Recovering Old Google Reader Data: Step-by-Step GuideGoogle Reader was discontinued in 2013, but many people still have valuable archives — starred items, subscriptions, notes, and highlights — tucked away in their Google accounts or exported files. This guide shows step-by-step how to locate, recover, and reuse old Google Reader data safely in 2025, whether you have an old Takeout archive, a stray RSS export, or just memories of feeds you once followed.
What you can expect to recover
- Subscriptions (OPML) — lists of feeds you followed.
- Starred items and highlights — saved articles you marked inside Reader.
- Shared/liked items and comments — posts you shared publicly from Reader.
- Cached article bodies — sometimes included in exports or third-party archives.
- Metadata (dates, tags, read/unread status) — may or may not be present depending on source.
Typical sources of old Google Reader data
- Google Takeout archives you downloaded around 2013.
- OPML exports you saved from Reader before shutdown.
- Backups or exports from third-party apps (NewsBlur, The Old Reader, Feedly imports).
- Browser bookmarks or saved pages containing feed URLs or content.
- Web archives (Wayback Machine) capturing Reader pages or shared items.
Step 1 — Locate any existing exports or backups
- Search your cloud storage and local drives for keywords: “Google Reader”, “Takeout”, “subscriptions.opml”, “reader”, “starred”, “shared”.
- Check old email attachments and downloads folders for files with extensions: .zip, .opml, .xml, .json.
- If you used Google Takeout in 2013, look for a ZIP containing a “Reader” folder with files like subscriptions.xml, starred.json, or shared.json.
If you find nothing locally, move to web sources below.
Step 2 — Check Google Takeout and your Google Account
- Google no longer provides active Reader data, but if you ever used Takeout and stored the archive in Drive, Dropbox, or your computer, it will contain the most complete exports.
- If you still have access to the Google account you used in 2013, check Google Drive, Gmail (search attachments), and Google Takeout history (takeout.google.com/settings/takeout) — past exports may be listed there.
Step 3 — Inspect and identify file formats
Common files you might find:
- subscriptions.xml or subscriptions.opml — list of feeds (OPML/XML).
- starred.json / like.json / shared.json — saved items and interactions (JSON).
- items.json / reader-items.json — cached article content (JSON).
- HTML or MHTML files — archived Reader pages or exported pages.
Open files with a text editor (VS Code, Notepad++, Sublime) to confirm contents. If files are compressed (.zip), extract before inspecting.
Step 4 — Import subscriptions (OPML) into a modern reader
If you have subscriptions.opml or subscriptions.xml:
- Choose an RSS reader that supports OPML import — examples: Feedly, The Old Reader, NewsBlur, Inoreader, FreshRSS (self-hosted).
- In that reader, find Import > Upload OPML and select your subscriptions.opml.
- Review imported feeds — some feed URLs may be defunct; unsubscribe or update as needed.
Tip: If OPML contains many dead feeds, use a feed checker tool (several online tools can validate URLs in bulk).
Step 5 — Recover starred/shared items and article content
If you have starred.json, shared.json, or items JSON files:
- Inspect the JSON structure to find fields like title, canonicalUrl/link, content/body, publishedTimestamp, and annotations/tags.
- If content/body exists, you can extract saved article bodies and save them as HTML/Markdown for personal archiving. Use a small script or an online JSON-to-HTML converter.
Example (conceptual) script steps:
- Parse JSON.
- For each item, extract title, url, content, and timestamp.
- Save as individual files named with date + slug, or append to a single archive file.
If you’d like, tell me whether you prefer Python, Node.js, or a no-code method and I’ll give a ready-to-run script.
Step 6 — Recover metadata (tags, read/unread status)
- JSON exports often include tags or labels that correspond to Google Reader’s folders/tags. Map those tags to your new reader’s tagging or folder system during import or by using a script to add them via the reader’s API (if available).
- Read/unread state is rarely preserved cleanly in old exports; if you need it, check for a boolean or status field in items.json/starred.json and convert accordingly during import scripting.
Step 7 — Use the Wayback Machine and other archives
If you lack exports but remember your shared items page or profile URL (for example, a public shared-items page), try:
- Searching the Wayback Machine (web.archive.org) for your Reader profile or shared-items URL.
- Looking up specific article URLs or feed URLs there to recover content or feed endpoints.
- Searching cached pages for lists of subscriptions or shared items.
Step 8 — Importing recovered articles into a modern system
Options to preserve recovered content:
- Import into a note system (Obsidian, Notion, Evernote) as Markdown or HTML files.
- Host a personal archive with a static site generator (Hugo, Jekyll) to browse older saved items.
- Use self-hosted readers (FreshRSS, Tiny Tiny RSS) to re-ingest content and keep it searchable.
If saving to Markdown, consider converting HTML bodies using pandoc or html-to-markdown libraries.
Step 9 — Automate cleanup and deduplication
Recovered archives often contain duplicates. Steps to clean:
- Normalize URLs (strip tracking parameters like ?utm_*, https/www variations).
- Deduplicate by canonical URL or content hash (MD5/SHA1 of article body).
- Remove dead links (HTTP 404 or domain gone) or mark them for manual review.
A simple Python pipeline can perform normalization, hashing, and dedupe — tell me if you want a script.
Step 10 — Privacy, licensing, and sharing considerations
- Your export likely contains personally saved items and possibly comments; treat it as private unless you intend to share.
- If you publish recovered article content, check copyright: saving personal copies for private use is generally fine; republishing full articles might infringe copyright unless permitted.
Practical examples and mini-recipes
- Quick: Import OPML to Feedly — Log in to Feedly > Organize > Import OPML > Upload file.
- Script idea (Python): parse starred.json, save each item as YYYY-MM-DD-title.html, include metadata frontmatter for tags and original URL. (I can provide full code on request.)
- Web-archive rescue: enter your old Reader shared-items URL in Wayback Machine, then bulk-download archived pages with a crawler to extract links.
Troubleshooting common issues
- No files found: search old email addresses and cloud accounts; ask family members who might have old backups.
- Corrupt ZIP: try 7-Zip or The Unarchiver; if still corrupt, partial extraction tools may recover files.
- Invalid OPML/XML: fix encoding issues by removing invalid characters or declaring correct UTF-8 in the file header.
When to ask for help
Send me:
- A sample of the JSON/OPML (paste a small snippet) and I’ll tell you what’s inside.
- Which platform you want to import into (Feedly, NewsBlur, self-hosted) and I’ll provide exact steps.
- Whether you want scripts (Python/Node.js) and I’ll provide them.
Recovering old Google Reader data is usually possible if you have any exported files or can locate archives; with a few scripts and imports you can rebuild your feed library and preserve starred articles for the future.
Leave a Reply