Question 1

What does Datadory cover in application software data?

Accepted Answer

Twenty-four primary datasets in four layers: platform stores (3,194,924 cross-store app rows, 2M-plus Play listings, 61,196 Mac App Store apps), open-source infrastructure (200M+ repositories, 11.72M packages, 4,247 SPDX-declared Android builds), vendor directories (86,267 Capterra profiles, 2,239 G2 categories, 500,000+ SourceForge projects) and a macro band from FRED, Eurostat, BIS, ECB and FFIEC. One adjacent record - PyPI - brings the working set to 25.

Question 2

How complete is cross-store app coverage?

Accepted Answer

The widest single table unifies 3,194,924 apps - 1.70 million from Google Play and 1.50 million from the Apple App Store - in 15 shared columns covering price, installs, ratings, category, developer and both monetization flags. Row-level stamps run through March 2026, and the listing dates underneath reach back to September 2008.

Question 3

Can I get install counts for Google Play apps?

Accepted Answer

Three ways. The live web catalog publishes an install bracket per listing - Spotify shows 1B+. The Extended snapshot records exact minimum and maximum installs for 2.3 million apps. And the cross-store metrics table carries an installs column across its 1.70 million Google rows. Bracket, bounds and point estimate, between them.

Question 4

Which datasets map npm and PyPI package dependencies?

Accepted Answer

Libraries.io indexes 11.72 million packages across 32 registries - npm 5.96 million, PyPI 908,000, Go 808,000, Maven 804,000, NuGet 681,000 - as seven related tables of projects, versions, tags, dependencies and repositories. The adjacent PyPI record adds 877,215 Python projects with 9.42 million releases and 44.8 TB of package weight.

Question 5

How far back do browser and operating-system share series run?

Accepted Answer

To January 2009, aggregated daily, weekly, monthly, quarterly and yearly across worldwide, continental and roughly 250 country-level scopes. The July 2026 worldwide browser table put Chrome at 68.22 percent and Safari at 16.47 - seventeen years of platform-shift evidence in one consistent shape.

Question 6

What trains an app rating or sentiment model?

Accepted Answer

Two standard benches. The Google Play snapshot pairs 10,841 app rows with 64,295 translated reviews scored for sentiment and polarity; the Apple 10k set contributes 7,197 iOS apps across 16 attributes with a joined description file. Scale up to the 2.3-million-row Extended table once the baseline holds.

Question 7

How deep does software directory coverage go?

Accepted Answer

Capterra documents 86,267 products across 1,025 category pages, each profile carrying an overall rating anchored by its review count, a stated starting price and buyer-rated feature sub-scores. G2 maps 2,239 segments under 43 top-level groups. SourceForge holds 500,000-plus projects with facets that restate their own counts - 202,909 Windows-tagged, 48,464 production-stable.

Question 8

How is application software data delivered?

Accepted Answer

Daily, weekly, or hourly - your call, as API pulls, bulk files, or warehouse-native writes. Every record lands as typed rows with stable column names and the field dictionary attached, so raising a feed's cadence is configuration rather than a rebuild. Name the stores, categories and ecosystems you need and a scoped sample goes out first.

The manifest

Play and Apple App Store Metrics (Hugging Face - appGoblin)

GitHub REST API (Repositories & Search)

Apple iTunes Search API

F-Droid Open Source Android Apps

Google Play Store (Web Catalog)

Mac App Store Apps Metadata (Hugging Face - macpaw-research)

The lay of the water

Pick your catch

Who fishes here

Frequently asked questions

See the rows before you commit