Amazon is currently cracking down on what it sees as inaccurate or reciprocal reviews and it appears, at least from recent posts I’ve read, a number of authors have been affected. Amazon are using an algorithm to identify what they term as suspect reviewing patterns, as well as identify reviewers who they believe know each other, and blocking those reviews. Once blocked, because Amazon believe the reviews broke their reviewing terms and conditions, the reviewer can no longer leave any future reviews. When challenged, Amazon have generally given automated responses along the lines of ‘we trust our algorithms and you have no right of appeal.’
Before I go any further, I want to make it clear I have every sympathy with the individuals concerned. Writing just one thoughtful book review takes time and effort, to do it over and over again is a considerable commitment. To be told, out of the blue, that everything you’ve written will be stripped from the site, would be one hell of a shock. Then there is the underlying accusation of cheating a system, one to which the reviewer has no right of reply, and the fact that once banned they can’t write any further reviews. It is being found guilty without trial and goes against everything we know as fair. Not only is it a personal affront, it means the authors of the books reviewed lose both the review and the rating as well. If (or possibly when) it happens to me, I would be furious too.
At the same time, Amazon has a problem. Customers no longer trust their review system and in some cases with justification. If you are selling a product on Amazon (any product, not just books) there are plenty of companies willing to give favourable reviews in return for a fee. The term for this type of behaviour is astroturfing and it happens on all the major sites where customer reviews play a part in the purchase decision making process. And it’s not just companies offering this service. I’ve been approached – both explicitly and implicitly – by authors asking to swap reviews. It hasn’t happened often, and I’ve always declined, but it does happen, and if a relatively obscure author such as myself has been approached then this is clearly something that some authors are happy to take part in. Astroturfing’s not a new process, it’s been happening since the first review sites were established and is employed by many companies large and small, but it has become so common in recent years it’s got to the point where customers have lost faith in product reviews.
The one thing you need to remember about Amazon is that their number one priority is to their customers. Everything they do is focussed on providing the best service to their customers. They are very good at this, and it’s the primary reason they have become so successful. Amazon regularly top the polls for best companies by as rated by consumers because they always put customers first, so it should come as no surprise that when their customers no longer trust the review system, Amazon decide to do something about it.
The problem for Amazon is how do they identify, out of the millions of products they sell and the tens or hundreds of millions of reviews on their system, which ones are the bad reviews. There are too many to analyse by person – I worked out that if you had one hundred million reviews and 1% were seen as problematic, it would take over one hundred man years to check them all – and it’s pure cost as far as Amazon are concerned. And it’s not easy. If you look at the books I’ve highlighted as Recommended Reads, each with a corresponding review on Amazon and Goodreads, almost all are by authors I don’t know personally, one or two are by authors I’d met previously through social media, and a number are by authors I’ve since got to know on social media, often because I’ve promoted their books. A handful have gone on to review my books. I’ve always been genuine with my praise and have never requested or expected a review in return – favourable or otherwise – but there are enough connections there for some people to question the validity of my reviews, adding to their mistrust of the system.
And customers regaining trust of the review system is at the heart of what Amazon are doing. So, rightly or wrongly, Amazon have decided to cut off the leg to save the body. They are doing this across all product areas, using an algorithm to identify suspect patterns and connections, and automatically removing reviews they believe could be false, banning those accounts highlighted from generating reviews. And this means some innocents will be caught up in the process. And while that’s bad news for those affected, Amazon see it as a small price to pay compared to their customer regaining trust in reviews.
But as an author, I want people to believe in the reviews of my books. I’m lucky enough to have received some great reviews, a number of them from other authors. I’d hate to lose these reviews but if it meant readers placed more trust in those that remain then so be it. We still have the opportunity of placing those reviews in the ‘Editorial Reviews’ section of the product page, it’s just they would no longer count as part of the overall star ratings and average.
Of course, I would prefer Amazon found a better way to clean up the review system, and as mentioned earlier, I have every sympathy for those wrongly caught up in this process, but if it provides a system readers trust, that can only be of benefit to all of us.
What do you think? Do you agree with me or do you think I’m completely wrong? What are your views on what Amazon is doing, or the review system in general? I’d love to hear from you.
Do you like intelligent thrillers? If so, join my mailing list and get one of my 5-star rated near-future dystopian thrillers absolutely free. The mailing list is guaranteed spam free and I will only contact you if I have a new book launch or an exclusive short story to share. To sign up, please click here.
