Authors have spent months building a KDP catalog, growing a readership, watching sales come in — only to log in one morning and find everything gone. No prior warning. No specific violation named. Just an email that says their account has been terminated for “activities that are attempting to manipulate our services,” with abnormal reading activity listed as one example.
Here’s what makes this particularly unsettling: the message doesn’t just hit KDP authors enrolled in Kindle Unlimited. Authors with paperback-only catalogs have received it. Authors who have never enrolled a single book in KU have received it. The phrase “abnormal reading activity” sounds specific, but in practice, it has become something broader — a generic response Amazon’s system sends when it finds anything suspicious about an account, with or without reading data to back it up.
What Amazon’s Termination Email Actually Says
When KDP terminates an account this way, the email typically reads:
“During a review of your account, we found activities that are attempting to manipulate our services. This is a violation of our Terms and Conditions, and as a result we are terminating our relationship with you and closing your account, effective immediately. Examples of manipulation include abnormal reading activity and attempting to publish books that violate our Content Guidelines.”
Then comes the list of consequences: account closed, outstanding royalties forfeited, all published titles removed from sale, and a permanent ban on opening new KDP accounts.
Note the language carefully. “Examples of manipulation include abnormal reading activity.” It’s listed as an example, not as the confirmed finding. There is no data attached. No specific book named. No date range for the activity flagged. Amazon’s system has concluded your account is suspicious, and this is the message it sends.
The PreviewBound team has reviewed enough of these cases to reach a working conclusion: “abnormal reading activity” in KDP termination notices often functions as a catch-all — a generic response triggered when Amazon’s algorithm flags an account as suspicious, regardless of whether actual reading manipulation occurred. The account looked wrong to the system, and this is the language that comes out.
What Originally Made It a KU Problem
The phrase has a real origin. Kindle Unlimited pays authors per page read, and that payment model attracted scammers almost immediately. Click farms — networks of bots or low-paid workers — would rapidly flip through enrolled books to rack up page reads without any reading happening. Amazon’s algorithm was built to catch those patterns: impossible reading speeds, borrows from freshly created accounts with no history, geographic clustering of suspicious activity.
That enforcement is legitimate. Fake page reads steal from a shared royalty pool that real authors depend on.
But the system that was built to catch KU fraud has expanded into something that triggers on account-level signals that have nothing to do with reading.
Why Paperback-Only Authors Get Hit Too
If you’ve never enrolled in Kindle Unlimited and you receive a termination citing abnormal reading activity, the reading data isn’t the issue. Something else about your account flagged the system.
The PreviewBound team’s assessment, based on author reports: Amazon’s review process covers the whole account, not just the book sales. When the automated system initiates a review and finds something that looks suspicious — whether that’s reading data, account details, identity verification, or behavioral patterns — the termination message that gets sent still references manipulation and may still list abnormal reading activity as an example, even when the actual trigger was something entirely different.
This is the part no one talks about clearly. The termination message is templated. What caused the review is not always what the template says.
Account Suspicion Triggers Beyond Reading Activity
Several things can cause Amazon to treat an account as suspicious during a review, even without KU enrollment:
Incorrect or inconsistent account details. Name mismatches, addresses that don’t correspond to the registered payment information, phone numbers that don’t verify, tax information that doesn’t align with identity documents — any of these can flag an account as potentially fraudulent during a review cycle.
Identity and face verification failures. Amazon periodically asks authors to complete identity verification, including in some cases submitting a photo or completing a video verification step. If the verification fails — the photo is unclear, the face doesn’t match the account’s history, the ID documents don’t satisfy the check — Amazon may treat the account as operating under false identity. The termination that follows can still cite manipulation and abnormal activity in its language, even though the actual trigger was a failed identity check.
Bot-like account behavior. Rapid publishing at volumes that look automated, metadata patterns that resemble mass production, or submission patterns that match known spam behavior can all raise an account’s internal risk score without a single page being read abnormally.
Device or IP signals. Multiple accounts accessed from the same device or IP address — even if you’re helping a family member set up their own account, or logging in from a shared network — can register as an attempt to operate multiple accounts.
Payment and banking anomalies. A bank account or tax identity that doesn’t match the registered author identity, or royalty payment details that change in a pattern that looks suspicious, can trigger a review.
The Generic Response Problem
This is the honest assessment: Amazon’s termination message for account-level suspicion appears to default to manipulation language and lists abnormal reading activity as an example regardless of what the actual flag was. Authors who have only ever sold paperbacks, who have never touched KU, have received the exact same termination email as authors who genuinely inflated page reads with click farms.
That means two things practically. First, if you receive this message and you genuinely haven’t manipulated anything, the problem may not be reading activity at all. It may be something in your account details, your identity verification, or your account’s behavioral signals. Second, the appeal process needs to address the actual likely trigger, not just defend your reading activity — because defending reading activity when your account was flagged for a failed identity check won’t help.
Termination vs. Suspension
A suspension is temporary. Your ability to publish is frozen while Amazon reviews the account, but it may be recoverable once you address whatever triggered it.
Termination is permanent. Books come down, royalties are forfeited, and Amazon’s terms prohibit you from opening a new account. They track accounts through banking details, IP addresses, and device identifiers. A second account opened after termination typically results in both being banned.
The manipulation and abnormal activity flag tends to go straight to termination. There usually isn’t a warning first.
Appealing
You can appeal by replying directly to the termination notice — not by opening a new support ticket.
If your account was flagged for reading activity, your appeal needs to address your promotion history, explain any unusual spikes with legitimate reasons, and demonstrate that your readership is real.
If your account was flagged during an identity review, your appeal needs to focus on verifying your identity clearly — providing accurate, matching documentation and explaining any discrepancies in your account details.
The problem with appealing a manipulation termination is that Amazon doesn’t tell you what the actual trigger was. You’re writing a defense without knowing the charge. Authors who succeed tend to be thorough — they address every possible angle: identity, promotion history, account details, any family or colleague who may have accessed the same account. Leaving a gap gives Amazon’s review team nothing to work with.
Reinstatement is possible but uncommon. Act quickly, stay polite, and be specific.
What Keeps an Account Safe
Keep your account details accurate and consistent. Name, address, phone number, tax information, and payment details should all match and be current. If anything changes, update it promptly. Mismatches are one of the cleaner ways to trigger a review you didn’t need.
Complete identity verification promptly. If Amazon sends a verification request, do it immediately and carefully. A blurry photo, an expired document, or a mismatch between your face and your submitted ID can be enough to trigger the account review that leads to termination.
Be careful with promotional services if you’re in KU. Services that promise guaranteed page reads or borrows without explaining their method are the ones that cause problems. Real promotion drives real readers. Fake promotion leaves a pattern in your data that Amazon’s system will eventually catch.
One account per person. Pen names don’t need separate KDP accounts. One account handles everything. The only exception is a separate legal business entity with its own tax ID.
Monitor your dashboard reports. If your borrows or page reads spike in a way you can’t explain, take note of it. Contacting KDP proactively to flag unusual activity you didn’t cause is better than letting the algorithm draw its own conclusions unchallenged.
The Honest Reality
“Abnormal reading activity” as a termination reason has expanded beyond what it literally says. It now appears in termination notices triggered by account suspicion at a level that has nothing to do with how readers are consuming your books.
The phrase has become Amazon’s generic flag for: we found something about this account that doesn’t look right, and we’re closing it.
That’s worth understanding, because it changes how you protect yourself. Keeping your publishing activity clean matters. But so does keeping your account details accurate, your identity verification current, and your account behavior recognizably human. Amazon is watching the whole account, not just the books.
Was this helpful?