Select your localized edition:

Close ×

More Ways to Connect

Discover one of our 28 local entrepreneurial communities »

Be the first to know as we launch in new countries and markets around the globe.

Interested in bringing MIT Technology Review to your local market?

MIT Technology ReviewMIT Technology Review - logo

 

Unsupported browser: Your browser does not meet modern web standards. See how it scores »

{ action.text }

On Amazon, consumers will pay a significant premium – at least five percent of the cost of a product – to purchase it from a seller with a good reputation. What’s surprising is that once you get beyond the star rating of a seller and the number of transactions they’ve completed, the text of user reviews of that seller can affect their sales in strange ways, said researcher Panagiotis Ipeirotis in a recent talk, embedded above.

For example, you would think that a user who said that something was “Best camera!” would be increasing the reputation (and sales) of that item. But this turns out not to be the case, and Iperotis has actually quantified the effect. A review like this reduces sales, on average, by 0.2 percent.

The reason, Ipeirotis hypothesizes, is that users take reviews in context. Reviews that say “Best camera!” don’t tend to include other information.

It’s the presence or absence of this information that makes buyers confident in a product, apparently. For example, negative reviews that are well written can actually increase sales of a product. One reason is that buyers gain confidence that “if this is the worst this product will throw at me, it must be pretty good.”

“Negative reviews that are specific actually tend to serve as risk mitigators,” says Ipeirotis, who has seen the effect in everything from hotels to video games.

On the other hand, the phrase “good packaging” makes it sound like there was nothing else in the transaction worth complimenting. Result: -0.56 percent in sales. “AAA++++ seller” is even worse: -2.93 percent in sales.

Another important discovery Ipeirotis made is that there is a significant disparity between what people talk about in reviews and what actually moves product. Reviewers love to talk about a camera’s zoom, for example, but in reality battery life has more influence with buyers.

Ipeirotis also discovered that grammar and spelling matter. This is an unconfirmed rumor, but he says that he heard that Zappos spent hundreds of thousands of dollars on Amazon Mechanical Turk to have the spelling and grammar of all the reviews on its site improved, at a cost of $0.10 per review, and that the result was increased sales.

Ipeirotis’s core message is that online, there are few useful assumptions about user behavior, and you always have to measure what’s actually going on in order to discover useful trends.

2 comments. Share your thoughts »

Tagged: Web, Amazon.com

Reprints and Permissions | Send feedback to the editor

From the Archives

Close

Introducing MIT Technology Review Insider.

Already a Magazine subscriber?

You're automatically an Insider. It's easy to activate or upgrade your account.

Activate Your Account

Become an Insider

It's the new way to subscribe. Get even more of the tech news, research, and discoveries you crave.

Sign Up

Learn More

Find out why MIT Technology Review Insider is for you and explore your options.

Show Me