Data Analysis of Wizkid's "Made In Lagos"

I decided to analyze Wizkid's (relatively new) Made In Lagos album to see if there were any interesting things I would find. The analysis is broken down into 2 parts: Audio and Lyrics. The things being stated below are mostly my opinions and things I inferred from the data I found.

Audio Analysis

I looked at the audio features from the Spotify API. These include “danceability”, “valence”, “loudness”, etc. For a full list of features and their meanings, check out the Spotify docs here.

• The song people are least likely to find boring is “No Stress”. It has high energy, danceability, tempo and loudness features. The song people are most likely to find boring is “Piece of me (feat. Ella Mai)”.

• The least distracting songs on the album are “Longtime (feat. Skepta)”, “Piece of me (feat. Ella Mai)”, and “True Love (feat. Tay Iwar)”. They have high values for acousticness and low values for danceability and energy. They’re good picks for concentration or deep work playlists.
• The songs most suitable for car rides or group listening are “Blessed (feat. Damian Marley)” and “Roma (feat. Terri)”. They are tracks with high valence, which sound more positive (happy, cheerful, euphoric).

Lyrics Analysis

Here is what I found by analysing the lyrics:

• Wizkid is probably the best listener in Afrobeats. The lyrics in Made In Lagos contain the bigrams “say na”, “she tell”, “she say”, “say she”, and “tell me” about 139 times. Meaning he said at least one of these bigrams approximately 10 times per song.
• The lyrics on the album are mostly positive. The only song with a negative sentiment polarity is “Reckless”. The song with the highest rating (most positive lyrics) is “Smile (feat. H.E.R.)”.
• The most used words (excluding stop words) on the album include “yeah”, “dey”, “say”, “love”, “tell”. Again proving that it was generally a good vibes album, and showing that Wizkid deffo listens anytime you “say” or “tell” him something.

Please feel free to email me and ask questions or suggest more things that could be derived from the data ✌🏾

Cover photo by Rob Rusling.