Consumer Privacy

Your E-Reader Knows Where You Stopped Reading

July 13, 2026 6 min read Haven Team

A paper book has no idea whether you finished it, skipped the middle chapters, or fell asleep on page 40 for three weeks running. A modern e-reader does, and for connected devices, that record does not stay on the device.


Amazon's Kindle, and most competing e-readers built on similar telemetry models, can log more than the fact that you own a book. Depending on the device and account settings, that can include the furthest page you have reached, how long you spend on individual pages, when you stop reading and for how long, and which specific passages you highlight or bookmark. None of this is hidden in the fine print exactly, it is described in Amazon's own privacy documentation, but it is also not something most readers think about while they are absorbed in a book.

This was reported in detail by The Wall Street Journal in a widely cited 2015 investigation, "Your E-Book Is Reading You," which walked through what several major platforms, Amazon, Barnes & Noble, and Kobo among them, were collecting about reading behavior at the time, and how that data fed back into product decisions ranging from which books got recommended to which manuscripts publishers chose to acquire based on where readers tended to abandon similar titles.

The Feature That Makes the Collection Visible

Kindle's "Popular Highlights" feature is the most visible evidence that this data is aggregated across readers, not just stored per-device. When a passage in a book has been highlighted by enough other Kindle readers, the book displays a small underline showing you where other people marked the same sentence, along with a count. That feature only works because Amazon is collecting and aggregating individual highlight data across its user base and matching it back to specific passages in specific editions.

It is opt-out, not opt-in, on many devices and can be turned off, but the default has historically been on, and most readers who have seen those underlined passages while reading never connected them to their own highlights being fed into the same aggregate pool for other readers to see.

What this reveals that a purchase record does not

A bookstore receipt tells you what someone bought. Reading telemetry tells you what they actually engaged with: which chapter they stopped at, which passage moved them enough to highlight it, whether they finished at all. That is a materially more granular signal about a person's interests, beliefs, and state of mind than the purchase alone.

Why Publishers and Platforms Want This Data

The commercial logic is straightforward and mostly not adversarial: aggregated abandonment data (where readers commonly stop finishing a book) has genuinely informed editorial decisions, like where a manuscript's pacing drags, and recommendation engines use reading pace and completion data to suggest what a specific reader is likely to finish next. Platforms have talked about this openly in industry contexts as a legitimate use of behavioral data to improve their product.

The privacy question is not whether this use is malicious. It is whether readers understand the trade, and whether the same data, once collected and retained, is available to be repurposed, subpoenaed, or exposed in a breach for reasons that have nothing to do with improving a recommendation algorithm. Reading history has a long-recognized sensitivity in American law, discussed in more detail in our piece on library patron privacy, precisely because what a person reads is one of the more direct signals of what they are thinking about, researching, or working through.

Data point Typically synced to account, not just stored locally
Furthest page reached per book Yes, used for "Whispersync" style cross-device resume
Highlights and notes Yes, synced across devices and, if enabled, aggregated anonymously into Popular Highlights
Reading session duration and pace Varies by platform and account settings
Library book loans through a licensed reading app Varies; governed by the app vendor's policy, not library confidentiality law

What Actually Reduces the Footprint

A few concrete settings changes meaningfully cut what is collected, without giving up cross-device sync entirely if you want to keep it:

Where Haven Fits

Haven does not make an e-reader, and this is not a problem our product solves directly. What connects it to the rest of our work is the same principle we apply everywhere else: a service should collect the minimum it needs to function, and anything genuinely private, in this case what you read and how you engaged with it, should not become a byproduct of convenience features you never explicitly agreed to.

Try Haven free for 15 days

Encrypted email and chat in one app. No credit card required.

Get Started →