Skip to: Main Content / Navigation

  • Facebook
  • Twitter
  • LinkedIn
  • Add This

Consumers take online reviews with a grain of salt

Article ID:
October 2013
David Ensing

Article Abstract

Online ratings sites have become an important part of the customer experience but should you believe what you read? The Maritz Research 2013 Online Customer Review Study surveyed 3,404 people about the use of 13 dedicated customer review and rating sites and found that most consumers are cautiously trusting.

Editor's note: David Ensing is vice president, voice of the customer, at Maritz Research, St. Louis. He can be reached at This article appeared in the October 28, 2013, edition of Quirk's e-newsletter.


Now more than ever, online ratings sites are an important part of consumers' purchase decisions. How many times have you searched for a review about a product you were thinking of buying, a hotel where you might be staying or a restaurant where you could be eating?


As the importance of positive online ratings grows, so has the incentive for businesses to make sure they are portrayed well at these sites. Most businesses do this properly - by improving the customer experience. But some businesses seem to take the late W.C. Fields' approach: If a thing is worth having, it's worth cheating for.


Indeed, recently we've seen a wave of stories about companies faking online reviews.


  • In September, the New York Attorney General settled a lawsuit with 17 companies that offered to post fake reviews for a fictitious yogurt store the AG's office had set up.
  • sued a social media reputation management firm for attempting to post fake reviews on Edmunds for its auto dealership customers.
  • Yelp sued a law firm for posting fake reviews.
  • Yelp sued the parent company of the Web sites and
  • Google removed large quantities of dealership surveys from its site because they were suspected of being fakes.

Given these developments, we at Maritz wanted to find out how consumers view information at rating sites. Do they trust it? How do they use it?


To answer those questions, we conducted an online panel study in April 2013 asking 3,404 people about their use of dedicated customer review and rating sites. Generally, we found four things:


  1. Of those who use rating sites, about 25 percent think the information presented is biased or unfair.
  2. Even those who think the information is fair use techniques to separate false or biased information from good information.
  3. Customer ratings of how much they trusted the information varied significantly from site to site.
  4. Many more people read reviews at rating sites rather than post reviews and there are significant differences between readers and posters.

Integrity of information


We asked rating-site users to rate the integrity of information at rating sites. Nearly 25 percent thought the information presented is unfair, with 16 percent saying it is overly negative and 9 percent saying it is overly positive. The remaining 75 percent thought the information presented on rating sites is generally fair. Interestingly though, even these respondents pointed out that they need to separate trustworthy reviews from non-trustworthy ones - usually due to concerns that some reviews are naturally biased or fake. For example:


"I can usually tell when a review is fake - if it is overly positive. I usually try to read a range of reviews. If there are not enough reviews, I don't usually take it seriously. I don't think Web sites can prevent all companies from posting fake reviews but by reading the various reviews, I can usually reach an accurate conclusion."


"I'm not sure how objective it is. I think that people tend to only share either the good or the bad, which usually does not give you a real balanced opinion."


"I think the reviews are for the most part honest but I believe that the site is selective in the reviews it posts."


"Some are credible and some are biased or questionable."


"Generally reliable if one disregards the top 5 percent and the bottom 5-10 percent of comments."


"Take everything with a little skepticism but if enough people say it then it's more believable."


"I believe that a lot of the information is put there by people who work for the companies to show positive information."


How much they trust


We also asked respondents to tell us how much they trust information at 13 high-profile rating sites using a five-point rating scale: Trust None of it, Trust Some of It, Trust About Half of It, Trust Most of It and Trust All of It.


Percentages of respondents who said they trust most or all of the information at the various sites ranged from 36 percent to 59 percent. Larger and more established sites seemed to be perceived as more trustworthy than newer and smaller sites but even at these sites, more than one-third of visitors are skeptical about much of the information. Figure 1 shows what percentage of those who read reviews trust most or all of the information.


Only read


We asked people about their reading and posting activity at each of 13 rating and review sites. Many more people only read reviews rather than write them (Figure 2).

