Google Doodles Scraper

Scrape public Google Doodles archive pages and individual Doodle detail pages at doodles.google. Each dataset record describes one Doodle.

It extracts the title and stable page slug, the displayed publication and appearance dates, introductory copy, the full public About text, the Doodle's own key-theme links, its artwork/media URLs, creator-credit lines when present, public outbound links embedded in the story, and the share and related-Doodle links exposed on that page. Each outbound link includes its source label, exact destination URL, hostname, and nearest preceding section heading. Doodles-site navigation and social-share controls stay in their separate fields. Detail-page extraction is anchored to the article or main containing the Doodle title. Header, navigation, and footer content is excluded from media, share, and related-card extraction. Artwork linked to another Doodle is recorded as a related card rather than mixed into the current Doodle's media, and text-only links to other Doodles are not labeled as related items.

Input

Direct detail-page example:

{
  "startUrls": [
    { "url": "https://doodles.google/doodle/celebrating-hispanic-heritage-month-2026/" }
  ],
  "maxItems": 1
}

Search query example:

{
  "queries": ["Halloween"],
  "maxItems": 2,
  "maxPages": 1,
  "includeDetails": true
}

startUrls accepts HTTPS archive, search, or /doodle/<slug>/ pages on doodles.google (and its www host). query or queries builds the public archive URL using its title_like filter. maxItems is the global number of Doodle rows. maxPages limits supplied archive/search source pages. includeDetails defaults to false: explicitly enable it to add one bounded detail request per archive card for rich page fields. Direct detail start URLs always extract the requested detail page. Google uses “Show More” instead of a stable numbered archive-page URL, so page is limited to 1.

For archive searches, the Actor follows the public “Show More” control within the selected limits. maxPages limits the number of archive/search pages; Google currently supports only page 1 for the page-number setting. includeDetails is off by default and adds one public detail-page request per discovered Doodle. You can optionally configure an Apify proxy for public requests. The Actor does not log in, solve CAPTCHAs, or bypass access controls. Google may limit public search results, so direct archive or detail URLs can be useful when a search does not return results.

Dataset fields

Enriched detail records expose Doodle-specific source facts; card-only rows are smaller. Aliases, date components, duplicate media representations and derived counts are not independent facts. Core properties include:

• Identity and navigation: doodleId, title, slug, pageUrl, canonicalPageUrl, archiveUrl (null for a directly requested detail page)
• Dates: publishedDate, publishedDateText, date components, appearedDate, appearedDateText, appearedDateUrl
• Story: description, clean descriptionHtml, aboutHeading, full aboutText and clean aboutHtml, launchText, whereAppearedText. HTML retains paragraph/list/emphasis markup, with active content, controls and all attributes removed; source prose is not truncated.
• Artwork and geography: artworkColor, artworkStyle, artworkForm, exact nationalityText and structured appearedCountries from the public title-matched metadata and appearance map. Country coordinates are source map reference points, not artist locations or exact event locations. Nationality wording is not treated as artist citizenship. Runtime/analytics dates are not used as publication dates.
• Themes and media: keyThemes, themeLabels, media, mediaImages, imageUrls, videoUrl, animationUrl, interactiveUrl, nullable media-type flags
• Credits and links: artistName, structured creatorCredits, externalLinks / externalLinkCount, shareOptions, relatedDoodles
• Collection metadata: themeCount, imageCount, shareOptionCount, externalLinkCount, relatedDoodleCount, contentLanguage

Unknown source values remain null; a missing boolean is not turned into false. When includeDetails is true, search/archive rows are expanded into public detail pages. If a detail is unavailable, valid card identity, date, and artwork remain in the result. Archive-only rows contain fewer fields than detail-enriched rows.

Output and access boundaries

Each dataset record represents one Doodle. Empty searches, blocked pages, and failed requests are reported in the run summary and logs rather than added as Doodle records.

You can download the dataset in formats such as JSON, HTML, CSV, or Excel. The run summary is available alongside the dataset in the Output tab. Detail enrichment adds extra requests and may increase compute use, so start with modest maxItems and maxPages values.