Reviews, Darned Reviews, and Statistics

A friend recently pointed me toward an article entitled, “Users who post ‘fake’ Amazon reviews could end up in court.” I’ve known for a long time that some authors do pay to get positive reviews for their books posted. In fact, some authors stoop to paying for negative reviews of competing works as well. Even though the actual technique used for cheating on reviews has changed, falsifying reviews is an age old problem. As the Romans might have said, caveat lector (let the reader beware). If there is a way to cheat at something, someone will most certainly find it and use it to gain a competitive advantage. Amazon and other online stores are quite probably fighting a losing battle, much as RIAA has in trying to get people to actually purchase their music (see Odd Fallout of Digital Millennium Copyright Act (DMCA) for a discussion of the ramifications of IP theft). The point is that some of those reviews you’ve been reading are written by people who are paid to provide either a glowing review of the owner’s product or lambaste a competitor’s product.

Of course, it’s important to understand the reasoning behind the publication of false reviews. The obvious reason is to gain endorsements that will likely result in better sales. However, that reason is actually too simple. At the bottom of everything is the use of statistics for all sorts of purposes today, including the ordering of items on sales sites. In many cases, the art of selling comes down to being the first seller on the list and having a price low enough that it’s not worth looking at the competitors. Consequently, sales often hinge on getting good statistics, rather than producing a good product. False reviews help achieve that goal.

I’ve spent a good deal of time emphasizing the true role of reviews in making a purchase. A review, any review you read, even mine, is someone’s opinion. When someone’s opinion tends to match your own, then reading the review could help you make a good buying decision. Likewise, if you know that someone’s opinion tends to run counter to your own, then a product they didn’t like may be just what you want. Reviews are useful decision making tools when viewed in the proper light. It’s important not to let a review blind you to what the reviewer is saying or to the benefits and costs of obtaining particular products.

Ferreting out false reviews can be hard, but it’s possible to weed out many of them. Reviews that seem too good or too dire to be true, probably are fakes. Few products get everything right. Likewise, even fewer products get everything wrong. Someone produces a product in the hope of making sales, so creating one that is so horrid as to be completely useless is rare (it does happen though and there are legal measures in place to deal with these incidences).

Looking for details in the review, as well as information that is likely false is also important. Some people will write a review without ever having actually used the product. You can’t review a product that you haven’t tried. When you read a review here, you can be sure that I’ve tried out every feature (unless otherwise noted). Of course, I’m also not running a test lab, so my opinion is based on my product usage—you might use the product in a different manner or in a different environment (always read the review thoroughly).

As you look for potential products to buy online, remember to take those reviews with a grain of salt. Look for reviews that are obviously false and ignore them. Make up your own mind based on experiences you’ve had with the vendor in the past or with similar products. Reviews don’t reduce your need to remain diligent in making smart purchases. Remember those Romans of old, caveat lector!

 

Author: John

John Mueller is a freelance author and technical editor. He has writing in his blood, having produced 123 books and over 600 articles to date. The topics range from networking to artificial intelligence and from database management to heads-down programming. Some of his current offerings include topics on machine learning, AI, Python programming, Android programming, and C++ programming. His technical editing skills have helped over more than 70 authors refine the content of their manuscripts. John also provides a wealth of other services, such as writing certification exams, performing technical edits, and writing articles to custom specifications. You can reach John on the Internet at [email protected].