How Trustpilot Extorts Businesses
medium.com
medium.com
I don’t think this is an inherent flaw in the idea of providing trustworthy third party reviews. I think the flaw is in the specific way that Trustpilot as a business operates. (Ie. Badly)
“Nice reputation you’ve got there. Would be a shame if something were to happen to it…”
Just for the sake of argument, how would one go about creating a trustworthy third-party review business? Yelp has been accused of similar tactics for years with little pushback. I just don't think it's possible to create an ethical review platform on the internet. Way too much can go wrong.
Either the reviewers are paying you, the businesses being reviewed are paying you, or you're advertising. Advertising is a crapshoot, content creators won't pay to make sure their review gets posted, so that leaves the businesses being reviewed. And if your goal, as the owner of FooBar Business Reviews, is to maximize your profits, these kinds of techniques are going to start looking interesting really quick.
I think the problem is that if you make it into a business it messes up all the incentives, but if you don't make it a business there isn't an incentive to put time and effort in to combat spam, lawsuits, and so on. Maybe a big open source undertaking, like Wikipedia but with reviews?
I'm genuinely surprised that Facebook hasn't picked up the idea of allowing user-submitted reviews of products, services, places, organizations, etc. Then you could search your friends, or friends-of-friends, for reviews. For Facebook this would be a wealth of information because it indicates which products you actually own and whether you like them. How is that not a huge wealth of information for an advertiser?
It'd be extra good if we could take Facebook out of the equation, but the social graph there is probably the largest currently in existence.
It wouldn't be good for an ad platform to show an ad for something and then show the prospective customer the debunking of said ad.
Facebook makes their money on the market being less efficient. If the market was super efficient and everyone bought what they wanted and was satisfied with it there would be a lot less need for advertising.
My desire for web-of-trust reviews to exist made it harder for me to think as a ruthlessly profit driven company thinks.
Wikipedia is certainly not immune from manipulation. A friend of mine worked for a firm that edited Wikipedia pages for a fee and typically catered towards the ultra-wealthy and politicians.
Spam is automatically taken care for you too (if you ignore the non-review posts).
I think this is the Angie's List model but I don't know how that is working.
I am concerned about the EBSCO model where business are ranked and then pay to be advertised as top offerings. It seems like it might not necessarily choose good businesses. Often their research is based on some kind of survey.
User-generated content for free leads to the Yelp/RipOffReport (and TrustPilot?) model.
Not sure if it's comparable but in the UK we have "Which?" magazine which provides reader-funded "supposedly impartial" product/business reviews. (It started as a printed magazine but is now largely an online venture I think).
I haven't done much due-diligence on them but they've been going for a long time (edit: wow, since 1957 it seems!) and traditionally have had a decent reputation in the UK.
I've personally paid for their print magazine and more recently online subscription for brief periods when wanting some analysis on a couple of important purchases, and found the information to be useful.
An example: https://terrylove.com/forums/index.php?threads/can-we-believ...
Nope, this will easily be gamed.
I don't think there is a financial model on the planet that would make reviews 100% trustworthy.
That's why I see only Google Reviews to be in the best position in this category. It is not perfect but way way better in terms of incentives etc. compared to Yelp and its ilk.
It's primary value for Google is the huge moat that it provides against Apple maps where Apple is dependent on the crappy Yelp reviews. I say crappy in the sense of terrible incentives as mentioned in the OP I am replying to. This poisons the whole review system of Yelp and Apple maps by association.
I definitely agree that it's a hard problem, but I would be wary about saying it's impossible. This shouldn't be rocket surgery!
> And if your goal, as the owner of FooBar Business Reviews, is to maximize your profits
What is that wasn't your goal? Perhaps such a site could be a non-profit...
The internet has now existed for 3 decades, it's become massively commercialized over the last decade and a half, and this problem still hasn't been figured out without resorting to the mob tactics described in the article. If it's not rocket surgery, why haven't we figured it out yet?
People still want to see the internet as this limitless utopia without boundaries or faults. If we're gonna be wary about throwing around the word "impossible", then we need to start seeing some concrete solutions to these very real, widespread problems. This is basically a modified version of the content moderation problem. You can either include all voices on an online platform and host propaganda and disinformation, or you can try to limit disinformation by removing some users from your platform. You can't do both.
