Browser Extensions 

brigade-featured

Review Brigades

December 2nd, 2016

Brigading is a term used when online trolls group together to flood another site (or subsection of a website) with their message.  On Amazon, this happens when a product is quickly flooded with negative reviews, often politically motivated, and likely by reviewers who have not actually used the product or read the book.

We’ve seen some pretty high-profile examples in the media recently:  

Here’s a few more lesser known examples:

So How Can We Detect Brigading?

There’s a few tell-tale signs of brigading that are very easy to notice once we’ve run a report on the reviews:

1. Rating from Unverified Purchasers is much lower than the rating from Verified Purchasers

uvp-brigade

Since the brigaders will not go out of their way to purchase the item before leaving a nasty review, you’ll often see an unusually high amount of unverified purchases, which are much lower on average than the verified purchase reviews.  

2. High number of Deleted Reviews

deleted-brigade

A high number of deleted reviews with a low average rating does not mean that Amazon is taking sides and trying to silence a group of people.  Amazon is simply doing their job of removing reviews from people who obviously have not used the product.

3. Lots of negative reviews appearing all at once

rt-brigade

Our Rating Trend test can be highly telling of when the brigades happen.  Usually, brigades are organized on different sites outside of Amazon (Reddit, Facebook, Twitter), and then inspire a flood of reviews all around the same date.  If the rating from reviews on High-Volume days is considerably lower than the rating from reviews on Normal-Volume days, it can be a sign of brigading.

4. Lots of reviews from Suspicious Reviewers

su-brigade

Because many who participate in brigades aren’t regular Amazon reviewers, brigades often trigger a lot of warns or fails for our Suspicious Reviewers test.

This test has four parts, and any or all might be a warning sign of a brigade:

  • One-Hit-Wonders: Reviewers who have only written one review on Amazon.
  • Never-Verified Reviewers: Users who have never left a “Verified Purchase” review.
  • Single-Day Reviewers: Those who have posted all their reviews on one day.
  • Take-Back Reviewers: Accounts who have reviews that have been deleted previously.

Our algorithm will look at these red flags and warn you about possible brigades.

brigade-warn

While our warning is wonderful tool to help you quickly identify when a product is being brigaded, it’s always a good idea to look through the rest of the report yourself and ultimately make your own decision about the authenticity of the reviews.


  • Dave Becker

    It should be ‘number of reviews’ not ‘amount of reviews.’

  • roxtoto

    doing god’s work, dudes…. ever since i noticed those pesky (horrid even) pro reviewers who are payed by receiving the reviewed items for free, i’ve been itching to find a counter. You’re brilliant at it, thank you thank you thank you. I still tag those fake reviewers “unuseful” out of contempt, but no more out of impotent rage. Soiling one of the last commerce-free spaces of the internet that is user reviews is quite the nasty move. You’re one of the few sites i will hold down my adblock for.

  • Di

    Not all reviews that are unverified purchasers are illegitimate. Some are people who have borrowed the book on KU and others are people who have bought it in their home country but their purchase doesn’t show up as Verified on other Amazons. I’m not sure why Amazon does this – you can buy the exact same product and yet your account on another Amazon doesn’t flag that you actually HAVE purchased that product. And, of course, the fake reviewers all ‘purchase’ the product when the seller puts it up for free so yes, it’s verified but that really doesn’t mean a whole lot in that context. Is there any way for ReviewMeta to distinguish people who have purchased the product on another Amazon? The username is the same and the product ID number is usually also the same so maybe it could be done?

    And thanks for your site. It’s a real gift to people who are trying to work out what products are genuine and that’s appreciated.

    • Hi Di-

      Yes, this is correct, just because a reviewer is unverified or verified doesn’t determine if the review is fake or not. Out of our 15 tests, only one looks at verified vs. unverified. All other tests completely ignore this attribute. We’re definitely seeing a lot of “Verified” reviews being generated from these underground review clubs – where the sellers offer a full paypal refund that Amazon has no way of tracking.

      As far as trying to confirm that someone has actually purchased the product on another Amazon, there is no reliable way of doing that. Yes, sometimes the product ASIN is the same across the different regional sites, but everything else is separate. The reviewer ID is going to be different even if it’s the same person. The reviewer name might be the same, but that’s something you can pick and change whenever you want. Also, there’s literally millions of reviewers named “Amazon Customer” across all regions.

  • Betty G.

    I am actually trying to leave a negative review on an item right now that has a negative brigade warning up now. I checked this site earlier b/c I suspected a lot of fake positive review on an item, but ordered it anyways despite the suspicious label b/c it was an Amazon lightning deal. I received the headphones and realized they’re garbage and am trying to leave a negative review about it b/c they are absolutely not even close to noise cancelling. I suspect that a lot of other folks who took advantage of the lightning deal may be legit reviewers who are negging the product b/c it is absolutely not delivering what the inflated reviews suggested it should have. Just one perspective.

    • 7998rt

      Anyone should reasonably be able to tell very easily just by the type of product whether the flood of negative reviews is “brigading” or not: If the product is a book, a movie, something political, or in some way controversial, then it’s likely “brigading”. If it is, as with your example, a pair of headphones, then it’s very likely that the negative reviews are valid. There’s probably not a political crusade against your headphones. Any computer algorithm like this or any other site uses can only do so much. It’s a formula. It can’t make subjective judgement calls. The user has to apply some common sense as well.

      • Absolutely. Controversial items are going to be more likely to be brigaded. However you also see it on some items that are from big brands that people love to hate, for example: Comcast.

  • evo34

    How about the opposite — a flood of fake positive reviews? DO you have an alert for that?

    • That’s pretty much what our standard report is trying to look for.