[erentz]

My experience out of four AirBnB stays was that two were as advertised and two were complete dirty slums, nothing like the photos or reviews suggested. We had to leave and get a hotel. The reviews were all 5 star raving about how beautiful, amazing, etc they were with 40-50 reviews each.

Has anyone else had this experience? It made us unable to trust the review system because in this case it was so clearly rigged somehow. And without being able to trust the review system AirBnB is next to useless.

[jressey]

I don't think the reviews are rigged per se. Here's what I think causes this. I was a superhost for about a year renting out a spare room.

Superhost status seems to increase your position in search results, and acts as an indication of high-performance/great accommodations leading to more renters being interested and you charging higher prices charged. Most hosts want to achieve this rating. Part of superhost status is maintaining an insanely good rating, I think > 4.5 avg for rolling 3 month periods. Renters seem to know this, and in fear of being somebody that gives low reviews, will inflate their reviews. I could be wrong, but I think my wife and I would review accounts and lease to renters who had given more positive reviews.

This may seem selfish but it only takes 1 3-star review from middle-aged people comparing you to a luxury hotel to learn.

[pavel_lishin]

Are you saying that I, as someone who needs a place to stay, am more likely to give a 5-star rating to a slum so that if I ever apply to stay with a superhost, they'll look at my rating history and be more likely to approve me?

[mattpavelle]

"my wife and I would review accounts and lease to renters who had given more positive reviews"

That is exactly what he/she is saying. I'm not actually surprised by this.

5-star ratings are a poor means of actually ranking things as they are generally currently implemented - especially when there is no guidance as to what each star means.

Big issues include:

As per this post, if PartyA needs high ratings and can see all the ratings PartyB provides to other PartyA types, they will certainly favor those PartyB types who rate higher.

The "average" on most of these 1 to 5 star scales is not a 3. E.g. the average Uber driver rating is 4.7 (or something) stars. So what does that mean to the user? I need to remember 4.7 is ok but 4.8 is good? Shouldn't the app provide guidance to users? Say to the user, "if you had a good overall drive - rate this 3 stars".

Why doesn't the AirBnB app "curve" ratings as well? Meaning apply my ratings to a bell curve with the average being a 3 - i.e. if my average rating is 4.8 with a high of 5 and a low of 4.6 - why doesn't the app just translate my 4.6 a "1", my 5 a "5" and curve the rest of my ratings so my average fits on the scale? Then users would know that a "3" is average - which for AirBNB might mean "good"? And a 4.8 is truly exceptional?

And of course none of this addresses the problem where my "exceptional" does not match yours. Maybe I like I like eco-tourism and camping, and you like Ritz Carltons and gold martinis. Or maybe I like slow and safe drivers and you like fast drives who run yellow lights?

Regardless, I'm tired of all these meaningless inflations.

[otterley]

What's the rationale for allowing Airbnb hosts to see the reviews that a renter has given other rentals?

[nilsonsfj]

This is disgusting and probably illegal in parts of the world. This is like healthcare companies refusing service for the ones more likely to need more health assistance.