Quick Links

Fake reviews are everywhere online and they've probably influenced some of your purchases. Fake reviews can be positive or negative, and they're obviously unethical and harmful. But they're the symptom of a larger problem with e-commerce and online platforms.

Reviews Are Worth Their Weight in Gold

In the world of e-commerce, reviews are the best sign of success or failure. Sure, a batch of good reviews shows that a product has sold well and that people enjoy it, but it's more than just that. Reviews are the ultimate form of advertising.

Businesses that have good reviews get a lot of free exposure. On websites like Amazon and Yelp, products and pages that have good reviews show up at the top of search results. They're suggested to users based on interests, and they're even sent out in email campaigns or marked with labels like "Amazon's Choice."

And while exposure doesn't guarantee sales, good reviews do. According to a BrightLocal study, 84% of people trust online reviews as much as they trust friends. That's an incredible statistic because it suggests that a poorly written 50-word review by a stranger can hold as much clout as a positive recommendation from someone that you trust. Essentially, the exposure and clout that businesses get from good reviews create a powerful e-commerce formula.

In some ways, this formula feels a little too easy. But it actually creates a highly competitive marketplace that's easy to exploit. Well-reviewed businesses can easily overshadow competitors, and products or pages with bad reviews are hidden from customers by website algorithms.

This makes sense. Obviously, Amazon and Yelp don't want you to associate their websites with crappy products. But bad reviews can completely break a reputable business, especially a small or new company that's struggling to find a foothold in the market.

You've probably realized where this is going. Businesses need good reviews to stay afloat, so they pay people to write fake reviews.

Who Writes These Fake Reviews Anyway?

Robots or AI don't do the majority of fake reviews; actual people do. It turns out that websites like Amazon are pretty good at catching bot activity, and it helps that most bot-written reviews stick out like a sore thumb (there are even websites, like Fakespot, that can catch AI-written reviews).

But, like other forms of online policing, fake-review takedowns are done manually. Usually, a website will simply target reviews that look "inauthentic." Yelp, for example, tends to take down poorly written reviews by inactive accounts, or accounts that are clearly run by a bot. The website also uses IP addresses to find suspicious reviews. If a restaurant in Idaho has 15 Yelp reviews from an Australian IP address, you can assume that there's some fraud going on.

So, if you're a business that wants to pay for some fake reviews, then your best bet is to find American residents that are active on websites like Amazon and Yelp. Ideally, your fake reviewers are easy to find and willing to do about 10 minutes of work for peanuts. It turns out that freelance writers fit the bill, and there are websites like Upwork, Fiverr, Guru, and Freelancer.com that exist solely for businesses to find and hire freelance writers, often for low-paying tasks.

Keep in mind that a lot of businesses actually hire fake reviewers through a marketing company, so an intermediary may be involved.

Freelance Websites Facilitate Fake Reviews

These freelance websites aren't lawless wastelands. They have rules, they have moderators, and they're totally above board. But oddly enough, fake review jobs don't always violate the terms of service on some of these websites.

Upwork, the largest freelance job board, doesn't explicitly mention fake reviews in its terms of service. Instead, the website's terms of service outlaws "illegal activity," and jobs that violate the "terms of service of another service, product or website."

As you've probably guessed, fake reviews violate Amazon's, Yelp's, Google's, and other websites' terms of service. Plus, a large-scale fake review job could easily qualify as fraud. Even though fake reviews don't explicitly violate Upwork's (and other freelance boards') terms of service, they're technically not allowed because of these two rules.

Well, there's a bit of a loophole. If job listers keep their business vague, they can hire employees for just about anything. Need an example? I searched "Yelp" on Upwork, and I found a job that makes a vague request for "experienced Yelp users." This specific employer is looking for 65 writers and wants to pay them $2 each. It's a painfully obvious fake review job, and it's been posted for more than a week.

Businesses use one more loophole to hire fake article writers, but it's less common. Most freelance job boards outlaw jobs that violate the terms of service of another website, like Amazon or Google. But according to Amazon's terms of use, businesses are allowed to hand out free products in exchange for a review. Other websites, like Yelp, work similarly---you can still review a restaurant even if you ate there for free. So, if an employer offers free products to review writers, then they can indirectly sponsor fake reviews.

That sounds fine and dandy, but most businesses don't want to give out their products in exchange for bad or lukewarm reviews. Businesses playing this game tell their prospective reviewers there's extra compensation for good, detailed reviews.

The Fake Review System Also Exists on Social Media

When most people think of fake reviews, they think about websites like Amazon, Yelp, or Tripadvisor. But to understand just how vast and complicated this phenomenon (and e-commerce as a whole) is, you have to look at how businesses use the system of "fake reviews" on social media.

Let's focus on Reddit. If you aren't familiar with Reddit, it's basically a forum-based website that contains every niche community imaginable. There are subreddits dedicated to boat enthusiasts, MMORPG fans, fashion lovers, and PC nerds.

These communities, along with most other subreddits on Reddit, are tempting targets for a business. If a company that manufactures gaming keyboards manages to get their product to the top of Reddit's PC Gaming forum, then they're effectively advertising their brand to 1.2 million self-described PC gamers. Additionally, a well-disguised Reddit post by said keyboard company could look like a testimonial by a hardcore gamer. And, as we already know, 84% of people trust online reviews and testimonials as much as they trust their friends.

