Actual fraud in Kindle Unlimited

The following information is not a guide to action, is informative in nature and does not in any way encourage violating the laws of the Russian Federation and the United States, as well as the rules of Payoneer and Amazon services.


In July 2016, I read an article on GT about how fraudsters make money on Amazon Kindle Unlimited. In August, I began to try to implement according to the described algorithm, but it turned out that everything was not at all as rosy as described, and besides, the actions outlined in some paragraphs were absolutely not transparent (for example, the paragraph about several versions of the book).

For a month of testing, having obtained the blocking of one account, on which I immediately tried to “generate” ~ $ 10,000, I managed to sort out more or less. Subsequently, testing continued, allowing me to earn ~ $ 8.000 per month, I spent about 1 hour per day on testing (and not every day).

At the moment I know two schemes on which you can earn (theoretically) stable ~ $ 15.000 per month. One of these schemes is obsolete, because some of Amazon’s abstract machine learning has learned and the accounts used are always blocked.

Payments on the example of four accounts

At the moment I do not deal with such manipulations, preferring to work honestly with KU, for this reason I decided to share information with the community.


I will introduce a few concepts to make it easier.

The reader is an Amazon account created to read books that have a trial activation period in Kindle Unlimited. E-mail address does not have to belong to you.

The writer is an Amazon account registered with Kindle Direct Publishing (KDP). E-mail address must belong to you .

KENP is a reading unit in KDP, which is taken into account if the book is released in Kindle Unlimited, each page is paid a reward of $ 0.0045-0.005 (depending on the total money pool, it is based on the results of Kindle Unlimited ads and subscribers for the past month).

Scheme # 1. Volumetric manual reading in depth

Registration 50-100-200 readers

Regular account registration in Amazon.

This stage is preferably divided into several days. I took about 10-30 e-mails successively and registered through Tor with “pens”.

Writers registration

It all depends on greed, in a short time I had an aggregate of about 50 writers (all of them were banned), at the same time there were no more than 12 active.

Regular registration of an account in Amazon, only it will be necessary to register at KDP, accepting the agreement, then fill in the information about the writer.

I chose random Russian names with random addresses (mainly Rostov-on-Don, Moscow, Nalchik, Stavropol).

Next, you must fill in the information for tax, in the TIN (TIN) input field, you can enter any TIN that you want, while everywhere indicating that the writer is an “Individual”.

Stage of "writing" books.

Consists of two nested stages:

Now you need to do a lot of books

The main point is to get a lot of similar books from the first one received.

I used MS Word macros, just swiping random pieces of the “tail” several times, then macro-stitched headlines in random pages for the table of contents (for example, “Chapter 8”) and automatically generated the contents.

It is advisable not to use one received book from the “tail” and “base” more than 7-10 times. At one time, I generated from one book and used more than a hundred pieces.

For each writer, it is desirable to load 6-7 books.

Download books to writers

Pre-generate titles and descriptions that will slightly correlate with the content of the “base” and at least somehow with the title of the book. Then just go to your account and download books. Images for the cover can not be chosen at all, either from the standard KDP library.

The download process is long and errors can occur, the main thing is to simply press the button further and not download anything again.

I downloaded from several browsers under different VPNs or several TOR instances at once.

Approximate view of the book shelf writer

Preparing to read

Log in to the reader account, fill in all the data previously. Data: card number, expiration date, address, zip code, telephone. Address and index data can be obtained from here . These phone numbers by city can be taken here . As cards use virtual cards. It is very desirable that the virtual cards always have an amount of $ 1.

Next, you need to go to Kindle Unlimited and register for the use of the trial period.

In advance you need to prepare a list of 20-50 random books released in Kindle Unlimited, which are similar in subject to yours.

I automatically collected address data and phone numbers in MS Excel macros. Cards generated "pens" using Yandex.Money and QIWI.


The process of reading itself is simple. Either know in advance the URLs of your books, or in advance in the search for the Kindle Store, drive in the name of your author and save the resulting URL. Next, in a crowd to open your books (6-7 pieces), in addition to open 3-4 random books of any other authors. After that, simply using Kindle Unlimited to rent these books for free and immediately open in the browser for reading ( Opera browser is not suitable for reading in it ).