> Perhaps such a site could be a non-profit...
Being non-profit doesn't automatically solve the problem. Wikipedia is a non-profit and its subjected to manipulation all the time. As long as there are people out willing to pay money to have a certain reputation online, there will be no such thing as a truly objective review forum.
You could do an open non-profit structure where all reviews can be analyzed, including "removed" reviews (have a second data feed that is raw, with the appropriate disclaimers").
clients would in some way have to be the people getting reviews of things. It would be more of an agent type tool that would go out and find things that fit the particular clients' wants. It does not admittedly seem like something that could work without a number of other things being added to it to make people want to pay for reviews.
I don't know, have we figured out a way to have trustworthy credit rating agencies, or trustworthy journalism of any other kind? The incentives for providing information lie in manipulating that information for profitable effects. That's why nominally informative media must be schizophrenically supported by being paired with the open manipulations we call advertising.
If you want a trustworthy review system you need something more like Consumer Reports--the user pays a subscription, the product company is not involved. I doubt anyone can get the inertia to make that work.
A pernicious business model that lives in the cracks between free speech, advertising, and honesty.
Trustpilot optimizes their SEO to rank for search terms that include the "negative review" phrase.
Once Trustpilot has your profile up on their site, they will quickly start ranking for the "negative" phrase. You will be tempted to squash down those negative reviews by redirecting your positive reviews there, which will then also give them higher ranking for all of your review-related phrases - including the negative one. At this point you're locked in - without your intervention, your average rating on Trustpilot will always be lower than on any other service, and in order to counterbalance that, you will be helping Trustpilot's SEO more than any other service. Game set match.
What to do? I would suggest four main principles:
1. They will eventually get your profile, you cannot escape that. But what you can do is create a profile yourself that will be crippled in SEO (eg: wrong TLD, or slight mispelling, perhaps an extra word, etc). You might have to take several swings at this, but the goal would be to come up with a name for your company that i) ranks for your correctly spelled name in their internal search, but ii) is not competitive in the global SEO race with other companies that have your correct spelling.
2. Pick a different review service that's more honest and will work with you on a good faith basis. I found the BBB to be the most business-friendly of all of them (a bit surprising, I know). They just simply don't operate as a VC-run business that has to grow fast at all costs, and as a result, they won't turn on your like other services do (eg: it's in the services' interest to keep the content on the site, and the BBB will take things down more aggressively than any other service I know). Once you pick your canonical 3rd party review service, make them the #1 in SEO for your name by ensuring that they always have an order of magnitude more content than any other service.
3. You might be pressured to send content to Trustpilot to make up for the bad reviews. Do as little of that as possible, and each time you do that, send 10x more content to your chosen service.
4. Some people (usually those meaning to inflict the most harm) will always choose to leave their 1-star reviews on an external site so that they cannot be taken down. But you can definitely lower that volume by making it exceptionally easy for people to leave a review on your own site. That approach is far more favorable because you have a lot more options to turn the sentiment of the reviewer from negative to positive (assuming you're willing to do some custom coding). You'll still need a 3rd party review service (eg: Yotpo) so that your ratings can show up in Google Shopping etc, but you can collect those by sending all your purchasers that standard "review your purchase" email. If they leave a bad review using that service, you'll have the full context and will be able to react accordingly. Apart from that email, make sure that people can leave a review on your site without having to verify their purchase. In my experience, a substantial portion of customers doesn't realize why unverified reviews are discouraged and often not even available, which drives them to Trustpilot. So regardless if you're using Yotpo or some other service (most of which don't allow unverified reviews), make sure that a customer can click on a "review this product" button right on your site. I'll let you figure out 1) how to make this happen, and 2) what to do if in that case someone selects a 1-star review (hint: put your best foot forward right then and there).
Left a pretty bad taste tbh, trust pilots only suggestion was to use their email setup where you CC them into customer comms and then later they email the customer themselves to ask for the review. Just seemed like a way to tie us in deeper with them.
That and their insistence to use a JS-based widget instead of just plainly copying review text smells like spyware. If it was my company I'd give them the middle finger and a C&D for using my brand's assets (logo, etc) on their site.
