Working with Custom Fields Filters

In sensortowerR: Interface to 'Sensor Tower' Mobile App Intelligence API

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  eval = FALSE,
  purl = FALSE
)

library(sensortowerR)
library(dplyr)

Introduction

Custom Fields Filters are a powerful feature in Sensor Tower that allow you to create complex, reusable queries to segment apps based on hundreds of different attributes. This vignette demonstrates how to use the sensortowerR package to work with custom fields filters.

Understanding Custom Fields

Sensor Tower tracks over 400 custom fields for apps, including:

• Game attributes: Genre, sub-genre, art style, camera POV
• Monetization: Free/paid, IAP, ads, subscriptions
• Technical: SDKs used, platform support, device compatibility
• Performance: Retention rates, ratings, download counts
• Metadata: Release dates, publisher info, update frequency

Basic Workflow

1. Discover Available Fields

First, explore what custom fields are available:

# See all available fields
all_fields <- st_discover_fields()
head(all_fields)

# Search for specific fields
game_fields <- st_discover_fields("game")
monetization_fields <- st_discover_fields("monetization")

# Show fields with their possible values
genre_fields <- st_discover_fields("genre", show_values = TRUE)

2. Create Simple Filters

Create filters for common scenarios:

# Filter for Word games
word_filter <- st_create_simple_filter(
  field_name = "Game Sub-genre",
  field_values = "Word"
)

# Filter for free apps with ads
free_with_ads <- st_create_simple_filter(
  field_name = "Free",
  field_values = list(),  # Boolean field
  global = TRUE
)

3. Retrieve Filtered Apps

Use filters to get app data:

# Get top Word games by DAU
word_games <- st_get_filtered_apps(
  field_name = "Game Sub-genre",
  field_values = "Word",
  measure = "DAU",
  regions = "US",
  limit = 20
)

# Use an existing filter ID
apps <- st_get_filtered_apps(
  filter_id = word_filter,
  measure = "revenue",
  regions = "US",
  date = "2025-07-01",
  end_date = "2025-07-31"
)

Advanced Filtering

Genre and Sub-Genre Filtering

# Get Puzzle games
puzzle_filter <- st_filter_by_genre(genres = "Puzzle")

# Get Word Puzzle games (multiple criteria)
word_puzzle_filter <- st_filter_by_genre(
  genres = "Puzzle",
  sub_genres = "Word"
)

# Exclude certain genres
no_action_filter <- st_filter_by_genre(
  genres = c("Action", "Shooter"),
  exclude_genres = TRUE
)

Monetization Model Filtering

# Free-to-play with ads
f2p_ads <- st_filter_by_monetization(
  free_only = TRUE,
  has_ads = TRUE
)

# Premium games (paid, no IAP)
premium <- st_filter_by_monetization(
  free_only = FALSE,
  has_iap = FALSE
)

# Subscription-based apps
subscription <- st_filter_by_monetization(
  has_subscription = TRUE
)

# Hybrid monetization (IAP + Ads)
hybrid <- st_filter_by_monetization(
  has_iap = TRUE,
  has_ads = TRUE
)

SDK and Technology Filtering

# Unity-based games
unity_games <- st_filter_by_sdk(sdk_names = "Unity")

# Apps using multiple SDKs
firebase_admob <- st_filter_by_sdk(
  sdk_names = c("Firebase", "AdMob")
)

# Apps NOT using certain SDKs
no_facebook <- st_filter_by_sdk(
  sdk_names = "Facebook",
  exclude = TRUE
)

Date-Based Filtering

# New releases (last 30 days)
new_releases <- st_filter_by_date(
  released_after = Sys.Date() - 30,
  region = "US"
)

# Apps released in 2024
apps_2024 <- st_filter_by_date(
  released_after = "2024-01-01",
  released_before = "2024-12-31",
  region = "WW"
)

# Established apps (>1 year old)
established <- st_filter_by_date(
  released_before = Sys.Date() - 365,
  region = "US"
)

Publisher Filtering

# Apps from specific publishers
ea_games <- st_filter_by_publisher(
  publisher_names = c("Electronic Arts", "EA Swiss Sarl")
)

# Exclude certain publishers
indie_games <- st_filter_by_publisher(
  publisher_names = c("Electronic Arts", "Activision", "Ubisoft"),
  exclude = TRUE
)

Working with Filter Collections

Use pre-built filter collections for common analyses:

# Get genre filter collection
genres <- st_get_filter_collection("top_genres")

# Use filters from collection
puzzle_apps <- st_top_charts(
  os = "unified",
  category = 0,
  custom_fields_filter_id = genres$puzzle,
  custom_tags_mode = "include_unified_apps",
  measure = "DAU",
  regions = "US"
)

# Get monetization model filters
monetization <- st_get_filter_collection("monetization_models")

# Compare F2P with ads vs F2P with IAP
f2p_ads_apps <- st_get_filtered_apps(
  filter_id = monetization$free_with_ads,
  measure = "DAU",
  regions = "US"
)

f2p_iap_apps <- st_get_filtered_apps(
  filter_id = monetization$free_with_iap,
  measure = "DAU",
  regions = "US"
)

