ナビゲーションにスキップ

How do I collect a WeChat account’s articles, text, engagement and comments for research?

Build the account’s article directory with wechat_mp.v2.fetch_account_articles, select the relevant articles, then retrieve text, engagement data and available comments only for the selected article links. Choose the research task first: finding material, reading an article, comparing engagement or understanding reader questions. Each research task needs different material, listed in the table below.

For example, to study a competitor’s product launch, list the account’s articles for the launch period, select the announcements and read the announcement text. Add engagement data or comments only when engagement data or comments help answer the research question.

Where should you get analytics for an account you operate?

For an account you operate, take analytics from the WeChat Official Accounts Platform, not from the public article workflow.

  1. Sign in to the WeChat Official Accounts Platform with an account that has analytics access.
  2. Select the Official Account and the reporting dates, then check the available reader, read and share columns.
  3. If your team has access to an official analytics integration, compare one published article for the same dates in the integration and in the dashboard.

Tencent Cloud lists daily send statistics, article totals and audience changes as separate operations; choose the operation your app exposes. Official Accounts Platform · Tencent Cloud’s Official Accounts API overview

Keep the platform’s original field names for deliveries, unique readers or reads. For other publishers’ accounts, use the public article workflow below. Record owned analytics and public interaction snapshots in separate source columns.

Start with the information you have

  • An exact article link: Read the article. The account identifier returned with the article supports account identification and later directory retrieval.
  • A verified account identifier: Build the account’s historical list. Review titles, summaries, links and publication times before you select articles to read.
  • Only an account name: Search for account candidates, check each candidate’s description and published articles, then continue with a confirmed article link. Checking the published articles keeps accounts with identical names apart.

To find account candidates by name, use the account category of data:wechat_search.v2.fetch_search. Replace the keyword in this request body for POST /v1/social-data/call:

{"action":"wechat_search.v2.fetch_search","params":{"keyword":"BRAND_NAME","business_type":"account","offset":0,"raw":false}}

Use the search results to identify the right account, then follow the identifier workflow to verify the account’s actual account ID. Keep display names, author bylines and account identifiers in separate columns.

Match the research task to the material

Research taskMaterialHow to use the material
Understand what an account covered during a periodArticle directorySelect publication dates and classify articles by topic
Check what an article saysReturned text and sourcesRead the original wording and cite the conditions that support a conclusion
Compare engagement for selected articlesReadership and interaction dataRecord collection time and compare publication age, placement and field definitions
Organize reader questionsComments and selected reply threadsPreserve original wording and parent/reply context; distinguish questions, experiences and additions
Verify an account and its linked Channels identityAccount profile and association evidenceBuild a sourced roster, then research articles and videos separately

Engagement data records one observation at retrieval time. Repeated retrievals create multiple observations; a single result is not a historical curve. Describe comments by the scope the comment action returned. Selected comments help investigate specific questions but do not represent every reader’s opinion.

Registered ownership, verification documents and follower estimates are separate tasks that the current profile workflow does not cover. Original-article counts, Channels associations and readership do not substitute for ownership, verification or follower data.

How do you fetch metrics only for the selected articles?

  1. List the verified account’s articles with wechat_mp.v2.fetch_account_articles and raw:false. Keep each article link, app_msg_id, idx and the pages actually retrieved.
  2. For each selected article, POST this JSON body to https://api.beatapi.io/v1/social-data/call with BeatAPI Bearer authorization:
{"action":"wechat_mp.v2.fetch_article_stats","params":{"url":"<confirmed_article_url>","raw":true}}
  1. In a successful result, check data.read_num together with the visibility and retrieval status. Record a missing read count as missing, separately from H5 comment counts, and save the observation time.
  2. To measure change, repeat wechat_mp.v2.fetch_article_stats for the same article at a later time.

To quote readers, use the article comment workflow and report the selected-comment scope that the comment workflow returned. Metric fields and definitions

Produce a research handoff

Give the research assignment as a concrete instruction, for example:

Collect this account’s product-update articles from the past month. Build a directory, select relevant items and read their text. Summarize each update, its conditions and supporting wording. Add engagement data if a comparison needs it. Deliver a source table and a findings table, with unanswered questions listed separately.

The source table holds title, account, original link, publication time, collection time and available material. The findings table holds your interpretation, supporting excerpts and next actions. Colleagues read the summary in the findings table and check each claim against the source table.

For updates, continue with incremental collection. For a local knowledge base, use Markdown import. To replace an old tool, read exporter alternatives.

For example, suppose an account’s directory lists a release announcement and a migration guide. To learn which customers need to migrate, read the migration guide’s text. To compare reads, save statistics for both articles at the same observation point and record each article’s publication time. To find questions the migration guide left open, read the migration guide’s returned comments. Keep the passage, the metric timestamp and the comment source in separate columns.

Keep the account inventory and the evidence for selected articles in two sheets:

  • Inventory sheet: one row per article, with the original URL and the available summary.
  • Review sheet: saved body text, metric snapshots or returned comments, only for selected articles.

During the next update, reuse the inventory sheet, and open previously saved evidence when the saved evidence already answers the question.

Control retrieval costs

Get the article list first, and select articles before you retrieve text, engagement or comments. Check the current pricing page, and record list pages and additional article requests separately. Reuse saved responses; local organization and conversion need no new retrieval.

WeChat questions

Give your connected assistant a concrete task

Review recent installation tutorials from this verified Official Account. List titles, dates and links, then select two articles for body reading. Report requirements, steps and sources. Label summary-only items as leads. If engagement comparison is needed, retain collection times and metric states.

The assistant uses the preview to choose objects. For quotes, the assistant reads the saved result or retrieves the specific object. Ask the assistant to report which material was reviewed and the actual coverage. Task walkthrough · Connection and result reading.