It is necessary to read the first, second, third pages of each book, making a break between turning over 40-90 seconds. Then go almost to the end of the book itself. Similarly, read 2-3 pages, turning to the beginning of the book.

Thus, one writer can get ~ 21,000 KENP for one such reader of 7 books. If you read two readers per day for a month, you will get a total of about ~ 1,260,000 KENP, and this, with the “course” of 1 KENP of $ 0.0047, is already ~ $ 5.922. One reading takes from 5 to 10 minutes.

Reading is desirable to distribute throughout the day, but you can read the “daily rate” at a time.

Optionally, you can delete the map data and delivery.

Subtotal and inoperability

In general, I brought all the points of the scheme. How many readers to use per day, how many writers to have, and how many books - it is better to calculate in advance and plan on the basis of the folding figures.

This is the scheme I used, for a couple of months I was able to earn and withdraw ~ $ 15,000 in this way. Over the last two months, I can say that this scheme with a probability of 99. (9)% ceased to function because Amazon had learned (perhaps my volumes of “content” contributed to this) and banned writers' accounts before they managed to withdraw money to score.

Question : Where is the reading "in depth"?
Answer : “Depth” is in every writer. You do not need a lot of writers and a lot of books, but you just need incredible volumes, within the framework of ordinary human logic, each book.

Advice : it is better to enter data on the bank account before the transfer from KDP somewhere in 1-2 days. The transfer is made within ~ 60 days after the end of the month (by experience - the 20-22 day of the month). For example, in May 2017, you will receive money at the end of July 2017 using the EFT transfer, while the funds themselves will be sent around July 20-22, 2017.

Withdraw money. Short and about

Now a little about withdrawing money to the accounts. I repeat that these are just approximate actions that could theoretically be performed. I do not call for their perpetration and did not do this in any way.

  1. Register with Payoneer with the name of your writer.
  2. Introduction of valid driver's license / passport data. I don't know where to get this data.
  3. We are filling in any address available to you, where you can send a card, where you can either get it yourself, or find out the number and CVC.

Cards are sent by mail to Russia and reach the recipient either quickly (Moscow and Moscow region), or slowly (Rostov-on-Don). It is better not to send more than 3-4 cards from different accounts to one email address. Different mailing addresses can be obtained very simply - in any branch of Russian Post there is a mailbox rental, the first three months are physical. face will cost ~ 400. Some branches of the card may not reach at all - you need to be prepared for this and in the “bad” post offices you can not rent mailboxes. The mailbox can only rent one per branch, but it is not forbidden to rent a lot of mailboxes in many branches. In general, bijective mapping to make it clearer.

Get the card and activate it - as the last step.

Repeat for each writer, probably.

Note : it would be desirable, probably, to have one account to which a real account is tied to one of the banks in order to accumulate funds from others on this account and then immediately transfer to your R / C in the bank. "But it is not exactly" .

Scheme # 2. Semiautomatic wide reading using db

Here I can’t give details of the technologies used, everyone is free to choose for himself - be it Excel or Postgres.

This scheme almost completely repeats the “white” earnings in KDP, but with some nuances, less investment, no ads, a large number of writers and a developed system of published books. Therefore, this scheme is more complicated, but with due effort, both from the technical side and from the “offline” side, it is possible to achieve a monthly salary of $ 15,000–20,000.

Partial repetition from the first scheme

From the first scheme, the following items remained:

CRM similarity

To manage accounts, as well as to understand where / when / how / what was used or done, you will have to write a CRM similarity that will pull data from the database. Excel here is no longer suitable due to the fact that automation is useful here.

Conditional CRM

In essence, we simplify our work with a large number of readers and writers. To do this, you can create a database for readers, writers and books. The functions on the update of the database are obvious, it is necessary to make automated decisions for them to implement the functions themselves, for example, the registration of the writer.

Reader table example

When a writer is created in CRM, he is entered into his table in the database, you also need to go to the browser and repeat the steps to register the KDP writer. This process is automated via Selenium WebDriver (works in both Chrome and Firefox with TorBrowser). The script for Selenium should be generated based on the entered data, after which it can be copied and run in the browser.

