Fixing Spelling, Grammar Of Customer Reviews Boost Revenue. But Will It Backlash?

Virtually every E-tailer today embraces the sales-increasing power of customer reviews—and almost no one thinks that tampering with those comments is a wise move. But a recent study from New York University throws a delicious new nuance into the discussion, having found that cleaning up the grammar and spelling of posted comments can sharply increase their sales impact.

The premise that consumers would find more articulate and intelligent-sounding comments more persuasive is as logical as it is disconcerting. It's disconcerting in that it reopens the "should we change the comments in any way and, if we do, how far should we go? And, who should do it and what should the rules be?" debate. Is the boost in sales worth the higher copy editing costs? Are automated systems (Zappos has been using Amazon Mechanical Turk for just such an effort) sufficiently accurate? Here's a frightening one: If you go the copy edit route, are you obliged to do it for all comments? What if you only cleaned up the positive reviews, leaving the negative ones verbatim, hoping the typos and poor grammar will dissuade other shoppers from being discouraged from buying?

The psychological approach-avoidance relationship E-tailers have had with user comments/reviews is legendary, whether it was Etsy toying with anonymity rules, Guitar Center paying for comments, Topix trying to charge for dealing with some comments or even Costco accepting them at all. Of course, attempts to game the comment system are nothing new, with Belkin probably the most blatant case. One marketer even tried gaming our comments system, until an IP address exposed the scheme.

The editing approach suggested by NYU is different, though. It doesn't get into changing the essence of what the user said—and it certainly does not promote the idea of publishing only favorable reviews, which often has the opposite impact because it sharply dilutes the credibility of the favorable reviews. No, this approach can be seen as merely doing the commenter and other customers a favor, in that almost everyone likes having their comments made to look more intelligent and other shoppers appreciate the easier to read clean copy. The added revenue is merely a side note. Sort of.

"A well-written review tends to inspire confidence about the product, even if the review is negative. Typically, such reviews are perceived as objective and thorough. If we have a high-quality negative review, this may serve as a guarantee that the negative aspects of the product are not that bad after all," wrote Panagiotis Ipeirotis, an associate professor at NYU's Stern School of Business. "For example, a negative review, such as 'horrible battery life. In my tests, battery lasts barely longer than 24 hours,' may be perceived as positive by other customers who consider a 24-hour battery life to be more than sufficient."

The academic posted the next logical step, based on some university research projects. "We noticed that demand for a hotel increases if the reviews on TripAdvisor and Travelocity are well-written, without spelling errors. This holds no matter if the review is positive or negative. We observed similar trends for products sold and reviewed on"

Regarding the Zappos example, Ipeirotis tried to make a standard P&L analysis. "Given that Zappos spent at least 10 cents per review, and that they examined approximately 5 million reviews, this is an expense of a few hundred thousand dollars. My archive on MTurk-Tracker kind of confirms these numbers. So, the expected revenue improvement should have been at least a few million dollars for this exercise to make sense. Ethical? Notice that they are not fixing the polarity or the content of the reviews. They just change the language to be correct and error-free. I can see the counter-argument that the writing style allows us to judge if the review is serious or not. So, artificially improving the writing style may be considered as interference with the perceived objectivity of the user-generated reviews. I still consider it fine to change the grammar, from the ethics point of view."

The dangerous part is setting limits. The dangerous part is setting limits. Is it limited to spelling? How about word choice? To properly fix word choice, the fixer needs to understand the writer's intent. That's a lot to ask of software, given functionality limits, and it's also a lot to ask of marketers, given ethical and morality limits. (Did I say that out loud?)

Some of this speaks to consistency issues, which is where Legal often gets involved. If you set a reasonable policy and precisely apply it consistently to all comments, it's hard to get into too much trouble. The very nature of copy editing—even when it is done by software—prevents such a strict consistency. If some edits are done more aggressively with some commenters than others, you have a potential headache.

Consider: You might set a very reasonable policy against obscenity and derogatory comments that could be considered personal. Sounds fair. For example, what if someone says the operating system is really slow. Some people get emotionally passionate about operating systems, and they might take offense at such a comment and consider it personally derogatory. What if the reader is from another country where obscenity definitions are very different from those that apply to the country of the commenter?

Such "Is this really worth it?" debates are why so many firms have a blanket rule that no changes are made and that all comments are immediately published. That policy may be far more dangerous, but Legal is happier because it's consistent.

It reminds me of the old host liability rulings, where party hosts have different liabilities if their guests get drunk. Some argue that paying a bartender is an act of responsibility, which should give the host more protection. Other courts ruled the opposite, determining that a professional bartender raises the standard and that the bartender should have detected the excessive drinking and stopped it. That's why some attorneys argue that hosts should have unattended open bars, so they can legitimately claim ignorance as to how much guests drink and, therefore, not prevent them from driving drunk.

Let's get back to word choice. Let's say a commenter said the product was "magnifilous." Did the commenter intend to say magnificent or marvelous? Assuming the commenter can't be reached (as an operational matter, that's not even a little practical), does a marketer choose the word that she/he wants to use?

Will posters object and post everywhere that retailer 1234 changed their reviews? Hello, backlash. Even more insidious: Could posters complain that their comments were not fixed? That's a hard complaint to make: "Hey! You told the world exactly what I told you to tell them. I'll see you in court."

Then again, if those posters could prove that you have an inconsistent policy and that you only fix favorable reviews, it might not be such a stretch.