Amazon Overhauls review system

[fjtorres]

From CNET:
http://www.cnet.com/news/amazon-upda...ning-platform/

Amazon is rolling out a big change to its customer reviews system in the US, introducing a new machine-learning platform it developed in-house to surface newer and more helpful reviews.

"The system will learn what reviews are most helpful to customers...and it improves over time," Amazon spokeswoman Julie Law said in an interview. "It's all meant to make customer reviews more useful."

The change, which started Friday, will probably go unnoticed at first, as the e-commerce giant's new platform gradually starts altering the star ratings and top reviews on product pages. The new system will give more weight to newer reviews, reviews from verified Amazon purchasers and those that more customers vote up as being helpful.

A product's 5-star rating, which previously was a pure average of all reviews, will also become weighted using those same criteria, and so may change more often.

There's no word yet on whether this new platform will roll out to other countries.

More at the source.

[Phogg]

Originally Posted by fjtorres

From CNET:
http://www.cnet.com/news/amazon-upda...ning-platform/






More at the source.

I read this as basically a scheme to bury critical reviews.

Already the system of users tagging reviews as either helpful or unhelpful flagged the better reviews. I suspect someone wants to sell the ability to circumvent the consensus of readers to those who will pay to have that done.

Of course, I also believe that Amazon bought Goodreads for that purpose.

[Canuck_in_Japan]

You shouldn't be able to write a review at all unless you are a verified purchaser. That would cut down on a lot of the paid for reviews.

[HarryT]

Originally Posted by Phogg

I read this as basically a scheme to bury critical reviews.

I see nothing in it which suggests that. Which specific part of the linked article leads you to that conclusion?

[DiapDealer]

I'd rather they not manipulate, or give more weight to, the reviews which make it in front of me at all. I'm perfectly capable of navigating/utilizing the good, the bad, and the irrelevant all by myself. I'm not an idiot.

[dickloraine]

Originally Posted by Phogg

I read this as basically a scheme to bury critical reviews.

Already the system of users tagging reviews as either helpful or unhelpful flagged the better reviews. I suspect someone wants to sell the ability to circumvent the consensus of readers to those who will pay to have that done.

Of course, I also believe that Amazon bought Goodreads for that purpose.

Why should they do this? Longterm it would only damage their business. For amazon it doesn't matter that much which book you buy for example. What matters is that you buy the books on amazon. Taking a little bit of money from producers to make reviews screwed would harm the overall business, since people won't trust amazon anymore. I can't believe they are that stupid.

[PatNY]

I really don't see the review placements changing much from these new rules as the reviews with the most thumbs up were already being shown front and center. I don't see how this would change much.

Critical reviews won't be buried at all if they are helpful. They could instead jump up to the front. Unhelpful critical reviews will be buried -- as they should be.

What I wish Amazon would try to change and minimize is their practice of having multiple product options on one product page -- which results in different product reviews being mixed together. This is very unhelpful and can be very misleading. This has bothered me for years.

--Pat

[robko]

Originally Posted by Phogg

I read this as basically a scheme to bury critical reviews.

Originally Posted by HarryT

I see nothing in it which suggests that. Which specific part of the linked article leads you to that conclusion?

I'm not sure if it's specifically designed to do this or not, but the natural tendency based on the weighting system is likely to mean that it will do so. On a "popular" product the majority of people are going to be looking for the "positive" reviews to confirm their intention to purchase and naturally are going to flag those as helpful. People that may have a legitimate concern or specific dislike not shared by the majority (say if the e-reader is IR vs capacitive touch, or has buttons vs not) are going to get buried because not nearly as many people flag it as helpful. In some ways this will also have the tendency to bury info functions not as many people are interested in or are lesser known amd might not be covered in the product info blurb (again because they aren't going to get the same amount of helpful flags).
The best example I have of this I've come across was a blog on the topic where the blogger was talking about a researching what he should buy for new digital camera. The one reviewer was bashing it because it didn't have this function or that function and basically said you'd be crazy to buy it. But, the negative review told the blogger everything he wanted to know because he was after a really simple point and shoot so the reviewed camera was exactly what he wanted.

I agree with DiapDealer. I can wade through and decide for myself. I actually spend more time looking at the negative reviews because they tend to give a lot of details that aren't in the description.

[fjtorres]

The fact that they are using machine learning algorithms tells us everything we need to know: they are looking for patterns.
The kind of patterns that come with 1-star campaigns, sock puppets, paid reviews, and other attempts to game the system.
Positive reviews or negative reviews won't matter any more than before; they just want honest and fair reviews.

