Trustpilot in the regulatory crosshairs

The AGCM - the Italian equivalent of the UK's Competition & Markets Authority - has fined Trustpilot EU 4 million as '[the] Online review platform provided consumer rating information not always representative of customers’ actual experiences.'

The ACGM's press statement is here. An extremely thorough English translation of the detailed findings of the Italian investigation is to be found here. 

In this post, we have taken extracts from both of these and made our own comments where necessary. 

The above is a summary press statement by the AGCM.

What follows is a translation and summary of the key findings by a European law firm.

Our comment

Overall

We continue to ask the question of any business subscribing to Trustpilot - or any other online review website - 'Why?'

• 'Why pay [a review site] when Google reviews are free?'

• 'Why pay [a review site] when Google reviews have far higher visibility?'

• 'Why pay [a review site] when Google reviews have greater credibility?'

We contend, as we always have, that if Google reviews had predated the online review sites, the latter would never have come into existence. 

Not covered by the above

Trustpilot's use of a 'Bayesian average' to calculate a business's 'Trustscore'. Here is what Google Gemini has to say on the topic...

The issue here can be broken down into 'Simple' and 'More complex':

Simple

Work this out if you can. We spent hours wondering how a business with 77 per cent of its reviews scoring it four or five stars could possibly result in a 'Trustscore' of 1.6, and couldn't come close to finding an answer

Most - all? - consumers will expect a score to be a mathematical average (as is the case with Google reviews and just about every other mechanism we have ever come across, including our own): the business has ten reviews scoring 5.0 = an overall score of 5.0. Not with Trustpilot, partly because of the seeding referred to above. All businesses start with a score of 3.5. The kind of score that will have most consumers moving swiftly on (to the next business). 

More complex

Click to read. Does anyone else think this sounds like 'Join and look great'?

Add to the above the fact that Trustpilot also applies a (positive) weighting to more recent reviews, and the water is muddied even further. This obviously benefits larger businesses and those that actively invite reviews from customers, which may just mean 'paying business members' (the ACGM evidently thinks so).

Overall

Some might say that all of the above work in favour of paying subscribers and against businesses that haven't joined Trustpilot. Certainly, Grizzly Research's examples, comparing similar businesses, some members and some not, give the distinct impression that 'join Trustpilot and look great, don't join and look terrible.'

The key here is two 't' words: 'Trust' and 'Transparency'. Reviews are relied on by the overwhelming majority of consumers these days to guide them toward reliable businesses. If businesses are offered an opportunity to tip the scales in their favour and that results in consumers being misled into using a business they might not otherwise use, it erodes trust. 

Tell us that a review such as this is not gold-dust? For all concerned: the business and those researching the business

We often hear people saying 'I don't trust reviews' or 'I wouldn't use reviews', and that is probably acceptable in the context of online shopping, but when it comes to choosing a medical, financial or legal adviser, it would be a great pity if the vital resource that online reviews can be was unable to be trusted.