News API - Blogs, Forums and Reviews Search, Webz.io

Search and export news articles, blog posts, forum discussions and customer reviews from across the open web into an Apify dataset, with your own Webz.io API token.

This actor wraps the Webz.io News, Blogs, Forums and Reviews APIs. It turns a Boolean query and a set of filters into clean, deduplicated rows you can download as JSON, CSV or Excel, schedule, or feed into other actors and AI agents.

What it does

• News article search: Boolean queries over news sites worldwide, with full extracted text, summary, sentiment, IPTC categories and topics, named entities with their own sentiment, stock tickers, source trust data, syndication and social engagement counts.
• Blog post search: the same query language over blogs, with named entities and tickers.
• Forum post search: discussions and replies, filtered by thread starters or replies and by reply count.
• Review search: customer reviews of products, hotels, apps and businesses, with rating, review text and the reviewed item.
• Filters as fields: languages (76), source countries, sites to include or exclude, domain rank, authors, news categories and topics, sentiment, organizations, people, locations, tickers, political bias, trust category, curated top news lists, breaking news and engagement.
• Time window: the last 3 days by default, any start within the last 31 days, and news history back about a year.
• Automatic paging: follows the provider's own next link until the results run out or your row cap is reached.
• Pay only for results: empty searches are recorded as a free row that says why, never charged.

Input

Field	Used by	What it does
mode	all	news, blogs, forums or reviews
queries	all	One Boolean query per line, read separately. Leave empty to search by filters only
since	all	Start of the window: 2026-09-30, a date and time, 24h or 7d. Empty means the last 3 days
newsHistory	news	Allows a start more than 31 days back (extra provider credits)
languages, countries, sites, excludeSites, domainRankMax, authors	all	Language names, country codes such as US, domains, Tranco rank ceiling, author names
categories, topics, sentiment	news	IPTC categories, finer topics, positive, negative or neutral
organizations, people, locations, tickers	news, blogs	Entities and stock tickers the post mentions
trustBias, trustCategory, topNews, breakingOnly, performanceScoreMin	news	Source bias, trusted or fake news, curated top news lists, breaking stories, engagement score
forumPosts, minReplies	forums	Thread starters or replies, minimum replies
ratingMin, ratingBelow, minItemReviews, highConfidenceOnly	reviews	Rating range, popular items, high accuracy parsers only
includeSyndicated	news, blogs, forums	Also return syndicated copies
sortBy, order	all	Crawl time (default), relevance, published, domain rank, rating, social shares and more
maxResults	all	Row ceiling for the run, across all queries
apiKey	all	Your Webz.io API token (secret). Or set the DATA_API_KEY environment secret

Every filter is checked against the provider's documented values before a request is sent. This matters because the API answers an unknown filter value with zero results rather than an error. Inputs a product does not use are left off and named in the log.

How to get an API token

Sign in at webz.io and copy the API token from your dashboard. Your account must be enabled for each product you search (News, Blogs, Forums or Reviews). If a product is not on your plan, the run stops with a clear message and you can ask Webz.io support to add it.

Example output

{
    "mode": "news",
    "recordType": "newsArticle",
    "identifier": "\"artificial intelligence\" regulation",
    "found": true,
    "id": "a83e94d134f35bc6439bdad39e318b4ef3319f8c",
    "title": "EU lawmakers agree on next steps for AI rules",
    "url": "https://www.example-news.com/tech/eu-ai-rules",
    "date": "2026-10-06T14:32:00.000+03:00",
    "source": "example-news.com",
    "author": "Jane Doe",
    "language": "english",
    "country": "BE",
    "sentiment": "neutral",
    "value": 1840,
    "siteType": "news",
    "text": "Full extracted article text ...",
    "summary": "European lawmakers reached a provisional deal on ...",
    "categories": ["Science and Technology"],
    "topics": ["Science and Technology->technology and engineering"],
    "entities": { "organizations": [{ "name": "European Commission", "sentiment": "neutral", "tickers": [] }] },
    "thread": { "site": "example-news.com", "domain_rank": 1840, "social": { "facebook": { "shares": 12 } } },
    "crawled": "2026-10-06T14:40:11.000+03:00",
    "note": null
}

Every field the provider returns is kept. value is the domain rank for news, blog and forum posts, and the rating for reviews.

FAQ

What is the Webz.io API?

Webz.io (formerly Webhose.io) is a web data provider that crawls news sites, blogs, forums and review sites and serves the content as structured JSON. Its News, Blogs and Forums APIs share one Boolean query language and response format, and its Reviews API uses the same query language. This actor calls /api/news, /api/blogs, /api/forums and /api/reviews on api.webz.io.

Do I need my own API token?

Yes. This actor is bring your own key: you need a Webz.io account with access to the products you want to search. Your token is stored as an Apify secret, sent only in the Authorization header, and removed from every log line. Without a token the run stops immediately with a clear message and no charge.

How do I write a query?

Enter one query per line, for example "electric vehicles" (Tesla OR BYD) -recall. A space means AND, operators must be uppercase, quotes and parentheses must be balanced, and you can add field filters such as title:bitcoin. The filter inputs add clauses like language:english, country:US or organization:nvidia for you.

How far back can I search?

News, blogs, forums and reviews are searchable for about the last 31 days, and the default window is the last 3 days. News history reaches back about one year when you tick news history, at 2 to 4 provider credits per request, and only in crawl order. Older data is sold separately by Webz.io as an archive product, which this actor does not cover.

How many results can I get?

Pages of up to 100 posts are read until your maxResults ceiling or until the provider reports no more results. Each page is one provider request and costs at least one credit from your Webz.io plan, and the provider allows about one request a second per token.

Can I filter by sentiment, company or person?

Yes, on news articles: overall sentiment, categories, topics, and organizations, people and locations mentioned. Blog posts take the entity and ticker filters too. Entities are extracted from English, German, Spanish and Italian posts.

Does it include segmentation counts?

No. The provider documents segmentation routes that count posts by site, language or entity, but on 2026-10-08 the live API host answered those routes with 404, so they are not wired. Only the four search routes that answer on the live host are used.

What do I pay?

The actor charges per returned record: 0.01 USD per news article, 0.008 USD per blog post, 0.006 USD per forum post and 0.008 USD per review, with Apify platform usage included. Empty results are free. Webz.io bills your plan separately in credits.

What happens if my token is wrong or my plan does not cover a request?

A rejected token, a product your account is not enabled for, an inactive account or ended trial, a plan-gated filter, or spent credits ends the run cleanly with the provider's own explanation as the status message. Rate limits are retried automatically.

Keyword map

news API, news search API, blog search API, forum search API, reviews API, customer reviews data, web data API, media monitoring, brand monitoring, social listening data, news sentiment, named entity news search, Webhose API, Webz.io API, news data export.

Terms and data use

You need your own Webz.io account, and your use of the data is governed by your agreement with Webz.io. Their published terms (webz.io/tos, sections 2 and 5) grant a non-transferable licence for internal use, forbid sharing or reselling the data with third parties including clients, forbid using it to compete with Webz.io, forbid catch-all queries that bulk-gather data, and forbid AI or LLM training on posts whose ai_allow field is false. Content belongs to the original publishers and authors, and a true ai_allow value is not a licence. Write targeted queries, check your agreement before republishing text or training models, and keep the export inside your organisation. This actor is an independent tool and is not made or endorsed by Webz.io.