So how do you get to the top of a Reddit forum? Well, content on Reddit that has a lot of upvotes gets more exposure, similar to how well-reviewed products on Amazon get more exposure. Conversely, Reddit posts with downvotes are hidden from users by the website's algorithms. Naturally, you can buy upvotes from a site like BoostUpvotes. And while you'd probably assume that an army of bot accounts does these upvotes and comments, they're done manually, just like fake reviews on Amazon and Yelp.

If you don't believe that businesses use Reddit for marketing purposes, then Google "how to market on Reddit," or "how to sell my product on Reddit." You'll run into some interesting guides, like Dreamgrow's "How to Use Reddit in Your Product Marketing Strategy." When a marketing guide has detailed sections like "Avoid Getting Banned," and "Create a Profile That Feels Real," you know that you're dipping into the realm of unethical business practices.

The fake review e-commerce machine has found its way onto social media pages, and that's kind of bizarre. It's a sign that misleading and unethical business practices have permeated the web, and that businesses are desperate to retain market share in the online world.

Fake Reviews Take Advantage of the Snowball Effect

We know that websites give more exposure to products and content that have good reviews, and they tend to hide poorly-rated content. But sites like Amazon and Reddit can't just show you the highest-rated content at all times; new products and posts need to be able to find their way to the top.

The success of an Amazon product or Reddit post depends almost entirely on its first few reviews. Website algorithms place extra emphasis on a brand new post or product if it has a few good reviews, which increases that product or posts' chances at success. Conversely, a brand new post or product with bad reviews will fall into obscurity and is bound for failure. This is called the snowball effect.

The snowball effect is an easy way to propel fresh, new content into popularity. It's also an easy way to weed out bad content and bad products. But here's the thing---it's tough for a brand new product to get any reviews, even if a well-known company manufactures it. This snowball effect primarily rewards people who pay for fake reviews---or, in the case of Reddit, fake upvotes.

Let's pretend you've just finished writing a romance novel. You put it on Amazon's Kindle store, but it isn't doing very well. It gets a few downloads, and it has a single three-star review. Naturally, your novel disappears into the swamp of unpopular self-published books, and you haven't made a single dollar from your work.

But, what if you decide to republish the book with 20 fake reviews? Well, Amazon might suggest it to customers. It could show up in the "similar" box under other books, or in marketing emails. Your book might run to the top of some search results, and if you make enough sales, your formerly unsuccessful romance novel could end up on a bestseller list.

This sounds like a wild "what-if" scenario, but it happens a lot. The snowball effect is so integral to a product's success that books full of literal nonsense can climb to the top of Amazon's Bestseller list---with the help of some fake reviews. As you can imagine, even legitimate and talented authors have to game the Amazon algorithms to gain exposure, especially when authors with similar books are relying on fake reviews for success.

You Can Also Buy Bad Reviews if You're Feeling Evil

Have you ever felt that weird sense of relief (or disappointment) after reading a negative product review? It's like, you wanted to buy this thing, but you know that you've dodged a bullet because of a review. Well, maybe you didn't dodge a bullet. Perhaps you were manipulated by a fake review.

Bad reviews, dislikes, and downvotes are meant to tell you when a product isn't worth your time. But, like good reviews, bad reviews factor into websites' exposure algorithms. When a product on Amazon or a post on Reddit has bad reviews, it gets tucked away, and it isn't suggested to users through search results. It's like the snowball effect in reverse. As strange as it sounds, you can buy bad reviews, downvotes, and dislikes to hurt competitors. This practice is so common that there are online guides on how to handle negative fake reviews "with class."

This unethical bad review technique can be devastating for a business, especially on Amazon. The website is dealing with a serious counterfeit problem, so it has a few automated systems in place to protect buyers from fraud. If a product gets a lot of bad reviews in a short period, then the person who's selling that product will get suspended from Amazon. Automatically.

In a fascinating video by The Wall Street Journal, Chinese business owners describe how fake reviews have sabotaged them. On important business days like Black Friday, aggressive businesses try to get their competitors suspended from Amazon. They'll buy up a bunch of bad fake reviews and temporarily dominate the market.

These techniques are grossly unethical, and they reinforce the fake-review status quo. When a business falls victim to sabotage-by-fake-review, they have to bounce back quickly. Otherwise, they could go out of business. How can a business bounce back quickly? Well, it can pay writers to leave good product reviews!

Fake Reviews Are a Symptom of a Larger Problem

It's easy to get caught up in the blame game. Fake reviews constitute a significant, unethical business with a lot of moving parts. But before you go pointing your finger at business owners, writers, marketing firms, freelance boards, or Jeff Bezos, consider the idea that fake reviews are a symptom of a larger problem.

E-commerce is all about exposure. You could call it exposure-commerce. A business is simply bound to fail if it can't retain an online presence. Since most online exposure is directly related to algorithms like the snowball effect, business owners are forced to hit the ground running and to avoid making even the smallest of mistakes. And at this point, the ethical decision to avoid fake reviews could even be a fatal mistake for any business.

With the exposure-commerce infrastructure we have now, fake reviews are the status quo. Should we accept fraud as the norm? No, of course not. But, unless online platforms rethink the review-based business model, we're going to keep seeing fake reviews.