# Changelog of YouTube Community Posts & Polls Scraper (`sian.agency/youtube-community-posts-scraper`) Actor

- **URL**: https://apify.com/sian.agency/youtube-community-posts-scraper/changelog.md
- **Full Actor documentation**: https://apify.com/sian.agency/youtube-community-posts-scraper.md

## Changelog

All notable changes to **YouTube Community Posts & Polls Scraper** will be documented in this file.

### \[2026-09-11]

#### 🖼️ Images

- 🖼️ **New `thumbnailUrl`.** Each row now carries a preview image, taken from the first photo on the listing, so the Output tab shows a picture per row.
- 🖼️ **The photos were always collected — now you can see them.** The full image gallery and its preview thumbnail appear in the Output tab and in every export, instead of only in the raw JSON.

### \[2026-05-22]

#### 🎉 YouTube Community Posts & Polls Scraper — Launch!

- **All-Three-Ops Bundle** — Channel community-posts listing + post detail enrichment + post comment threads in a single actor (no chaining required)
- **Polls Fully Structured** — Parse `pollChoices[]` array + `pollTotalVotes` integer so you stop scraping poll blobs by hand
- **Multi-Image Carousel Unwrap** — Get a flat `imageUrls[]` of best-resolution URLs whether a post has 1 image or 10
- **Handle Resolution Built-In** — Paste `@MKBHD`, get the canonical `UC…` channel ID free — no pre-resolution step required
- **`commentsContinuationToken` USP** — Every `postDetail` row carries a chain-into-comments token that saves 1 upstream call when fanning out
- **44 Typed Output Fields** — Clean camelCase columns with consistent types, plus an `extra` blob for any unmapped upstream fields
- **Bulk Channel + Bulk Post Support** — Process up to 200 channels per listing run, or up to 500 post IDs per detail/comments run
- **HTML Run Report** — Every run saves a clean HTML summary (post-type breakdown, row counts, channel meta) to the key-value store — even on fatal errors

#### 💎 User Benefits

- **For creator-management agencies**: track 30-channel talent rosters in a single weekly run instead of stitching 3 separate scrapers
- **For brand-sentiment researchers**: aggregate poll-result signals before sponsorship deals without manual transcription from screenshots
- **For AI training pipelines**: collect long-form creator text + structured poll data at scale, with consistent typed columns
- **For social-listening platforms**: ingest cross-creator poll trends via clean JSON rather than scraping HTML
- **For academic researchers**: walk post → comment-thread graphs with full reply counts and channel-owner reply flags

#### 🎯 Use Cases

- Creator-management agency analysts tracking community engagement across 30+ talent channels in one weekly run
- Brand-sentiment researchers aggregating poll-result data from creator community tabs before signing sponsorship deals
- AI training data engineers bulk-collecting long-form text + structured poll signals for model training corpora
- Social-listening platforms aggregating poll choices across competing creators to detect emerging audience preferences
- Academic researchers studying creator-audience interaction dynamics via post → comment thread engagement graphs