I thought their prices were crazy initially, but the amount of times we hear "oh I saw your good reviews on trust pilot so we decided to go with you". Just got to suck it up and pay the toll.
At this point, TrustPilots business model is essentially reputation laundering.
If you pay them enough, they will use their mostly neutral reputation on the consumer side to launder your potentially shitty reputation to a more positive one. It's not like the average consumer knows any better and I would not be that surprised that having a positive score on TrustPilot is not the decision maker for a customer, but having a very bad one - is a disqualifying factor. Hence the reputation laundering scheme...
[1] https://en.wikipedia.org/wiki/Stiftung_Warentest
[2] https://www.heise.de/newsticker/meldung/Verbraucherschutzmin...
If the companies are the ones footing the bill, I see no way to have a fair system where a few richer players couldn't exploit the system or the review company itself charge companies as much as they can just for leaving them alone and not rigging the reviews against them.
> whether we liked it or not, we were now listed, and anybody could leave a public review that we have no control over
Why should a business have control over public reviews? Unless, that is they are libellous or otherwise in breach of some law, in which case the business has a fair case against the publisher.
However...
I feel from personal experience that TrustPilot is a huge scam, having had my genuine review suppressed by one of their paying customers. It's a huge loophole - dispute the review to get it off the front page, by the time it's reinstated, it's further down and they've had a chance to harass the customer to get it changed (incidentally, when companies do that I tend to drop another star if possible and edit the review saying so.)
It's also a privacy nightmare for anyone shopping online (at least here in the UK) if you use different email addresses for each company, because you have to opt out every single time. And if you don't leave a review, the send a reminder (as though it's somehow my job to leave reviews).
I ended up blocking their domain trustpilotmail.com in my email service, and if I see the Trustpilot logo on a company website, I will look for an alternative provider.
Not what I'd call 'better' and possibly not even 'business' (and to the extent that bureau denotes a government agency, definitely nope). So maybe there's nothing new under the sun here?
I would even go so far as to call this an inherent problem of trust and scale. As soon as the world's too large to experience things like business personally or with your own circle of trusted people, it's desirable to make some way to scale that, and it's inherently a cursed problem that always becomes abusive. (or possibly a cursed problem that more often than not becomes abusive)
Consumers continually say they want and use them, trusting some type of collective wisdom of folks willing to take time to give a review, assuming some honesty and relevance (reviewer has same needs, experience, and use pattern as I do for the item).
Yet, like price engines before them, consumer demand alone doesn't make success (some might also say that reviews are a feature, not a product, but table that for now).
From Zagat to Angie's list, the BBB in the US, TripAdvisor, yelp, 4square, plus a zillion others (along with on-site reviews powered by Bazaarvoice and others)... So many have come and gone. Amazon reviews are gamed, unclear what drives "Google Reviews", blog reviews have faded...
Did social kill them? Maybe, but the few people who respond to a "Hey, is X any good" on any forum or group seems less helpful than these past offerings.
I've wondered about these "come and gone" tools for a while, since price engines and reviews to leverage collective knowledge were hallmarks of internet 1.0 lore, and it seems like we failed at really delivering on those hopes. Still difficult to find the lowest price on items, and still difficult to get a clear read on product quality or value... And restaurants and other service offerings are still harder.
Any thoughts on what commerce model would work for reviews to be unbiased but collective, more than just the Consumer Reports model of few experts and some surveys?
IDing users wouldn't be too hard to implement. There are plenty of businesses that provide instant ID verification services, many of the new online banks and money transfer apps use these services. The two issues would be cost and the fact that the number of people using review sites that do this would be pretty low.
My current thinking is that you need the reviewers to have their reputation on the line somehow when they review something. There are quite a few options to do this algorithmically, happy to chat if it's something you are interested in too.
Even with all the gamed reviews on sites like this and Amazon, if you spend the time you can usually figure out what's up, but it takes time.
Often it's garbage (this won't do X--hey, if you had paid attention you would know they never claimed it would do X!) or irrelevant (it's not the fault of the product if it was late, damaged in shipping etc) and sometimes it's just a tread-with-caution (I've seen a lot of that on Amazon--customers burned by combined-inventory or bogus stuff). If it's legitimate gripes you've found something to avoid.
So unless the review cater to me personally by covering those points, I usually prefer a more detailed review (or a large number of reviews/discussions by Googling in the hope that someone covers those points).
