Artist Lifecycle

How your relationship with artists evolves over time: sticky favorites, comebacks, and lost loves

Artist Overview

Total Artists Heard

3,787

Currently Active

311

Comeback Artists

1,259

Drifted Away

386

Sticky Artists

Stickiness = fraction of days between first and last listen that you actually played this artist. A score of 1.0 would mean you listened every single day since discovery.

Long-running Favorites

Artists with the longest active span (days between first and last listen).

Comeback Artists

Artists who disappeared for 90+ days and then returned. More comebacks = a more cyclical listening pattern.

Drifted Away

Artists you used to love but haven’t listened to in 180+ days — sorted by how much you played them historically.

Artists You've Drifted Away From

Not heard in 180+ days · Sorted by historical plays · Max 30 shown

Full Artist Lifecycle Table

All Artists

3,787 artists in listening history

Last updated on 2026-04-04

--- title: "Artist Lifecycle" description: "How your relationship with artists evolves over time: sticky favorites, comebacks, and lost loves" --- ```{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 lifecycle <- load_lifecycle() max_date <- max(lifecycle$last_listen, na.rm = TRUE) active_count <- sum(lifecycle$is_active, na.rm = TRUE) total_artists <- nrow(lifecycle) comeback_artists <- sum(lifecycle$comeback_count > 0, na.rm = TRUE) drifted_artists <- sum(!lifecycle$is_active & lifecycle$total_plays >= 10, na.rm = TRUE) ``` ## Artist Overview ```{r lifecycle-value-boxes} layout_column_wrap( width = 1/4, value_box( title = "Total Artists Heard", value = scales::comma(total_artists), showcase = bsicons::bs_icon("people"), theme = "primary" ), value_box( title = "Currently Active", value = scales::comma(active_count), showcase = bsicons::bs_icon("activity"), theme = "success", ), value_box( title = "Comeback Artists", value = scales::comma(comeback_artists), showcase = bsicons::bs_icon("arrow-counterclockwise"), theme = "info" ), value_box( title = "Drifted Away", value = scales::comma(drifted_artists), showcase = bsicons::bs_icon("moon"), theme = "secondary" ) ) ``` ## Sticky Artists > **Stickiness** = fraction of days between first and last listen that you actually played this artist. A score of 1.0 would mean you listened every single day since discovery. ```{r sticky-artists} #| fig-width: 12 #| fig-height: 6 sticky <- lifecycle %>% filter(total_days >= 10) %>% arrange(desc(stickiness)) %>% head(15) %>% mutate( artist = factor(artist, levels = rev(artist)), span = as.integer(last_listen - first_listen) + 1L ) p <- sticky %>% ggplot(aes(x = artist, y = stickiness, fill = stickiness)) + geom_col_interactive( aes( tooltip = paste0( "", artist, " ", "Stickiness: ", round(stickiness, 3), " ", "Days heard: ", total_days, " / ", span, " day span ", "Total plays: ", scales::comma(total_plays), " ", "First listen: ", format(first_listen, "%b %Y") ), data_id = artist ), alpha = 0.9 ) + coord_flip() + scale_fill_gradient(low = "#4A90E2", high = "#1DB954", guide = "none") + scale_y_continuous(labels = scales::percent_format(accuracy = 1), expand = expansion(mult = c(0, 0.1))) + labs( title = "Top 15 Stickiest Artists", subtitle = "Artists you've listened to on the highest fraction of days since discovery (min. 10 listen-days)", x = NULL, y = "Stickiness Score" ) + theme_minimal() + theme( plot.title = element_text(size = 15, face = "bold"), plot.subtitle = element_text(size = 11, color = "gray60"), panel.grid.major.y = element_blank() ) make_girafe(p, height_svg = 6) ``` ## Long-running Favorites > Artists with the longest active span (days between first and last listen). ```{r long-running} #| fig-width: 12 #| fig-height: 6 long_running <- lifecycle %>% mutate(span_days = as.integer(last_listen - first_listen)) %>% filter(total_plays >= 5) %>% arrange(desc(span_days)) %>% head(15) %>% mutate(artist = factor(artist, levels = rev(artist))) p <- long_running %>% ggplot(aes(x = artist)) + geom_segment_interactive( aes( xend = artist, y = first_listen, yend = last_listen, tooltip = paste0( "", artist, " ", "First: ", format(first_listen, "%b %Y"), " ", "Last: ", format(last_listen, "%b %Y"), " ", "Span: ", span_days, " days ", "Total plays: ", scales::comma(total_plays) ), data_id = as.character(artist) ), linewidth = 3, color = "#1DB954", alpha = 0.7 ) + geom_point_interactive( aes(y = first_listen, tooltip = paste0("", artist, " First listen: ", format(first_listen, "%b %d, %Y")), data_id = paste0("first-", artist)), color = "#4A90E2", size = 2.5 ) + geom_point_interactive( aes(y = last_listen, tooltip = paste0("", artist, " Last listen: ", format(last_listen, "%b %d, %Y")), data_id = paste0("last-", artist)), color = "#FF6B6B", size = 2.5 ) + coord_flip() + scale_y_date(date_labels = "%b %Y", date_breaks = "6 months") + labs( title = "Artist Active Spans — Top 15 Longest Running", subtitle = "Blue = first listen · Red = last listen · Green bar = active period", x = NULL, y = NULL ) + theme_minimal() + theme( plot.title = element_text(size = 15, face = "bold"), plot.subtitle = element_text(size = 11, color = "gray60"), panel.grid.major.y = element_blank(), axis.text.x = element_text(angle = 45, hjust = 1) ) make_girafe(p, height_svg = 6) ``` ## Comeback Artists > Artists who disappeared for 90+ days and then returned. More comebacks = a more cyclical listening pattern. ```{r comeback-artists} #| fig-width: 12 #| fig-height: 6 comebacks <- lifecycle %>% filter(comeback_count > 0) %>% arrange(desc(comeback_count), desc(total_plays)) %>% head(15) %>% mutate(artist = factor(artist, levels = rev(artist))) if (nrow(comebacks) > 0) { p <- comebacks %>% ggplot(aes(x = artist, y = comeback_count, fill = comeback_count)) + geom_col_interactive( aes( tooltip = paste0( "", artist, " ", "Comebacks: ", comeback_count, " ", "Max gap: ", max_gap_days, " days ", "Total plays: ", scales::comma(total_plays), " ", "First: ", format(first_listen, "%b %Y"), " · Last: ", format(last_listen, "%b %Y") ), data_id = as.character(artist) ), alpha = 0.9 ) + coord_flip() + scale_fill_gradient(low = "#FFD700", high = "#FF6B6B", guide = "none") + scale_y_continuous(breaks = scales::pretty_breaks(n = 6), expand = expansion(mult = c(0, 0.1))) + labs( title = "Top Comeback Artists", subtitle = "Artists you've rediscovered after gaps of 90+ days", x = NULL, y = "Number of Comebacks" ) + theme_minimal() + theme( plot.title = element_text(size = 15, face = "bold"), plot.subtitle = element_text(size = 11, color = "gray60"), panel.grid.major.y = element_blank() ) make_girafe(p, height_svg = 6) } else { cat("No comeback artists found yet (no gaps > 90 days followed by return).") } ``` ## Drifted Away > Artists you used to love but haven't listened to in 180+ days — sorted by how much you played them historically. ```{r drifted-away} #| message: false drifted <- lifecycle %>% filter(!is_active, last_listen < max_date - days(180), total_plays >= 5) %>% arrange(desc(total_plays)) %>% head(30) %>% mutate( days_since = as.integer(max_date - last_listen), year_month_last = format(last_listen, "%B %Y") ) if (nrow(drifted) > 0) { drifted %>% select(artist, total_plays, total_days, days_since, year_month_last, max_gap_days) %>% gt() %>% tab_header( title = "Artists You've Drifted Away From", subtitle = paste0("Not heard in 180+ days · Sorted by historical plays · Max ", nrow(drifted), " shown") ) %>% cols_label( artist = "Artist", total_plays = "Total Plays", total_days = "Days Heard", days_since = "Days Since Last", year_month_last = "Last Listened", max_gap_days = "Max Gap (days)" ) %>% fmt_number(columns = c(total_plays, total_days, days_since, max_gap_days), decimals = 0) %>% data_color( columns = days_since, colors = scales::col_numeric(palette = c("#FFF9C4", "#FF6B6B"), domain = NULL) ) %>% data_color( columns = total_plays, colors = scales::col_numeric(palette = c("#E3F2FD", "#1DB954"), domain = NULL) ) %>% opt_interactive(use_search = TRUE, use_filters = TRUE, page_size_default = 10) } else { cat("No drifted-away artists found yet.") } ``` ## Full Artist Lifecycle Table ```{r lifecycle-table} #| message: false lifecycle %>% mutate( span_days = as.integer(last_listen - first_listen) + 1L, days_since = as.integer(max_date - last_listen), status = if_else(is_active, "Active", "Inactive") ) %>% select(artist, status, first_listen, last_listen, total_plays, total_days, stickiness, comeback_count, max_gap_days) %>% gt() %>% tab_header( title = "All Artists", subtitle = paste0(scales::comma(nrow(lifecycle)), " artists in listening history") ) %>% cols_label( artist = "Artist", status = "Status", first_listen = "First Heard", last_listen = "Last Heard", total_plays = "Plays", total_days = "Days", stickiness = "Stickiness", comeback_count = "Comebacks", max_gap_days = "Max Gap" ) %>% fmt_number(columns = c(total_plays, total_days, comeback_count, max_gap_days), decimals = 0) %>% fmt_number(columns = stickiness, decimals = 3) %>% fmt_date(columns = c(first_listen, last_listen), date_style = "wday_month_day_year") %>% data_color( columns = stickiness, colors = scales::col_numeric(palette = c("#E3F2FD", "#1DB954"), domain = c(0, 1)) ) %>% tab_style( style = list(cell_text(color = "#1DB954", weight = "bold")), locations = cells_body(columns = status, rows = status == "Active") ) %>% opt_interactive(use_search = TRUE, use_filters = TRUE, page_size_default = 15) ``` --- *Last updated on `r Sys.Date()`*