Musical Discovery

Monitor your musical exploration: new artists, new tracks, and discovery trends

Discovery Overview

Total Artists Discovered

3,787

Total Tracks Discovered

9,288

New Tracks — Apr 2026

New Artists — Apr 2026

Cumulative Discovery

Artists
Tracks

New Discoveries Per Month

New Artists
New Tracks

Discovery Rate Over Time

Discovery Rate = proportion of today’s songs that were first-ever listens. A rate of 0.5 means half of what you listened to today was new to you.

Discovery Score

Discovery Score combines the track discovery rate (how much is new) and the inverse of repetition (how little you repeat) into a single 0–100 score. Higher = more exploratory listener.

Formula: (discovery_track_rate × 50) + (50 / repetition_kpi)

Discovery Heatmap

Discovery Stats Table

Monthly Discovery Statistics

New artists, tracks, and discovery metrics by month

Last updated on 2026-04-04

--- title: "Musical Discovery" description: "Monitor your musical exploration: new artists, new tracks, and discovery trends" --- ```{r setup} #| message: false #| warning: false library(here) source(here("src", "utils", "utils.R")) load_common_libraries() library(viridis) ``` ```{r load-data} #| message: false #| warning: false discovery <- load_discovery() daily <- load_daily_processed() monthly <- load_monthly_processed() weekly <- load_weekly_processed() # Monthly discovery aggregates monthly_discovery <- discovery %>% mutate(year_month = floor_date(date, "month")) %>% group_by(year_month) %>% summarise( new_artists = sum(new_artists), new_tracks = sum(new_tracks), total_songs = sum(total_songs), discovery_artist_rate = round(mean(discovery_artist_rate), 4), discovery_track_rate = round(mean(discovery_track_rate), 4), .groups = "drop" ) # Weekly discovery aggregates weekly_discovery <- discovery %>% mutate(week_start = floor_date(date, "week", week_start = 1)) %>% group_by(week_start) %>% summarise( new_artists = sum(new_artists), new_tracks = sum(new_tracks), total_songs = sum(total_songs), discovery_track_rate = round(mean(discovery_track_rate), 4), .groups = "drop" ) # Discovery score = weighted combo of track discovery rate and inverse repetition KPI # Score is 0–100: higher = more exploratory discovery_with_rep <- discovery %>% left_join(daily %>% select(date, repetition_kpi), by = "date") %>% mutate( exploration_score = round( (discovery_track_rate * 50) + (50 / pmax(repetition_kpi, 1)), 1 ) ) # Current month stats current_month <- floor_date(Sys.Date(), "month") current_disc <- monthly_discovery %>% filter(year_month == current_month) previous_disc <- monthly_discovery %>% filter(year_month == current_month %m-% months(1)) current_new_tracks <- if (nrow(current_disc) > 0) current_disc$new_tracks else 0 current_new_artists <- if (nrow(current_disc) > 0) current_disc$new_artists else 0 prev_new_tracks <- if (nrow(previous_disc) > 0) previous_disc$new_tracks else 0 total_unique_artists <- max(discovery$cumulative_new_artists, na.rm = TRUE) total_unique_tracks <- max(discovery$cumulative_new_tracks, na.rm = TRUE) ``` ## Discovery Overview ```{r discovery-value-boxes} layout_column_wrap( width = 1/4, value_box( title = "Total Artists Discovered", value = scales::comma(total_unique_artists), showcase = bsicons::bs_icon("person-bounding-box"), theme = "primary" ), value_box( title = "Total Tracks Discovered", value = scales::comma(total_unique_tracks), showcase = bsicons::bs_icon("vinyl"), theme = "success" ), value_box( title = paste("New Tracks —", format(current_month, "%b %Y")), value = current_new_tracks, showcase = bsicons::bs_icon("star"), theme = if (current_new_tracks >= prev_new_tracks) "success" else "secondary" ), value_box( title = paste("New Artists —", format(current_month, "%b %Y")), value = current_new_artists, showcase = bsicons::bs_icon("person-plus"), theme = "info" ) ) ``` ## Cumulative Discovery ::: {.panel-tabset} ### Artists ```{r cumulative-artists} #| fig-width: 12 #| fig-height: 5 p <- discovery %>% ggplot(aes(x = date, y = cumulative_new_artists)) + geom_area(fill = "#1DB954", alpha = 0.2) + geom_line_interactive( aes( tooltip = paste0( "", format(date, "%b %d, %Y"), " ", "Total unique artists: ", scales::comma(cumulative_new_artists), " ", "New today: ", new_artists ), data_id = as.character(date) ), color = "#1DB954", linewidth = 1.2 ) + labs( title = "Cumulative Unique Artists Discovered", caption = "Total distinct artists ever heard, growing over time", x = NULL, y = "Unique Artists" ) + scale_x_date(date_labels = "%b %Y", date_breaks = "6 months") + scale_y_continuous(labels = scales::comma) + theme_minimal() + theme( plot.title = element_text(size = 14, face = "bold"), panel.grid.minor = element_blank() ) make_girafe(p, height_svg = 5) ``` ### Tracks ```{r cumulative-tracks} #| fig-width: 12 #| fig-height: 5 p <- discovery %>% ggplot(aes(x = date, y = cumulative_new_tracks)) + geom_area(fill = "#4A90E2", alpha = 0.2) + geom_line_interactive( aes( tooltip = paste0( "", format(date, "%b %d, %Y"), " ", "Total unique tracks: ", scales::comma(cumulative_new_tracks), " ", "New today: ", new_tracks ), data_id = as.character(date) ), color = "#4A90E2", linewidth = 1.2 ) + labs( title = "Cumulative Unique Tracks Discovered", caption = "Total distinct tracks ever heard, growing over time", x = NULL, y = "Unique Tracks" ) + scale_x_date(date_labels = "%b %Y", date_breaks = "6 months") + scale_y_continuous(labels = scales::comma) + theme_minimal() + theme( plot.title = element_text(size = 14, face = "bold"), panel.grid.minor = element_blank() ) make_girafe(p, height_svg = 5) ``` ::: ## New Discoveries Per Month ::: {.panel-tabset} ### New Artists ```{r monthly-new-artists} #| fig-width: 12 #| fig-height: 5 p <- monthly_discovery %>% ggplot(aes(x = year_month, y = new_artists)) + geom_col_interactive( aes( tooltip = paste0( "", format(year_month, "%B %Y"), " ", "New artists: ", new_artists, " ", "Total songs: ", scales::comma(total_songs) ), data_id = as.character(year_month), fill = new_artists ), alpha = 0.85 ) + geom_smooth(method = "loess", color = "#FF6B6B", linewidth = 1, se = FALSE) + scale_fill_gradient(low = "#4A90E2", high = "#1DB954", guide = "none") + labs( title = "New Artists Discovered per Month", caption = "Red line = LOESS trend", x = NULL, y = "New Artists" ) + scale_x_date(date_labels = "%b %Y", date_breaks = "3 months") + theme_minimal() + theme( plot.title = element_text(size = 14, face = "bold"), axis.text.x = element_text(angle = 45, hjust = 1), panel.grid.minor = element_blank() ) make_girafe(p, height_svg = 5) ``` ### New Tracks ```{r monthly-new-tracks} #| fig-width: 12 #| fig-height: 5 p <- monthly_discovery %>% ggplot(aes(x = year_month, y = new_tracks)) + geom_col_interactive( aes( tooltip = paste0( "", format(year_month, "%B %Y"), " ", "New tracks: ", new_tracks, " ", "Total songs: ", scales::comma(total_songs) ), data_id = as.character(year_month), fill = new_tracks ), alpha = 0.85 ) + geom_smooth(method = "loess", color = "#FF6B6B", linewidth = 1, se = FALSE) + scale_fill_gradient(low = "#4A90E2", high = "#1DB954", guide = "none") + labs( title = "New Tracks Discovered per Month", caption = "Red line = LOESS trend", x = NULL, y = "New Tracks" ) + scale_x_date(date_labels = "%b %Y", date_breaks = "3 months") + theme_minimal() + theme( plot.title = element_text(size = 14, face = "bold"), axis.text.x = element_text(angle = 45, hjust = 1), panel.grid.minor = element_blank() ) make_girafe(p, height_svg = 5) ``` ::: ## Discovery Rate Over Time > **Discovery Rate** = proportion of today's songs that were first-ever listens. A rate of 0.5 means half of what you listened to today was new to you. ```{r weekly-discovery-rate} #| fig-width: 12 #| fig-height: 5 p <- weekly_discovery %>% ggplot(aes(x = week_start, y = discovery_track_rate)) + geom_line(color = "#4A90E2", linewidth = 0.8, alpha = 0.6) + geom_smooth(method = "loess", color = "#1DB954", linewidth = 1.2, se = TRUE, alpha = 0.2) + geom_point_interactive( aes( tooltip = paste0( "Week of ", format(week_start, "%b %d, %Y"), " ", "Discovery rate: ", scales::percent(discovery_track_rate, accuracy = 0.1), " ", "New tracks: ", new_tracks, " ", "Total songs: ", scales::comma(total_songs) ), data_id = as.character(week_start) ), color = "#4A90E2", size = 2, alpha = 0.7 ) + labs( title = "Weekly Track Discovery Rate", caption = "Proportion of songs that were first-ever listens · Green = LOESS trend", x = NULL, y = "Discovery Rate" ) + scale_x_date(date_labels = "%b %Y", date_breaks = "6 months") + scale_y_continuous(labels = scales::percent_format(accuracy = 1)) + theme_minimal() + theme( plot.title = element_text(size = 14, face = "bold"), panel.grid.minor = element_blank() ) make_girafe(p, height_svg = 5) ``` ## Discovery Score > **Discovery Score** combines the track discovery rate (how much is new) and the inverse of repetition (how little you repeat) into a single 0–100 score. Higher = more exploratory listener. > > Formula: `(discovery_track_rate × 50) + (50 / repetition_kpi)` ```{r monthly-discovery-score} #| fig-width: 12 #| fig-height: 5 monthly_score <- discovery_with_rep %>% mutate(year_month = floor_date(date, "month")) %>% group_by(year_month) %>% summarise( avg_score = round(mean(exploration_score, na.rm = TRUE), 1), avg_disc_rate = round(mean(discovery_track_rate, na.rm = TRUE), 4), avg_rep_kpi = round(mean(repetition_kpi, na.rm = TRUE), 2), .groups = "drop" ) p <- monthly_score %>% ggplot(aes(x = year_month, y = avg_score)) + geom_line(color = "#FFD700", linewidth = 1.2) + geom_smooth(method = "loess", color = "#FF6B6B", linewidth = 1, se = TRUE, alpha = 0.2) + geom_point_interactive( aes( tooltip = paste0( "", format(year_month, "%B %Y"), " ", "Discovery score: ", avg_score, " ", "Avg disc. rate: ", scales::percent(avg_disc_rate, accuracy = 0.1), " ", "Avg rep. KPI: ", avg_rep_kpi ), data_id = as.character(year_month) ), color = "#FFD700", size = 3 ) + labs( title = "Monthly Discovery Score", caption = "Higher = more exploratory · Red = LOESS trend", x = NULL, y = "Discovery Score (0–100)" ) + scale_x_date(date_labels = "%b %Y", date_breaks = "3 months") + theme_minimal() + theme( plot.title = element_text(size = 14, face = "bold"), axis.text.x = element_text(angle = 45, hjust = 1), panel.grid.minor = element_blank() ) make_girafe(p, height_svg = 5) ``` ## Discovery Heatmap ```{r discovery-heatmap} #| fig-width: 14 #| fig-height: 6 # Calendar-style heatmap: new tracks per day (last 2 years) heatmap_data <- discovery %>% filter(date >= Sys.Date() - years(2)) %>% mutate( week_of_year = isoweek(date), year = factor(year(date)), weekday = wday(date, label = TRUE, abbr = TRUE, week_start = 1) ) if (nrow(heatmap_data) > 0) { p <- heatmap_data %>% ggplot(aes(x = week_of_year, y = weekday, fill = new_tracks)) + geom_tile_interactive( aes( tooltip = paste0(format(date, "%b %d, %Y"), " New tracks: ", new_tracks), data_id = as.character(date) ), color = "white", linewidth = 0.3 ) + facet_wrap(~ year, ncol = 1) + scale_fill_gradient(low = "#2E4057", high = "#1DB954", name = "New Tracks") + scale_x_continuous(breaks = c(1, 13, 26, 39, 52), labels = c("Jan", "Apr", "Jul", "Oct", "Dec")) + labs( title = "New Tracks Discovered — Calendar Heatmap", subtitle = "Last 2 years · Hover for date details", x = NULL, y = NULL ) + theme_minimal() + theme( plot.title = element_text(size = 14, face = "bold"), plot.subtitle = element_text(size = 11, color = "gray60"), panel.grid = element_blank(), strip.text = element_text(face = "bold", size = 12) ) make_girafe(p, width_svg = 12, height_svg = 5) } else { cat("Insufficient data for heatmap.") } ``` ## Discovery Stats Table ```{r discovery-table} #| message: false monthly_discovery %>% arrange(desc(year_month)) %>% left_join( monthly %>% select(year_month, repetition_kpi, unique_songs), by = "year_month" ) %>% left_join(monthly_score %>% select(year_month, avg_score), by = "year_month") %>% mutate(year_month_str = format(year_month, "%Y-%m")) %>% select( year_month_str, new_artists, new_tracks, total_songs, discovery_track_rate, repetition_kpi, avg_score ) %>% gt() %>% tab_header( title = "Monthly Discovery Statistics", subtitle = "New artists, tracks, and discovery metrics by month" ) %>% cols_label( year_month_str = "Month", new_artists = "New Artists", new_tracks = "New Tracks", total_songs = "Total Songs", discovery_track_rate = "Discovery Rate", repetition_kpi = "Repetition KPI", avg_score = "Discovery Score" ) %>% fmt_number(columns = c(new_artists, new_tracks, total_songs), decimals = 0) %>% fmt_percent(columns = discovery_track_rate, decimals = 1) %>% fmt_number(columns = c(repetition_kpi, avg_score), decimals = 1) %>% data_color( columns = avg_score, colors = scales::col_numeric(palette = c("#FFE5E5", "#1DB954"), domain = NULL) ) %>% data_color( columns = discovery_track_rate, colors = scales::col_numeric(palette = c("#FFF8E1", "#4A90E2"), domain = NULL) ) %>% opt_interactive(use_search = TRUE, use_filters = TRUE, page_size_default = 12) ``` --- *Last updated on `r Sys.Date()`*