Reviews + AI Insights: Steam, Agoda, Google Play

LMSlay3 days ago

TubeText maintainer note (AI-assisted): comparing two Google Play review exports? A larger complaint percentage can come from a different sample window or denominator—not necessarily a worse app. Our free-to-read guide shows how to keep a shared collection brief, separate usable review rows from summaries/errors, define a small theme codebook, and report each theme as a count plus its sample denominator. It includes a blank analyst worksheet and clearly fictional examples: https://tubetext.dev/google-play-review-theme-comparison/ Start with Play Console's own review tools if you manage the app, or use an existing permitted export; the method doesn't require running this connector. The worksheet labels are analyst-added, not MCP/Actor output. Optional AI insights may cover a different subset, so don't assume they analyzed your filtered sheet. This is a documentation-based guide, not a live-tested workflow or a claim about all users. Collection and optional AI use have separate Apify charges; check current pricing before running. This resource note does not establish connector health or successful execution.

LMSlay2 days ago

Reading a Steam review sample beside the store score Our existing scope checklist is for readers comparing a displayed Steam score with a selected review list: https://github.com/LMSlay/tubetext-examples/blob/main/recipes/steam-review-score-scope-checklist.md Keep the score's language, recent/overall label, observation time and review count with it. For an exported or manually selected list, separately retain the language/date/recommendation filters, sort order, returned count and exclusions. That list is not automatically the population behind the headline score. The guide's fictional example has 9 of 30 manually chosen negative reviews mentioning crashes. That describes 30% of this selected sample, not 30% of players. An AI-generated theme likewise needs review against the original passages; it is not a new Steam score or proof of what caused a score change. Start with the native page and a small scope record when those answer the question. A paid export is optional. This static guide does not establish that our Actor reproduces Steam's scoring denominator, returns language-specific headline scores or produces a representative sample. No new collection, Actor run or MCP integration test was performed for this note. Prepared by TubeText Labs with AI assistance. Preserve source links and reviewers' rights. Consult current Store documentation and pricing before any paid collection.