirement aims to solve existing limitations in filtering and aggregating data over days, weeks, months, and other date hierarchies.
Identify User Behavior Trends Over Time: Understand how user engagement changes by day of week, time of year, and season.
Support Marketing Campaign Evaluation: Analyze the impact of promotions, new releases, and features on listener activity.
Enable Custom Time-Based Filters: Allow users to apply intuitive date filters like “Last 7 Days” or “Q2 2025” in reporting tools.
Improve Dashboard Performance and Consistency: Centralizing date logic improves efficiency and avoids duplication across reports.
Enhanced Reporting: Teams can now generate more meaningful insights using filters like month, quarter, or day of the week.
Strategic Planning: Helps marketing and content teams plan releases and campaigns based on historical engagement patterns.
Time Intelligence Functions: Enables use of advanced DAX or SQL functions for period-over-period comparison, running totals, and growth metrics.
Improved Data Quality: Reduces logic duplication across dashboards and ensures consistent use of time data.
Purpose: Stores event or activity data with metrics tied to specific points in time.
Key Fields:
spotify track uri
parameter_value (optional metric)
ts (timestamp of the event)
platform( device being played)
track_name
artiste_name
Purpose: Supports flexible, time-based slicing of data.
Derived From: The "ts" column in the parameter table.
Key Fields:
One-to-Many:
Each unique date in the date table corresponds to multiple timestamps in the Data table.
The join is established on:
date_table.date → data_table.date (extracted from ts column)
This relationship enables slicing data by any time attribute without duplicating logic.
Supports use of date hierarchies in Power BI, Tableau, or SQL queries.
Time Intelligence: Enables year-over-year, month-over-month, and week-based comparisons.
Efficient DAX or SQL Logic: Time-based calculations become cleaner and reusable.
Visual Clarity: Enhances reports with calendar-based trends and dashboards.
Power BI Desktop – For data transformation, modeling, and visualization
Power Query Editor – For data cleaning and shaping
DAX (Data Analysis Expressions) – For calculated columns and measures
CSV Dataset – Spotify track-level data
Imported the dataset into Power BI using the Get Data → CSV option.
Removed duplicate rows based on track_name and artist_name.
Removed rows with missing or null values in essential columns.
Converted release_date and ts to proper Date/Time format.
Added a new calculated column:duration_min = duration_ms / 60000
A dedicated Date Table was created to support time-based filtering and advanced analysis.
A one-to-many relationship was established between DateTable[Date] and SpotifyData[date] (derived from ts).
This enabled dynamic filtering, drill-through, and time-series analysis.
Created custom DAX measures to enrich analytical capabilities like;
Total Tracks by Artiste
Total Album by Artiste
Total numbers of Artiste
Developed an interactive dashboard with the following visuals:
Bar Chart: Top 5 Artists by Artiste Count,Top 5 Album by Album Count,Top 5 Tracks by Track Count
Line Chart: Yearly Trend of Album/artiste/Tracks
Scatter Plot: Relationship between Average time played and Frequency
Slicer Filters: Skipped, Shuffled, Year.
Included Drill-through Pages
Tracks with higher energy and danceability generally had higher popularity scores.
Most releases are concentrated in Q2 and Q4, possibly aligning with marketing cycles.
Pop and Hip-Hop dominate top artist lists.
Valence and tempo varied significantly by year, suggesting evolving music trends.
Copyright © 2025 Acme Products| All Rights Reserved
Every project starts with a chat, Emmanuel is keen to conversation and open to discussing your project. he will pull in the right expertise when needed