We extract articles, live sports statistics, entertainment feeds, and multimedia metadata across the Globo network. Delivered as clean JSON, CSV, or Parquet to S3, BigQuery, or Snowflake on your cadence.
Structured, schema-consistent data across all major object types — delivered clean, typed, and ready to query.
Complete list of extractable fields for G1 News Articles objects from globo.com. All fields typed and schema-versioned.
"article_id": "g1-8492018", "headline": "Copom mantém taxa Selic em 10,50% ao ano", "author": "G1 Economia", "publish_date": "2026-05-12T18:30:00Z", "region": "Nacional", "category": "Economia", "is_paywalled": false
| # | article_id | url | headline | subheadline | author | publish_date |
|---|---|---|---|---|---|---|
| 1 | ||||||
| 2 | ||||||
| 3 |
Complete list of extractable fields for ge.globo Sports objects from globo.com. All fields typed and schema-versioned.
"match_id": "ge-bra-1029", "tournament": "Brasileirão Série A", "team_home": "Flamengo", "team_away": "Palmeiras", "score_home": 2, "score_away": 1, "status": "IN_PROGRESS", "minute": "78"
| # | match_id | tournament | team_home | team_away | score_home | score_away |
|---|---|---|---|---|---|---|
| 1 | ||||||
| 2 | ||||||
| 3 |
Complete list of extractable fields for gshow Entertainment objects from globo.com. All fields typed and schema-versioned.
"article_id": "gs-991823", "title": "Resumo da novela: próximos capítulos", "show_name": "Renascer", "celebrity_tags": "['Marcos Palmeira', 'Juan Paiva']", "publish_date": "2026-05-12T10:00:00Z", "comments_count": 342
| # | article_id | title | show_name | celebrity_tags | publish_date | image_url |
|---|---|---|---|---|---|---|
| 1 | ||||||
| 2 | ||||||
| 3 |
Complete list of extractable fields for Authors & Journalists objects from globo.com. All fields typed and schema-versioned.
"author_id": "auth-102", "name": "Andréia Sadi", "role": "Comentarista", "department": "Política", "article_count": 1420, "twitter_handle": "@AndreiaSadi", "profile_url": "https://g1.globo.com/autor/andreia-sadi/"
| # | author_id | name | role | bio | article_count | recent_articles |
|---|---|---|---|---|---|---|
| 1 | ||||||
| 2 | ||||||
| 3 |
Complete list of extractable fields for Video Metadata objects from globo.com. All fields typed and schema-versioned.
"video_id": "vid-882319", "title": "Gols de São Paulo 2 x 0 Corinthians", "duration_seconds": 184, "show_category": "Futebol", "is_premium": false, "publish_date": "2026-05-11T23:45:00Z"
| # | video_id | title | duration_seconds | views | publish_date | thumbnail_url |
|---|---|---|---|---|---|---|
| 1 | ||||||
| 2 | ||||||
| 3 |
Our Globo scraper navigates the complex structure of Brazil's largest media conglomerate, handling regional subdomains, live-updating sports pages, and infinite-scroll news feeds.
Extract headlines, body text, timestamps, author metadata, and embedded media links across G1, O Globo, and Valor Econômico.
Poll ge.globo match pages for real-time score updates, possession statistics, card events, and minute-by-minute commentary.
Capture reality TV updates, soap opera summaries, and celebrity news from gshow with full tag extraction.
Execute JavaScript to load paginated content on section fronts and author pages, ensuring no historical articles are missed.
Extract user comments, upvotes, and timestamps from article pages for sentiment analysis and public opinion tracking.
Target specific state or city subdomains (e.g., g1.globo.com/sp) to build localized datasets.
Scrape video titles, durations, categories, and view counts from embedded Globoplay and ge.globo players.
Configure sub-minute execution schedules for breaking news alerts and live sports match tracking.
Automatically detect and flag articles gated behind the Globo Mais subscription, extracting visible teaser content.
Brief in. Clean data out.
Provide G1 categories, ge.globo tournaments, or specific author profiles. We design the extraction schema together.
We configure Scrapy / Playwright crawlers, proxy rotation, and infinite-scroll handling for the Globo network.
