Travolution have asked the question - How do you know when you’ve read a fake hotel review on a travel website? (see their full post)
In response, one comment says “When you read it on an online travel agency website”.
Yikes - have user generated reviews got that bad a reputation? Have companies been so manipulative of it in the past that it no longer holds any weight?
It was only last January (7 months ago) that Nielsen//Netratings published research that said that sites posting user reviews are considered the most trustworthy (see reference)
When asked which online source do you trust to give you ‘the most reliable information of all’ user generated content sites like TripAdvisor were cited by 21% of respondents ahead of 15% each for review sites like TimeOut.com and official local information sites such as Visit Scotland.com. Travel agents’ own sites such as ThomasCook.com were considered the most reliable by 12% of Britons online compared with just 11% who selected search engines.
Personally I have always been a little wary of trusting user generated reviews as I don’t know whether it is really a customer or a hotelier / tour operator / airline in disguise. However, I write as someone who once booked a hotel in London and then checked TripAdvisor. All the reviews said to stay well clear of the hotel….. but I had booked by then…. when I got to the hotel - there was no cold water (let alone hot water!). Lesson learnt.
Do I have any ideas to help with reducing “gaming” by travel companies?
Firstly there is a difference between two categories of user data - and therefore two styles of gaming:
- Attention or use data - what the user does - such as user navigation paths on a website, items put on a shopping basket, time spent on a page etc
- Action data - such as posting a review or “buying” something
Both can be gamed - by which I mean changed or used to a individual or companies advantage. However, mostly attention or use data will only impact the experience of that individual user so the incentive to do so is minimal.
An example of gaming an ecommerce website in this way would be:
- Go to a website selling mobile phones
- Go straight to a page with an expensive phone on
- Add the expensive mobile phone to your shopping basket
- Remove the expensive mobile phone from your shopping basket
- Navigate to a web page with a less expensive phone
- Add the less expensive phone to your shopping basket
The chances are that, on the website, a special offer for the expensive phone you first added to your shopping card could be shown to you - perhaps with some money off or similar. This is because the website will have identified that you are interested in buying - but has evaluated that you are buying on price - so could be upsold to with a product you have previously expressed an interest in….. You also need to find a website powered by an expensive ecommerce platform such as Broadvision - as very few sites actually do this at the moment!
That is gaming for personal benefit - but the problem with user generated review sites is that the gaming is mainly for commercial benefit - i.e. getting good reviews of your product infront of potential customers
The real problem is one of identity. How do you know the user is who they say they are?
An answer to this is to ensure that the incentive to provide a correct identity is driven by something other than having access to the system - perhaps by an ongoing relationship where you have to provide correct information for other purposes.
Here is an example of what I mean - do you recognise the question format?

This question came up in Linkedin.com’s business social network - not the first place I would go to ask a travel question.
However, what interests me about this question is:
- I can see who asked it - it is not a travel company from Limerick “trolling” a question - in order to answer it later with a good response pushing their product - but a supply chain manager from Dell in the USA.
- I can see who has answered it - a law professor from Houston, a bank employee and assorted others
I know exactly who these people are - I don’t know them personally - but I trust that they are who they say they are - because LinkedIn has solid profiles of all users. I know they are not travel companies.
This is something that Facebook, MySpace, Bebo and all the other travel social networks DONT have.
So to summarise - user generated review sites are only as good as the identity mechanism they have. The identity mechanism doesn’t just have to convince the website that the user is who they say they are - but also must be able to convince the websites’s users….. this, without making all sorts of personal information very public, is not something that many user generated review sites will find very easy to achieve.
Now if I were Linkedin - I would be looking to utilise this solid profile information to create something bigger and better than what they already have……….. they are one of the few sites that people actually give their real information to (although it only really attracts those of working age)
Ummmm…….. something to ponder
If you want to be notified next time something is published sign up for email alerts or subscribe to the RSS feed. Thank you for reading!


Blog home



In short, how do we separate user reviews from abuser reviews? On a social site like eBay or Tripadvisor, volume helps. If a highly-rated eBay seller has only 5 reviews, I might suspect undue influence by relatives and friends. If he has 1000 reviews, the danger diminishes.
User reviews are no longer so often called peer reviews, but that’s precisely the advantage provided by a site like LinkedIn, as Alex describes — they make it more likely that I’m hearing the opinion of a person “like me.” Like me can mean many things, of course — similar design tastes, similar desired activities, similar sensitivity to price, etc. Or it can mean simply like me in the sense that the author of the review is an actual traveller and not a supplier or a PR house.
The disadvantage of the networking site as a travel advisory is that it will take longer and you’ll get fewer replies — because there is not sufficient volume (at least in that aspect of the site). And (to come full circle) if the volume does pick up, and people begin to favor network sites for travel advice, then you can bet the supplier and the PR house will begin creating fake identities for bank employees and law professors from Houston.