Just remember that amazon is about more than books. And that last year's "world's best whatever" can easily be this year's "overpriced junk" so yes, newer reviews should weigh marginally more than older ones.

I just ran into a perfect example today with my WD TV streaming box; a few months ago it received a firmware update that changed the GUI. Much improved. Today I noticed it enabled MIRACAST support. Two year old product. Hundreds of reviews. A year ago it didn't even mention MIRACAST. Now it not only has it, it works well. The new review is more relevant to the product as it exists so it *should* weigh slightly more.

Anybody remember these?

http://www.amazon.com/Hewlett-Packar...+560+pocket+pc

Great deal?

All they are doing is what Google does constantly; look for bad actors trying to misuse the system and try to cancel them out. Somebody will always scream and grumble but at least they're trying. I don't hear eBay or Rakuten even trying.

[Josieb1]

Originally Posted by Canuck_in_Japan

You shouldn't be able to write a review at all unless you are a verified purchaser. That would cut down on a lot of the paid for reviews.

I review books on Amazon, I post the same review as I do on Goodreads. I don't always buy my books at Amazon though, I buy at Allromance or the publisher direct so making reviews only for verified purchases would make me only be able to review a book if I bought it from Amazon. That's unfair to the authors and in a way makes Amazon a monopoly seller and in many ways it's already that

[fjtorres]

Originally Posted by Josieb1

I review books on Amazon, I post the same review as I do on Goodreads. I don't always buy my books at Amazon though, I buy at Allromance or the publisher direct so making reviews only for verified purchases would make me only be able to review a book if I bought it from Amazon. That's unfair to the authors and in a way makes Amazon a monopoly seller and in many ways it's already that

Plus, people review library books, gift books, loans from friends, books they read in the past...

[pwalker8]

Originally Posted by robko

I'm not sure if it's specifically designed to do this or not, but the natural tendency based on the weighting system is likely to mean that it will do so. On a "popular" product the majority of people are going to be looking for the "positive" reviews to confirm their intention to purchase and naturally are going to flag those as helpful. People that may have a legitimate concern or specific dislike not shared by the majority (say if the e-reader is IR vs capacitive touch, or has buttons vs not) are going to get buried because not nearly as many people flag it as helpful. In some ways this will also have the tendency to bury info functions not as many people are interested in or are lesser known amd might not be covered in the product info blurb (again because they aren't going to get the same amount of helpful flags).
The best example I have of this I've come across was a blog on the topic where the blogger was talking about a researching what he should buy for new digital camera. The one reviewer was bashing it because it didn't have this function or that function and basically said you'd be crazy to buy it. But, the negative review told the blogger everything he wanted to know because he was after a really simple point and shoot so the reviewed camera was exactly what he wanted.

I agree with DiapDealer. I can wade through and decide for myself. I actually spend more time looking at the negative reviews because they tend to give a lot of details that aren't in the description.

The lack of transparency does bother me, especially since I've noted that Amazon's ebook suggestion engine seems to be more and more weighted towards Amazon digital services books. Of course, it's probably more likely an attempt to handle some the little loopholes that people have found to game Amazon's review/rating system. I question how successful it's going to be.

[HarryT]

Originally Posted by Josieb1

I review books on Amazon, I post the same review as I do on Goodreads. I don't always buy my books at Amazon though, I buy at Allromance or the publisher direct so making reviews only for verified purchases would make me only be able to review a book if I bought it from Amazon. That's unfair to the authors and in a way makes Amazon a monopoly seller and in many ways it's already that

The problem is, though, that there's no way to distinguish your reviews from fake ones who are out to either promote their friend's book or to trash it for some reason. Verified reviews are more trustworthy in the sense that you know that the reviewer has at least bought the book.

[thewitt]

Though Amazon will weigh reviews from verified sellers higher than those who purchased or received the book elsewhere, they will still accept all reviews. I don't see any issue with this. It's not a reader awareness and service site like Goodreads, it's an online store.

[harriska2]

I've had my issues with amazon's review process. We bought a Traeger Jr pellet BBQ at Costco. The design is one which could burn something down so we had to re-engineer it with a different switch. I posted that to Amazon and gave it a 2 start review. You could tell all the fake reviews there - generic, one name user, few previous reviews. But funny enough, people believed my review and it gained thumbs up (or whatever it they call it). Low and behold, a year later they no longer sell that model and modified it to do what we had suggested (and done ourselves) so it doesn't burn backwards into the pellet hopper.

The review process can work and I do think that the steps they are taking could end up being worthwhile.