Complex Filter Combinations

Create sophisticated filters by combining multiple criteria:

# Create a complex filter manually
complex_filter <- st_custom_fields_filter(
  custom_fields = list(
    list(
      name = "Game Genre",
      values = list("Puzzle"),
      global = TRUE,
      exclude = FALSE
    ),
    list(
      name = "Free",
      global = TRUE,
      true = TRUE
    ),
    list(
      name = "SDK: Unity",
      global = TRUE,
      true = TRUE
    ),
    list(
      name = "Day 7 Retention (Last Quarter, US)",
      values = list("30% - 100%"),  # High retention
      global = TRUE,
      exclude = FALSE
    )
  )
)

# Use the complex filter
high_retention_puzzle <- st_get_filtered_apps(
  filter_id = complex_filter,
  measure = "DAU",
  regions = "US"
)

Analyzing Filtered Results

Once you have filtered apps, analyze them:

# Get comprehensive analysis
analysis <- st_analyze_filter(
  filter_id = word_filter,
  measure = "DAU",
  regions = "US",
  top_n = 10
)

# Access analysis components
print(analysis$filter_criteria)
print(analysis$total_apps)
print(analysis$top_apps)

# Custom analysis on filtered data
word_games %>%
  mutate(
    dau = dplyr::coalesce(dau_30d_us, users_absolute),
    retention_d7 = dplyr::coalesce(retention_7d_us, `aggregate_tags.Day 7 Retention (Last Quarter, US)`)
  ) %>%
  filter(!is.na(dau)) %>%
  summarise(
    total_dau = sum(dau, na.rm = TRUE),
    avg_retention = mean(retention_d7, na.rm = TRUE),
    median_dau = median(dau, na.rm = TRUE)
  )

Integration with Other sensortowerR Functions

Custom fields filters integrate seamlessly with other package functions:

# Use with st_top_charts
top_word_games <- st_top_charts(
  os = "unified",
  category = 0,  # Required when using custom filter
  custom_fields_filter_id = word_filter,
  custom_tags_mode = "include_unified_apps",
  measure = "DAU",
  regions = "US",
  date = "2025-07-01",
  limit = 50
)

# Use with st_app_tag
matching_apps <- st_app_tag(
  app_id_type = "unified",
  custom_fields_filter_id = word_filter
)

# Combine with metrics retrieval
if (nrow(matching_apps$data) > 0) {
  app_metrics <- st_metrics(
    app_ids = matching_apps$data$unified_app_id[1:10],
    metrics = c("dau", "revenue", "downloads"),
    regions = "US"
  )
}

Best Practices

1. Cache Filter IDs

Filter IDs are permanent and reusable. Save them for repeated use:

# Save commonly used filters
my_filters <- list(
  word_games = "603697f4241bc16eb8570d37",
  puzzle_games = st_filter_by_genre(genres = "Puzzle"),
  f2p_games = st_filter_by_monetization(free_only = TRUE),
  unity_games = st_filter_by_sdk(sdk_names = "Unity")
)

# Save to file for later use
saveRDS(my_filters, "my_sensor_tower_filters.rds")

# Load in future sessions
my_filters <- readRDS("my_sensor_tower_filters.rds")

2. Validate Filters

Always verify what a filter contains:

# Check filter details
filter_details <- st_custom_fields_filter_by_id(
  id = "603697f4241bc16eb8570d37"
)

print(filter_details$custom_fields)

3. Handle Large Result Sets

When working with broad filters:

# Use pagination for large result sets
all_puzzle_games <- st_app_tag(
  app_id_type = "unified",
  custom_fields_filter_id = puzzle_filter
)

# Process in batches if needed
if (!is.null(all_puzzle_games$last_known_id)) {
  # Get next page
  next_page <- st_app_tag(
    app_id_type = "unified",
    custom_fields_filter_id = puzzle_filter,
    last_known_id = all_puzzle_games$last_known_id
  )
}

Troubleshooting

Common Issues

1. Empty Results: Check that the filter criteria are not too restrictive
2. API Errors: Ensure your authentication token has necessary permissions
3. Performance: Use specific filters rather than broad ones for better performance

Debugging Filters

# Test filter creation
tryCatch({
  filter_id <- st_create_simple_filter(
    field_name = "Invalid Field Name",
    field_values = "test"
  )
}, error = function(e) {
  message("Filter creation failed: ", e$message)
})

# Verify field exists
fields <- st_discover_fields("Invalid Field")
if (nrow(fields) == 0) {
  message("No matching fields found")
}

Conclusion

Custom Fields Filters provide powerful segmentation capabilities for app market analysis. By combining different filter criteria, you can create precise queries to identify specific app segments, analyze market trends, and track competitor strategies.

Key takeaways:

• Use st_discover_fields() to explore available filtering options
• Create reusable filters with utility functions like st_filter_by_genre()
• Combine filters for complex queries
• Integrate filters with other sensortowerR functions for comprehensive analysis
• Cache filter IDs for improved performance

For more information, see the function documentation:

?st_custom_fields_filter
?st_get_filtered_apps
?st_filter_by_genre

sensortowerR documentation built on March 18, 2026, 5:07 p.m.