> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.beatapi.io/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.beatapi.io/_mcp/server.

# How should I choose a data workflow for WeChat, Channels, Douyin or Xiaohongshu?

> Match WeChat, Channels, Douyin, and Xiaohongshu tasks to authorized reports or BeatAPI public actions, and compare workflows on one small sample before scaling.

Choose the data source by task. To review your own account, start with the management reports you can access or with an application that holds the required grants. To research public content, read a small sample with BeatAPI actions first, then decide whether recurring collection is worth planning. For example, to plan content for a storage product, first find relevant works and viewer questions, then decide whether you need account inventories or metric snapshots.

## What can your source answer?

| Task                                              | Starting point                                              | What to verify                                         |
| ------------------------------------------------- | ----------------------------------------------------------- | ------------------------------------------------------ |
| Review your own publishing, audience or campaigns | Available management reports or an authorized application   | Permission, metric definitions, dates and export scope |
| Read a few known public works                     | Original works or links, with details or comments as needed | Work identity, relevant wording and context            |
| Find repeatable public samples by topic           | Defined keywords, platforms and pagination                  | Objects returned, gaps, duplicates and stop reasons    |
| Reuse an existing collection program              | Run the parser and export on one saved response             | Maintenance, failures, field changes and repeatability |

For an application integration, check the scopes of the actual app and the actual grant to see which accounts and objects the application can read. Signing users in and reading their authorized profiles is a different task from finding public discussion about rental storage, and the two tasks need different data routes.

## Where do I start on each platform?

* **Xiaohongshu**: For public notes, use [keyword search](/guides/xiaohongshu-notes-research), or a [creator inventory](/guides/xiaohongshu-creator-note-inventory) for an identified creator. Retrieve note details or comments after selecting notes. For collaboration quotes and audience evidence, see the [Pugongying shortlist](/guides/xiaohongshu-creator-shortlist).
* **Douyin**: Choose a keyword, account or work route with the [Douyin data workflows](/faq/douyin). If you already have comparable index exports, use the index exports for [topic and history review](/guides/douyin-topic-signal-review).
* **Channels**: Use the [account inventory](/guides/wechat-channels-video-inventory) for an identified account, or [WeChat Search](/guides/wechat-topic-search) to find candidates for a topic. Confirm the work ID before requesting comments.
* **Official Accounts**: Start with the account's [article history](/guides/wechat-account-articles), or use a confirmed article link to read the article body, statistics or comments. See [Official Account data routes](/faq/wechat-data-options).

## Compare workflows on the same small sample

For example, for a weekly storage roundup:

1. Start with one term and one page.
2. Record the pages requested, usable objects, verifiable sources, missing columns, candidates that still need details and the review time.
3. Deliver the initial sample once the sample answers the research question.
4. When replacing an existing program, rerun the old export on the same saved material and check that the output columns and identifiers still work.

Estimate data calls, maintenance and review time separately. Count the actual list, detail and comment requests with [pricing](https://beatapi.io/pricing). Compare workflows against the same deliverable, such as ten sourced candidates and three quotable viewer questions.

Record the old and the new workflow in one sheet: total candidates, candidates with traceable sources, candidates still needing details and time spent. If workflow A returns ten candidates and three of the ten lack a source, while workflow B returns eight candidates that all have sources, choose by the number of verifiable candidates your team needs and the effort needed to fill the gaps.

## What should I retrieve when I already have a dashboard report?

Start from the field you need to verify. When a dashboard from your own or an authorized source gives ranks, trends and estimates, save each metric's definition, dates and coverage. When the dashboard lacks original evidence, retrieve the selected public works or reviews. Compare similarly named metrics from Xingtu, Pugongying or index tools only after checking each source’s definition and purpose. For third-party platforms such as Chanmama, Qiangua or NewRank, the available access depends on your installed plan or service contract.

Before choosing a workflow, run one saved sample through the full handoff. Keep the request ID, received rows, gaps, reproducibility and measured usage, and verify charges against your own request records. The sample run shows which steps deliver usable evidence and which steps need manual review.

[Choose a Xiaohongshu data source](/guides/xiaohongshu-data-access) · [Choose a Douyin data source](/guides/douyin-data-access)

[Choose a Channels data source](/guides/wechat-channels-data-access) · [WeChat articles and statistics](/faq/wechat-data-options)