For years, Fakespot has been my go-to tool for screening online reviews on shopping sites for authenticity. You are likely aware that Amazon, Walmart, and other e-commerce sites are packed with fake reviews and shoddy products. It can be hard to tell what is real. I’ve relied on its AI analysis for ratings of both products and vendors when purchasing online. In fact, I have written about it before and taught it in my online shopping class.
Unfortunately, Fakespot was discontinued in all formats on July 1. Based on Reddit threads where Fakespot fans lament its retirement and beg for a subscription version, the service was very popular. Conspiracy theories abound regarding why Mozilla would retire such a popular service. Regardless of the reason, it’s time to find an alternative for spotting fake reviews.
Review of Free Tools
If you search for “Fakespot alternative,” you will find lots of tools that say they can do what Fakespot does. Many of them are either paid tools or are geared toward businesses that monitor reviews about their own products.
The same free tools were suggested on multiple sites. Disappointingly, many are shut down despite those articles having recent publication dates. Here are the tools I looked into. The starred tools are the ones I report on below.
- ReviewMeta – The site still exists, but it couldn’t find any of the links I pasted into the search. Anecdotal reports online suggest this site stopped including new content.
- The Review Index – every search was met with a server error on the website, and the chrome extension is no longer supported
- Wisely – provides a summary, but did not judge the integrity of reviews for the product pages I tested
- Vetted AI – can’t copy/paste a link. But, if you type in the product name or genre, it looks for reviews around the web
- *Null Fake
- *RateBud
- *ChatGPT
Testing Method and Results
My method to test the tools was simple. I found what seemed to be an obvious scam on Amazon and ran it through Fakespot. I plugged the same link into the other sites I tested (except Vetted AI – see above) and compared the results.
The original item was a tricked-out smartwatch selling for under $20. Here are the grades given to the item and the seller by Fakespot:
These were exactly the results I expected.
Here is the same item rated by other sites:
Null Fake
Unbelievable. Literally. So we’ll stay away from Null Fake until the results are more useful. This is a new open source tool, so it may take time to catch up.
RateBud
This is the kind of result I was hoping for! It is very close to Fakespot’s rating.
There is another red flag here that we may only notice when running the same product through different tools over multiple days. One of the other screenshots said there were 30 reviews, while Ratebud stated there were 4 reviews (the next day). The Ratebud report notes that all 4 reviews came on the same day. The product was exactly the same, so we can reasonably assume that the first listing started to receive negative reviews, leading them to delete that product and start a new page for it when they could preload positive reviews.
ChatGPT
My results varied when using ChatGPT. In my initial phase of research, I used the prompt “analyze the reviews on this page.” After it gave results, I entered the prompt, “Are these reviews genuine?” The response it gave was very promising!
However, when I tried this again on the product I evaluated with other tools, the answer included instructions on how to spot a fake review and suggested I use Fakespot to analyze the page. Perhaps helpful, but not what I was looking for.
In my limited testing, Ratebud was the closest to Fakespot.
DIY Vetting
If you’d rather evaluate the reviews yourself, here are some red flags that may indicate shady dealings.
The Review Text
- Very brief one or five-star reviews without context or overwhelmingly positive (or negative) language. We normally think of fake reviews as intended to inflate the overall rating, but some companies may use negative reviews to decrease trust in a competitor’s product.
- Reviews that use poor grammar or unnatural-sounding language, such as an overly formal tone
- Generic word choice and vague or basic descriptions. When real people take time to write a review, they tend ot be more specific about their opinions.
- Multiple reviews using identical language
- Overuse of personal pronouns (me or I)
- Any review that points the reader to a different product
The Reviewer
You can also learn a lot about a review by inspecting the reviewer.
- Is the name of the reviewer either overly generic or just a string of letters and numbers? Probably not real. The use of a middle initial is a dead giveaway for a fake name
- Click on the reviewer’s profile to check their review history
- How many reviews have they written? Beware the single review
- Do all of their reviews use similar phrases?
- Are they all one or five-star reviews?
- Have they reviewed multiple items for the same seller?
- Does the user have a profile picture? If not, or if it looks like a stock photo of a very good-looking person, use caution
Other Clues
Beyond the review text and reviewer, here are some other warning signs:
- Check the time stamps to see if the listing has many reviews posted on the same day
- If it has very few reviews compared to a similar product
- Overly staged/stock photos or overproduced video indicate the review may have been paid for
- Look for the Verified Purchase label. If it isn’t there, assume it’s safe to disregard the review
To limit review results to verified purchases, click on the stars at the top of the listing, then scroll to and click “see more reviews.” The next screen will have this dropdown field with a filter:
AI Detection Tools
Another way to check an individual review to verify if it was created by AI is to use an AI detection tool. There are many available for free (or freemium), but results vary. Much like when using AI to generate content, these tools can get it wrong while seeming very convincing. To learn more about AI detection tools, check out this article by PC Mag.
Fake Reviews Are Illegal
It may be hard to believe since they are everywhere, but fake reviews are officially illegal in the US. According to last year’s FTC ruling, the following deceptive practices are banned in online stores and on social media sites:
- Buying reviews
- Reviews from people who do not exist
- Reviews written by company insiders (when not clearly disclosed as such)
- Suppression of negative reviews
- Fake social media indicators (likes and comments from bots or people who do not exist)
- Review sites that claim to be independent, but are affiliated in some way with the items being reviewed
Reporting Fake Reviews
Amazon has an internal reporting system for fake reviews. To report a review, scroll to the end and click “report.”
Choose “fake” and mark it as inauthentic. Then submit. This flags the review for investigation and possible removal.
You can also file a report directly with the FTC. This FTC page contains a link to their reporting site, as well as instructions on how to report fake reviews on other social and review platforms.
Summary
The internet is filled with fake reviews and testimonials. With the Fakespot tool retired, how are you to know whether the reviews you see on e-commerce websites are authentic and can be trusted when you’re making decisions about products to buy? Try the tools and suggestions above to vet online reviews and report fakes.