Schema validation, null-rate checks, and payload inspection before full pipeline activation.
JSON / CSV / Parquet pushed to your S3 bucket, BigQuery dataset, or Snowflake stage on agreed cadence.
Media sites employ complex front-end frameworks and regional routing. Here is how we maintain stable extraction across the Globo ecosystem.
Globo heavily relies on infinite scroll for its news feeds and category pages. We run Playwright to simulate user scrolling, intercepting the underlying XHR requests to extract structured JSON payloads directly rather than parsing the DOM.
Certain video metadata and regional news content behave differently based on the visitor's IP. We route requests through Brazilian residential proxies to ensure we capture the exact content served to local audiences.
For ge.globo match tracking, standard crawling is too slow. We target the specific WebSockets and background API endpoints powering the live match dashboards, achieving sub-second latency on score updates.
Articles on O Globo and Valor Econômico frequently hit the Globo Mais paywall. Our pipeline detects the paywall boundary, captures the available teaser text, and flags the record as 'is_paywalled: true' to maintain dataset integrity.
A standard news article, a live blog, and an interactive feature on G1 all use different DOM structures. We maintain distinct selector chains for each article archetype, preventing null values when the editorial team uses custom layouts.
Agencies track brand mentions, executive quotes, and crisis developments across all regional G1 subdomains in real time.
Analysts aggregate ge.globo match statistics, possession metrics, and historical performance data to feed predictive models.
Data teams extract article comments and reaction metrics to gauge public sentiment on political and economic news.
Machine learning engineers use decades of high-quality Portuguese editorial text from Globo to train regional LLMs.
Publishers monitor G1 publishing velocity, topic coverage, and author output to benchmark their own editorial operations.
Investors track local economic developments and infrastructure news by scraping specific state-level G1 portals.
"Globo dictates the news and entertainment cycle for over 200 million people — but capturing that unstructured feed requires a dedicated extraction layer."
Extracting data from Globo's varied properties requires managing infinite scrolls, complex video metadata structures, and regional content variations. DataFlirt handles the heavy lifting of proxy rotation and JavaScript rendering so your analysts can focus on the underlying media signals.
Everything supported by our globo.com scraper — rendered SPA elements, auth walls, rate-limit evasion and beyond.
Open-source tooling on proven cloud infra — no vendor lock-in, full observability.
Scrapy handles crawl orchestration and deduplication. Playwright manages JavaScript execution for infinite scrolls and dynamic media embeds.
We maintain pools of Brazilian residential proxies to ensure accurate geo-targeting and prevent rate-limiting by Globo's CDN.
Pipelines run on AWS Lambda for burst extraction of breaking news, managed by Airflow for strict SLA adherence.
Data delivered to where your team already works — no new tooling required.
About globo.com scraping, legality, and pipeline operations.
Ask us directly →Scraping publicly accessible news articles, sports statistics, and metadata is generally permissible. DataFlirt extracts only public data, respecting paywall boundaries. We do not bypass DRM, extract premium video content, or violate authentication mechanisms. Clients must ensure their downstream use complies with copyright laws.
G1 serves localized content via subdomains. We configure pipelines to target specific state or city endpoints, using Brazilian residential proxies to ensure the correct regional variations are served by the CDN.
Yes. We bypass the HTML DOM and directly poll the backend APIs powering the ge.globo live match dashboards, allowing us to capture events, possession stats, and scores with sub-minute latency.
Yes. We intercept the requests to Globo's third-party commenting infrastructure, extracting user handles, comment text, timestamps, and upvote metrics for sentiment analysis.
Our pipeline identifies the paywall boundary, extracts the visible headline and teaser text, and flags the record with an 'is_paywalled' boolean. We do not use compromised credentials to bypass the paywall.
For critical news monitoring, we configure high-frequency pipelines that poll G1 front pages every 60 seconds, delivering new article URLs and headlines via Webhook instantly.
No. We extract video metadata — titles, durations, tags, and view counts — but we do not download or extract the actual video files or bypass Globoplay DRM protections.
20-minute scoping call. Pilot dataset within the week. Production within two. Whether you need a one-off historical news dump or a continuous live sports feed — we scope, build, and operate the pipeline. Tell us what you need.