Creating a writer in CRM

Similarly, readers are created. It will be a little more difficult to write the Selenium script template for the upload of your works in KDP.

There are a lot of data for the book.

In addition, you can add a small table to the database, where URLs of books from other “white” writers are entered, so that you can quickly “take” these pages for your needs. Periodically check for error 404.


The reading process has hardly changed, but there is also automation going on through Selenium. After registration in the reader's KU, the Selenium opens the books, then the initial reading takes place (towards the end of the book), the pages are scrolled through at a random time in the region of 90-150 seconds.
After that there is a choice (which one is better - I cannot say for sure):

  1. Finish the script and go for a walk, after the reader session “falls off”, restart it manually, the KDP “reader” will reboot, then just go to the end of the book and again turn the pages towards the beginning of the book.
  2. The script waits for another random 90-150 seconds, then goes to the end of the book and turns the pages towards the beginning of the book every 90-150 seconds, after 3-4 pages the script ends.

The main nuance in the read script is the correct handling of tab transitions within a single browser.

In this case, reading can be done in one of the options:

  1. Multiple TorBrowser instances on the same machine. This option is not very good if you have weak characteristics, and also that the OS does not change.
  2. Many virtual machines. At the same time, it would be great to have different operating systems and their versions in each VM. This greatly complicates the automation of control. Difficult option, but the most correct.

"Writing" books

Returning to the content part, you will need to get a lot of unique small books with a size of 150-250 A4 pages in Arial 12 font. Content for books 18+ is easily in the public domain, but this content will take a lot. Also, this content will need to be well formatted in order to look more or less “solid”. In this case, I do not see how to automate the process.

Another nuance is that the automatic translation will have to be corrected at least a little, and it will also have to be done manually.

The final figures are for 100 writers: about 500-700 such books will be needed, which is a lot.

Problems and difficulties of circuit # 2

In this case, there are many difficulties:

  1. Automate the deployment of virtual machines and their management.
  2. Initial technical training (script templates, creating a database and CRM, writing management) can take a lot of time.
  3. The initial content preparation (if you make a full preparation for all 100 writers) can take more than one month.
  4. Small difficulties may arise with the withdrawal of money, because need a lot of data and postal addresses.
  5. Get a large number of e-mail addresses in the systems GMail, Yahoo may cause questions, but there are ways to solve it. There are no problems with addresses in Outlook, for example.

One of the solutions to the problem with e-mail addresses

The main difficulty: there is no possibility to automate the stage with virtual cards, that is, almost every reader will have to walk and manually receive virtual cards, which is very time consuming (about 30 minutes a day). There are also restrictions on services for receiving cards, which leads to the need to have additional accounts. Maybe there are some services that will simplify this stage.


As the end of the descriptive part of this scheme, we can say that we have to work a lot here, both technically and content. Full automation cannot be achieved. An extensive "network" of writers and books can be scary.

The question is : where is the reading in “width”?
Answer : “width” due to the volume of the number of books, so you have to use a lot of readers every day.

Question : how much can each writer receive?
Answer : empirically it was found that even a threshold of $ 800 per writer is a lot, so in Scheme # 2 it is advisable not to exceed the threshold of $ 150-200 per writer. Thus, you can have those same $ 15,000-20,000 per month.

Question : Why is this scheme similar to the "white" system of earnings in KDP?
Answer : the books that you need to get in this scheme, in fact, are content on 3 points out of 5, that is, readable and which your readers will sometimes have to read. Also the volume of books is normal (1 book is really read for an ordinary person in 1 day). From the non-“white” there is just the use of automation and “their” readers, instead of using advertising and promoting their books. Well, even the "white" way for an ordinary writer is difficult to earn more than $ 5,000.


The two schemes here are just a description and an example. You can not break anything. It’s rather a way to use the Kindle Unlimited promotion a little more than necessary. It is important to understand that as soon as Amazon closes the KU trial period, you can say goodbye to these methods.

But you can get an invaluable experience of the tester when writing scripts for Selenium.

I will ask in the comments not to ask questions, the answers to which may lead to violations of the rules of the Payoneer system or violations of the legislation of the Russian Federation.


All Articles