Interestingly, sites seem to vary quite a bit on the ratio of readers to posters.


  • Sites like Yelp, TripAdvisor and OpenTable have relatively low ratios of about three readers for every poster.
  • Other sites like Edmunds, Priceline and Urbanspoon have higher ratios of between six and 10 readers for every poster. These differences are probably due to site characteristics.
    • Edmunds is primarily a professional review site where customers can also leave their own reviews. Many visitors probably go there to read professional reviews.
    • Priceline is not primarily in the review-generating business. Reviews are secondary to the site and people need to have booked through Priceline to leave a review.
  • TripAdvisor and Yelp are first and foremost review sites. Thus, anyone can leave a review there and both companies take countermeasures to try to identify reviews left by people who did not actually experience the product or service they are rating.

Finally, we looked at the demographics of people who posted a review on at least one site versus those who only read reviews. We found that, compared to readers, posters tended to be younger and more affluent. As many researchers know from experience, these characteristics (especially being younger) correspond with being harder graders on surveys of customer experience. Thus, it would not be surprising if information left by real customers at rating sites is overly negative.


Not a crisis


The upshot seems to be that a significant minority of visitors feel rating-site information is unfair. However, those who believe it generally is fair still take precautions to weed out false information from good information - and even information left by real customers may be biased due to self-selection when choosing to post. While there is not a crisis of confidence yet, review sites must do more to improve the quality of information offered.

Comment on this article

comments powered by Disqus

Related Glossary Terms

Search for more...

Related Events

May 4-6, 2015
RIVA Training Institute will hold a course, themed 'Advanced Moderating,' on May 4-6 in Rockville, Md.
April 7 at 11 a.m. CDT, 2015
20/20 Research Inc. will host a complimentary one-hour Webinar, titled 'The Quant+Qual Paradigm: 3 Integrative Strategies for Today' on April 7 at 11 a.m. CDT.

View more Related Events...

Related Articles

There are 1129 articles in our archive related to this topic. Below are 5 selected at random and available to all users of the site.

Watch out Apple: Nintendo Wii rates highest among America's youth
According to data from Harris Interactive, Apple-branded products dominate the personal-device arena among America's youth but Nintendo Wii rates highest overall in brand equity.
Data Use: Adaptive choice is a good choice
Using an example of a transportation company client, the author outlines why adaptive choice-based conjoint analysis is a useful tool for developing market segmentations.
Using morphological content analysis to mine insights from qualitative interviews
The author details the business-to-business application of morphological content analysis, which provides a systematic approach to analyzing interview transcripts. An example in the consumer electronics industry is provided to illustrate the main concepts.
Consumers give Mannington a winning formula for new vinyl flooring product
Mannington Resilient Floors used several research strategies to develop its successful flooring product. In addition to a telephone survey, Mannington used HTI Custom Research's monthly mail omnibus study to determine some basic purchase dynamics and the demographics of purchasers of floor coverings. Consumers were asked about their level of satisfaction with different kinds of floor coverings and what the coverings’ strong and weak points were. Mannington also studied retailers and others in the trade to gauge perceptions compared to its competitors and to determine how to increase and improve its industry profile.
Pay to play: Two-thirds of Web users shell out for content
Research from the Pew Internet and American Life Project indicates that most consumers are not opposed to paying for online content. Nearly two-thirds of Internet users have paid to download or access some kind of online content, namely music, software and apps.

See more articles on this topic

Related Suppliers: Research Companies from the SourceBook

Click on a category below to see firms that specialize in the following areas of research and/or industries



Conduct a detailed search of the entire Researcher SourceBook directory

Related Discussion Topics

Online Focus Group sessions
10/28/2013 by Sally Hooper
Market research report
08/20/2013 by Aarkstore Store
SSI - Panel Quality
07/19/2012 by Paul Ponsford
Indexing GAP
07/26/2011 by Felipe B. Fehse
E-mail reminders
04/03/2011 by Ruben Alcaraz

View More