Suffolk Scribblings 
Interesting read, Dylan. Like you, I’ve met people on social media and have reviewed their novels, not asking for anything in return. Naturally I’ve been delighted if some of these have subsequently left reviews of my novel, and would be disappointed if these disappeared. But I agree that the Amazon review system is open to abuse and it would be a small price to pay for a fairer system.
I would be very disappointed if they disappeared, too. I think it all depends on how stringent Amazon are. The stories we’ve heard are by those authors affected but we’ve no idea how many have been checked and found to be OK. Only time will tell.
As indie authors, we’re told we need to build our audience ‘one reader at a time’ through social media, and most notably through blogging. As an indie author, if you don’t blog, readers won’t find you – it’s that simple. If you do blog, and if you do it effectively, it creates personal one-to-one connections with readers and fellow writers (that would be… um… people you then ‘know’). From that, ideally, comes their interest in books you have written. That’s how it’s supposed to work!
If indie authors had to guard against inadvertantly receiving reviews from people they ‘know’ through social media, they would have to become unresponsive to comments and feedback on their blogs, ceasing to engage with anyone who might – heaven forfend – accidentally evolve from being an itinerant commenter into somebody they ‘know’ who might then – oh, no, here we go – buy, read, enjoy and then write a review of their book! Or else they would have to specifically ask visitors to their blog ‘not’ to buy and review their books! Then what would be the point of blogging to, oh… promote yourself as a writer?
My tongue is a little in my cheek here, as you probably guessed, and I’m not in principle opposed to Amazon’s attempts to clean up its reviews. But I think it will be a tragedy if a review from a fellow writer who happens to have connected with you through the blog you write (because you want to promote your writing), is interpreted somehow as fraudulent or misleading.
It’s a tough one, and as I’ve mentioned in another comment, how much this approach will be affected will depend on exactly what criteria Amazon are looking for and how stringently they apply it. For example, many of us who have published through KDP have used our Amazon customer logins to set up the account. It wouldn’t take a lot of processing power to work out that the Author Dylan S Hearn is the same as the customer Dylan S Hearn. They could easily remove my reviews if they were targeting authors leaving reviews but haven’t (as yet). Therefore there must be more to what they’re looking for. We’re hearing from those authors affected but have no idea on how many have been judged and found OK, or even what issues they found with those who’ve been banned. Only time will tell if this is a blanket crackdown or a targeted one with a few innocent casualties.
I don’t know, but I suspect the “blocking people you know from reviewing your book” is really “blocking you from leaving reviews of your own book” because they’re tracking IPs. It’s Amazon trying to be polite about it, instead of coming straight out and accusing people of writing dummy reviews. I seriously doubt (but again, I don’t know) that Amazon’s algorithms are tracking FB friends, or LinkedIn connections, or blog followers.
Let’s hope not, because that would be truly sinister…
I know Amazon have blocked individual reviews for a while that have been generated from the same post code/zip code as the author, although they haven’t blocked the individuals writing those reviews.
I think in this case they’re looking for, what are in their eyes, serial offenders. Both the cases I’ve heard of have involved people who’ve written a lot of reviews (in the thousands in one case). However, nobody really knows for sure and Amazon aren’t saying.
Reblogged this on Jan Hawke INKorporated and commented:
Some more food for thought on the ‘review for a friend’ policy on Amazon debate… looks like this one will run and run!
Thanks so much for sharing. Yes, it’s a tough one. I hate seeing innocent people being caught up in this but at the same time I can see Amazon’s dilemma and can’t think of how else they could (cost effectively) handle it.
I agree that is a problem. A lady I know, whose poetry I have read and admired on her Blog has just published a book of poems which I have purchased and would normally review, but knowing she has already reviewed both my books, favourably bless her, I am reluctant to post a review for her on Amazon for exactly the reasons you state
I think it’s a dilemma we all face, Peter. It’s only human nature to want to read books by authors (or poets) who have enjoyed your own. Where I’ve read and enjoyed a book by somebody who has already reviewed one of my own, or if I’ve been involved – as a beta reader for example, I always make it clear at the beginning of the review. That way, if somebody feels it may have an influence, they can take it into account.
Reblogged this on Armand Rosamilia and commented:
Amazon Review Deletions
Thanks for reblogging this 🙂
It’s a double-edged sword. Naturally, as self-published authors we want our family and friends to be honest about our work which is why when I released Dead Heat I really could not reiterate enough to the friends I sent the copies to, the need to give an honest and accurate appraisal on Amazon – to write the sort of review with enough information that strangers would be informed as to whether or not it is their sort of thing. Thankfully, those who did posted some great reviews. All of these people know me, so hopefully they will not be affected.
