How to Export a Weibo Main-Feed Snapshot

The Weibo Scraper exports posts from Weibo's main feed with text, engagement, media, and public author information. I maintain the Actor. The current input lets you set a post limit and proxy configuration; it does not offer keyword, profile, hashtag, or date targeting.

Use this workflow for a feed snapshot you can inspect and categorize. It is not a method for collecting every mention of a brand or every post by a chosen account.

Understand what the input can select

limit is the content-scope control and defaults to 100. The example requests 50 posts. Raising that number asks for a larger feed sample, not a different topic or population.

Keep the proxy enabled. The deployed schema says Weibo does not respond to Apify's shared platform IPs without it. The default datacenter proxy is the starting configuration; residential access is a fallback to consider if the log shows blocks.

No Weibo cookies or profile URL belong in this example's input. Adding unrecognized search fields will not turn the Actor into a targeted search scraper. If your task depends on a particular account or date range, choose a tool that explicitly supports that scope.

Budget for the sample

Pricing as of September 6, 2026. The active price is $0.001 per returned post. Fifty posts would cost $0.05. Proxy usage may be billed separately by Apify. A pricing change is scheduled for 2026-09-09; recheck the rate before a later run or recurring schedule. Check current Actor pricing and the run charge breakdown; these calculations describe Actor events, and separately billed platform or proxy resources may also apply.

Export fifty feed posts

Open the Actor on Apify. In Input, switch to the JSON editor and paste:

{
  "limit": 50,
  "proxyConfiguration": {
    "useApifyProxy": true
  }
}

Click Start, then open the completed run’s output dataset.

After completion, review text and post links in the dataset, then compare the returned count with the requested limit. A short or empty result needs a log review before you infer anything about feed activity.

Export JSON first to retain Chinese text, nested author fields, and media arrays. If you use CSV, import it with UTF-8 encoding and check a few Chinese names and paragraphs for corruption.

Read the returned fields

The historical sample has author.followersCount: null, despite other author fields being present. Do not turn that null into zero or compute an engagement rate that divides by it. A public author object is not a full profile scrape.

createdAt can be a source-formatted timestamp containing a timezone, such as +0800. Preserve the original and parse it into a separate normalized date field. Media links in the sample contain expiry parameters, so keep the source post URL as the durable reference rather than relying on every image or video URL remaining usable.

Turn the snapshot into a content-review table

Create a working table with postId, postUrl, text, author ID and name, publication time, likes, comments, and shares. Add your own collection timestamp because the documented post sample does not include one. Keep long IDs as text to avoid spreadsheet rounding.

Deduplicate by post ID. Add manual topic labels after reading each post, preserving the original Chinese text beside any translation. Review translations for names, slang, irony, and references before assigning sentiment. An automatic label without that context can misread a post.

Summarize topics as counts within this collected feed sample. Keep the collection date and sample size beside the chart. Do not generalize those proportions to all Weibo users or describe the feed as a representative opinion survey.

If a post merits deeper investigation, open its source URL and review the surrounding context. A comments count tells you how many comments are displayed, not what those comments say. This export does not establish the sentiment of the discussion below a post.

For repeated observation, save each snapshot and match post IDs. Preserve multiple observations when comparing engagement changes. A post absent from the next feed snapshot may simply no longer be selected for that feed, rather than deleted or unpopular.

Troubleshoot connections and missing data

For no output, inspect connection and proxy errors first. There is no query or date setting to broaden here. Verify that the proxy remains enabled before increasing the limit or changing proxy type.

Missing follower counts and media arrays can be valid source limitations. Do not infer that an author has no followers or that an empty image array means the post has no video. Examine images and videos separately, and treat expired media URLs as a reason to return to the post page.

For saving Weibo feed snapshots, see the Node.js example. Use the current input fields above if the example differs.

Frequently asked questions

Can this Actor search a keyword or profile?

The current deployed input exposes only limit and proxyConfiguration. It does not support keyword, hashtag, profile, or date targeting.

Is a proxy optional?

The deployed schema says a proxy is required for access from Apify. Start with the default enabled configuration.

Can I calculate engagement rates from follower counts?

Not reliably when follower counts are missing. The documented sample returns author.followersCount as null, which must not be treated as zero.