So, to break it down:
You can send free ARCs to a bunch of readers (your street team) and let them know when the book launches and ask for an honest review if they’ve read the book;
You cannot ask them to post a 5-star review (that’s influencing the review);
You cannot require them to post a review in exchange for an ARC (so you can’t email them saying: “I gave you a free ARC, now you need to give me a review!”)
You cannot incentivize readers to post a review with anything other than a free ARC.
And finally, your friends and relatives can’t post reviews of your books.
But how does Amazon identify biased reviews (i.e. ones left by friends and relatives)?
That’s the million-dollar question. And while no one has a definitive answer (except the almighty ‘zon), I came across an interesting theory on Dave Chesson’s blog that may shed some light.
Amazon’s time-stamped search URLs
Say you want to share the Amazon link to one of your books. Maybe you’re sending it to a friend. A natural thing you might do is:
Head to Amazon.com;
Enter your book title into the search bar;
Click on your book and land on its Amazon page;
Share the URL of that Amazon page with your friend.
To give you an example, I just did this with David Gaughran’s excellent book Let’s Get Digital. I searched for “lets get digital” and clicked on the first result. I got the following URL:
As you can see, the full URL contains the keywords I searched for. But it also contains an important parameter that I've highlighted: QID.
QID is basically a timestamp of when the search was made. It records the number of seconds that have passed since January 1, 1970. If I were to do the exact same search a second later, I’d be getting qid=1532003912.
This means that every link you generate this way is unique. Amazon knows when the search was made, and what keyword it was related to. If its algorithms detect that a large percentage of a book’s reviews come from readers who got to the book through the same time-stamped URL, they’re likely to flag these reviews as biased and remove them.
For example, such a link was shared on Reddit by Trump supporters who wanted to ‘brigade’ Megyn Kelly’s book with one-star reviews. Because Amazon could identify all the reviews that came from that single link, they were able to delete them (news story here).
So, which link should you use? You want one that is stripped of any parameters — get rid of everything that comes after the first long number. In the case of Let’s Get Digital, it would be:
You can also use affiliate links from them Amazon Associate program since those have no keyword tagging nor timestamps.
Is this enough to guarantee safety from any review deletion? Probably not, since Amazon is likely to have other automated ways of flagging biased reviews. But it might reduce the risk by quite a lot.
Till next week!