I can attest to that. There is a lot of it on Upwork and I have been approached more than once. I didn’t want to use tie my Amazon account to dodgy reviews for this very reason.
I remember one author on twitter stating that if you only ever wanted to receive positive reviews, don’t publish, just give your book to your mum to read. I replied that he’d clearly never met my mum. 😉
I would have thought that the reviews to be taken with a pinch of salt are the reviews that are bought, and I suspect you can pick them out because the reviewer won’t have read the book. What’s the point if the author has paid for you to say wonderful things about it? Reading it is a waste of time; all you need is to be liberal with the awesomes and I-read-it-in-one-sitting-type comments. A review that mentions specific points and gives reasons for liking particular aspects means that the reviewer has read the book. To my mind, those are the indicators of a fair review, even if the reviewer might have kept quiet about some of the short-comings. Amazon shouldn’t assume that just because a reviewer might be from the same background as the author that she/he is going to misrepresent the book.
I’m hoping that the Amazon algorithm is a little more sophisticated than just looking to see if people are related but the reality is that the only people who know how it works are Amazon. I completely agree with you that bought reviews are a terrible waste of time. Not only are they bad from an ethical perspective, they’re a false economy. If you’ve written a good book, you will get good reviews. If you haven’t, you won’t. The reviews may come a lot slower than you’d like, but if the book is good enough, the reviews will come.
This is a tricky one, to be sure. Personally I’m against trusting algorithms as the end-all be-all answer to astroturfing. Not because it might be unfair, but because it’s a Scotch-tape answer to the leak in the dam. I get the theory behind it and what Amazon is trying to do with it (and after all, our Electronic Overlords know all, yes? 😉 ), but too often I’ve seen “this is what the system says” trump the logic of a situation. You could simply be a blissfully happy reviewer who reviews everything they read (and I know plenty of people who do so). Who’s to say a simple and honest “I loved my buddy Jim’s book” won’t trip an alarm and nuke their entire review collection?
Not to say it’s not a bad idea…just saying it seems a flawed quick fix that could easily backfire. And I’m just more of a fan of well thought-out fixes instead, even if they do cost a bit more. 😉
I’m in danger of coming across as an Amazon apologist – which I’m not – and like you I don’t like the thought of using an algorithm as the sole arbiter of defining the legitimacy of reviews, but at the same time I can’t think of an alternative that wouldn’t cost the earth. Having worked for large multinationals for many years, I can say that these decisions aren’t taken lightly. They may not always be the right decisions, but somebody, somewhere, has calculated the pros and cons of using this method and decided that it’s a risk worth taking. That said, a little more transparency on the process wouldn’t go amiss.
Oh, totally understandable, and I agree, a bit more transparency might be good. I’ll admit, I’m speaking from the POV of working in the EDI division of a major bank (translation: all that fun fiddly processing that goes on behind the scenes with business e-banking), so I’ve seen both perfect fixes for problems and “there I fixed it” kludges. Amazon’s move here seems more of the latter than an attempt at the former. I don’t have an answer here either, but I’m sure someone somewhere does! 🙂
My guess is that they’ve decided to start tough then tweak depending on feedback. As you say, it’s tough on those unfairly caught up in it.
I should also say, many congratulations on posting the 1500th comment on this blog. As a prize, you get to receive my undying thanks 🙂
Like you, I think there is some good that could come from this, but though I understand they want to rid books of reciprocal author reviews, I would think this is less of a problem then those who pay sites to generate lots of reviews for them. I hope that gets equal, if not more, attention.
But also like you, I hate to see an innocent caught up in it. I, too, have reviewed books of authors who have subsequently reviewed my mine, but it was never a “I’ll review your book if you review mine” type of thing. I read a lot of indies, and just like traditionally published books, I approach each review I leave as a reader, not a writer, and I choose the books I read from the same vantage point. If too many innocents get caught up, then many of us who read indies and leave reviews might be hesitant to do so in the future, and that would be a shame.
Thanks for sharing this, Dylan. Informative as always.
I’m with you on pay sites, definitely. And I agree, it would be a shame if authors (who are also readers) feel reluctant to leave reviews for books they’ve enjoyed. Many of my favourite books over the past couple of years have come via recommendations of other authors, and not the ones that wrote them.
Tough question. From Amazon’s pov, it’s a fine line between gaming the system and using the internet to find new authors to read and posting reviews of the one’s you like. I’ve met some great people and read some great books, and reviewed them. It’d be a shame to throw a wet blanket over that.
Amazon algorithms have never struck me as particularly clever, but I suppose they’d rather cut deeper and take no chances.