> If Trustpilot does not own the reviews, how can grant (and more importantly deny) how you use those reviews?
is readily answered by their terms:
"You hereby grant us the worldwide, non-exclusive, perpetual, irrevocable, royalty-free right and license to publish, display, reproduce, modify, create derivative works of and commercially exploit any material, information, notifications, reviews, articles or other types of communication (hereinafter referred to as the "User-Generated Content" or "UGC") which you create on the Website as a Registered User. We may freely use and transfer the UGC and disclose the UGC to third parties."
It seems like the important word in the legalese you posted above is "non-exclusive."
Say a user wrote a review on TrustPilot, and copy/pasted it to an email to the business. Then the business posted the review from the email on their website. Are they now violating TrustPilot's TOS?
You might want to get a release before putting it up, and being right doesn't mean Trustpilot knows you're right. Might make that blue box defamation, but that'd be a long, uphill battle against a well-funded opponent.
As soon as word that negative reviews are removed you're dead as a review site.
The flip side is that I suspect that people make the effort to leave a review when they are unhappy so reviews are likely heavily skewed negatively unless the company joins the site and actively requests verified reviews from its customers.
I don't think people realize or care.
When you're using Twitter/Facebook/Instagram/YouTube/TikTok/etc., you're opening it many times a day and seeing dozens and maybe hundreds of ads. When you use Yelp, you're going to the site with a specific intent and trying to leave very quickly. You visit Yelp once a week or once a month and you try and use it as little as possible.
No one is doom-scrolling a review site and seeing ad after ad after ad. They're going to the review site, finding something, and leaving fast. So they simply don't get the kind of traffic that you can monetize into a giant business. But they have something that is valuable to businesses: their reputation. So you try to monetize that, but that's difficult. You don't want to let businesses pay to remove negative reviews so what are you selling them? Also, what lies might your salespeople be telling businesses that you're selling them? Are commission-driven salespeople telling businesses that paying will get bad reviews removed?
Angie's List worked off a different model where people paid to get access. That meant that they didn't have to look to businesses for payments and the reviews came from someone "real" enough to have a credit/debit card and have paid them money. However, it looks like they've pivoted their business model and it's no longer a paid service (and now called Angi).
You'd need something like a paid service that's backed by a guarantee (some credit cards basically offer this for the high-end cards; concierge service that will recommend a restaurant (that takes the card, of course) and if you complain to them afterwards they'll make it right or refund the cost).
But for most things people don't even bother - everyone knows gyms are scammy with cancellations and people still sign up, for example.
Consumers are generally not primed to pay for reviews and when they dont pay they become the product.
There's a company I used to be customer of, with a free tier but who act in an underhand manner to their paying customers and for quite a while they've been getting bad reviews from people warning about their dodgy practices. Free-tier users are occasional users in the main because if you use it a lot you're pretty quickly getting 'commercial use' detected issues. However in the last six months they'be been guilting those free-tier users into giving glowing reviews and their positive review count has soared while the negative reviews from paying customers has been drowned out. So in this case, Trustpilot is also useless for 'real' customers.
Did you read the article? The reviews of Trustpilot are mentioned.
The usual common sense concept is that if there was even the suspicion that a business like Trustpilot would abuse their power, their reputation would be so tarnished that they would not only not do, but absolutely make sure it was almost impossible for the perception of abuse to happen.
But instead, it looks like nothing happens when they flagrantly breach trust, so it looks like in practice you can't really trust any company at all.
I remember back in the days when they'd call us up and say they could fix us a nice score if we subscribed
[0] https://support.cloudflare.com/hc/en-us/articles/200170086-U...
Get Satisfaction, Or Else...
- article: https://signalvnoise.com/posts/1650-get-satisfaction-or-else
- discussion: https://news.ycombinator.com/item?id=540540
Follow-up on "Get Satisfaction, Or Else..."
- article: https://signalvnoise.com/posts/1661-follow-up-on-get-satisfa...
- discussion: https://news.ycombinator.com/item?id=543431
If you really want it to be efficient display a static "cached" image of the rating and swap it for the iframe once the image is loaded and displayed. This ensures fast response time and accuracy of the fresh rating at the same time.