Read-it-later statistics: the honest numbers (2026)

How many saved articles actually get read?

Track 250 bookmarks long enough and you learn the truth: only 16% ever get revisited. The other 84% are saved once and never touched again. That's not a vibe — it's the largest controlled study of bookmark behavior ever run (Bergman, Whittaker & Schooler, 2021).

The detail that should worry every "save it for later" user: of the few bookmarks that did get revisited, only 4% were reached through the bookmarks menu — the rest depended on an always-visible bookmarks bar. Out of sight really is out of mind.

Pocket, in numbers: 2B saves couldn't save it

• ~17 million registered users in April 2015, at its Series funding announcement (VentureBeat, 2015).
• ~20 million users and 2 billion saved articles circa 2016, built by a team of just 20 people (First Round Review).
• May 22, 2025: Mozilla announces the shutdown; July 8, 2025: the service stops (TechCrunch, 2025).
• November 12, 2025: data export disabled; all user data queued for permanent deletion (Mozilla support).

Note what's missing: a user count at shutdown. Mozilla never published one, and the numbers circulating online are unsourced. Two billion saves and it still wasn't a business — the save button was doing all the work, and the reading never happened. That gap between saving and reading is the category's structural disease.

The graveyard keeps growing

• Omnivore — the open-source "free Readwise" favorite — had 500,000 users worldwide when ElevenLabs acquired the team on October 29, 2024. The hosted service closed on November 15, 2024 and user data was deleted.
• With Pocket gone eight months later, two major read-later services died within twelve months of each other.

The full casualty list — 13 dead apps and services, from Summify (2012) to Pocket (2025), with dates and causes of death — lives on our read-it-later graveyard timeline. To estimate how your own pile compares, the graveyard calculator does the math.

The survivors, measured

• Instapaper — launched January 28, 2008 by Marco Arment, survived three ownership changes (Betaworks 2013, Pinterest 2016, Instant Paper 2018), and in July 2025 was chosen by Kobo to replace Pocket on its e-readers. Eighteen years old and still standing.
• Readwise Reader — entered public beta December 14, 2022 at $7.99/month (annual) beta pricing.
• Matter — still operating in 2026 (getmatter.com), one of the survivors actively courting ex-Pocket users.

Read that chart carefully: the biggest "save for later" tool on this list isn't a read-later app at all. OneTab — a button that shelves your open tabs — has roughly 2 million users, five times Raindrop's extension install base. People aren't saving articles because they have a reading plan; they're saving to make the tab guilt go away.

The bigger picture: overload without the reading

• A Carnegie Mellon study presented at ACM CHI 2021 (ScienceAlert coverage, n=103) found ~30% of participants self-described as tab hoarders, and more than half couldn't bring themselves to close tabs they had open.
• Pew Research (2016, n=1,520): only 20% of Americans say they feel information overload — down from 27% a decade earlier — while 77% like having so much information at their fingertips. We don't feel overloaded; we feel behind. That's what the save button monetizes.

Where these numbers come from

All statistics were re-verified in July 2026 against the linked primary source. Chrome Web Store install counts are point-in-time snapshots (July 8, 2026) and drift over time; they measure extension installs, not a product's total user base. Where a company never published a number (e.g. Pocket's user count at shutdown), we say so instead of repeating unsourced figures. Found an error? Open an issue — corrections ship within a week.

Frequently asked questions