Sorry for the late reply. For some reason your comment ended up in spam!
Yes, I agree with you. It would be a shame if that happened but I still believe that if it was a choice between having some (or even all) author reviews banned and a system the majority of readers trust, compared to a system few trust, I’d take the former.
So my problem is that, you state that Amazon is all about the customer and yet they are turning around and calling their customers liars, cheaters, and scammers with no recourse for the customer. I realize there is a real problem they are trying to address but when regular customers get caught up in the attempted solution with no ability to defend themselves, that’s a bigger problem. It’s Amazon saying, “We’ll take your money but we think you are the Bernie Madoff of reviews so we won’t let you post them ever again.” How is that being customer focused?
Hi Melissa, I think Amazon are all about doing what’s best for their customers as a group but are willing to risk letting individual customers go if the vast majority benefit. I’m not saying I agree with their approach, and the lack of transparency or right of appeal with their process is a major problem, I’m just saying that if only a handful of customers from millions are affected, they would probably see that as a price worth paying.
As I said in my post, I have the utmost sympathy with those unfairly caught up in this. What I hope is that Amazon will change their review policy, explaining to individuals why they have been blocked and allowing them the right of reply.
Reblogged this on Traci Bold and commented:
This is an excellent blog about Amazon’s new review policy and it worth reading and thinking about. I do agree with Dylan Hearn on all his points here. The whole scenario is a catch 22 for Amazon and I have faith that they will clear it up professionally and amicably to ultimately benefit the authors they serve.
Thanks for reblogging this. I, too, hope Amazon clear up the review system in a professional way. I’m concerned for those that have been unfairly caught up in the process with no right of reply and I’m hoping these are the few exception and not the sign of things to come.
I have heard stories via all the blog conversation happening around this one of reviewers who are not authors being banned because they reviewed a book by someone they knew through social media. I understand Amazon is concerned about review swapping, but in this type of case, all that’s happened is that they know each other. I find that very disconcerting!
This part is worrying. I read a blog post this morning http://consumerist.com/2015/07/06/amazon-is-data-mining-reviewers-personal-relationships/ saying Amazon are data mining social media interactions of their customers. If it’s true, it’s a major breach of trust and I’m sure consumer groups and governments will get involved. However, there is nothing in the article providing proof of the claim so I’d be careful jumping to any conclusions until we know more.
Yes, right now it’s all anecdotal…we’ll have to see how the story develops! Companies are so driven to mine social media these days that accusations like this are bound to come up more often. The thing about social is that it is (largely) public, so if Amazon was doing this I don’t think it would be illegal, but it is certainly creepy.
Stipulating that you must have read the book you want to review should be good enough. I can’t help thinking that a lot of authors will be going elsewhere..mainly those who don’t get many reviews and treasure the ones they do get from family and friends.
First, I can’t let this pass. “There are too many to analyse by person – I worked out that if you had one hundred million reviews and 1% were seen as problematic, it would take over one hundred man years to check them all.”
It would take a woman half as long. 😉
Agree on all points, Dylan (except for the bit about Amazon looking out for the customer. Amazon is always looking out for turning a profit, and if that goal intersects with the customer’s interest, it’s incidental). I’ll say it, even if you didn’t in your post: the only folks whining about this policy are the people who were trying to scam the system. And it’s not just self-published authors who were pulling these stunts. I just read a novel published by Penguin, a zillion five star reviews, you would think it was the Greatest Story Ever Told, and it absolutely sucked. And not in the “not my cup of tea” range of suckage–in the I-can’t-believe-this-author-isn’t-hiding-in-shame-after-attaching-his-name-to-this-piece-of-dreck range. Cringe-inducing dialogue, insipid plot.
Sorry! Of course you are correct. Half the time and twice as good, I believe. 🙂
*strikes that phrase from his personal lexicon*
You’re right about it not just being indie authors who do this. Publishers have a long tradition of getting authors from their own stable to write gushing blurb for the work of others. Some may well be genuine but I’ve heard enough accounts to know it’s not all. Whether they take part in astroturfing is another point but given how many companies do I would;t be surprised (although I hope I’m wrong).
That’s another really interesting post, Dylan. Thanks!
You’re welcome, Marcus!
When I first tried to open this post, it showed up as page not found. The second attempt was the same. So I was beginning to think Amazon had even managed to delete your posts! (Not that I’m into conspiracy theories, you understand…)
It could be, although as a proud believer in science it’s important to remember that correlation doesn’t mean causation 🙂
A nice, balanced outlook, and one that I share. Thanks for that 🙂