This is a preview build of the app. It uses a separate database to the production app.

Member Safety and Freeloaders

Design

Note: this article discusses predatory behavior, sexual assault and harassment.

It is well known that Couchsurfing™ has a problem with creeps using the platform to harass individuals. While this behavior occurs between all types of people, its most prevalent form is through some men using the platform as a way to attempt to have sex with women. While some people might be comfortable defending themselves in these types of situations, it has no doubt been a barrier to entry for numerous others, and has caused even experienced users to move away from Couchsurfing™. The behavior is so common that it even has its own term, SexSurfing. To be clear, sex is not bad for the community, but users' expectation of sex is. In recent years, there has been an influx of users signing up to Couchsurfing™ with perverse motives that directly contradict the ethos of couch surfing.

Although such problems plague any online platform where individuals may connect, the problem is particularly endemic to Couchsurfing™ due to its need to turn a profit and its chosen model of doing so. Under a model where the majority of revenue comes from new user verifications, there is a need to attract new users to the site regardless of how it affects the existing community or even whether those users become committed users. The simple fact is that the idea of free accommodation and sex sell, and this is good business for Couchsurfing™. This is one of the main harmful outcomes of the conflicting incentives playing out at the moment.

Countless people have been attracted to the site with the primary value proposition of a free place to sleep or an easy way to get laid. These people, unless corrected, are bad for the community and turn others away. Maintaining such a status quo — by not aggressively rejecting such narratives, or otherwise — will attract certain types of people to the platform, and some of them will pay for verification. Additionally, there is no profit in keeping good people around. This is extremely toxic to the community as a whole. Sex is of course not necessarily bad or unwanted. Couch surfing facilitates a large number of genuine interactions between people from different walks of life, and such interactions do sometimes lead to fully consensual sex, love, flings, or other non-platonic interactions. These experiences can be great, but it is imperative for the health of the online couch surfing community that people do not flock to the service expecting to have sex. It's unacceptable to attempt to use the platform to coerce others into sexual acts or otherwise exhibiting sexually inappropriate behavior that makes people feel uncomfortable.

It would be easy to write this off as a product of there being a lot of creepy people out there in the world, but there is a sentiment that over time, this has gotten worse. Couchsurfing™ has also done very little to limit it. They may have instituted policies that act as guidelines for people to follow, but this doesn't change the structure of how people interact with the app. As mentioned above, there is a chance that the lack of interest from Couchsurfing™ to act on the problem is because it would destroy a major source of revenue for them.

In one of the more obvious cases of creepy behavior, it is extremely common for women to be inundated with creepy requests and messages from men, which is amplified when women post public trips. There are many ways that Couchsurfing™ could approach fixing this. Automatic filters could control explicit messages. Users could have the option to alter message settings in different situations. For instance, someone could opt to not accept messages from unrated or unverified users that find them from public trips. However, these aren't trivial to implement and don't yield profit.

Possibly more worrying is the more insidious case of users who act inappropriately only in person, after agreeing to meet up. In some situations, this behaviour can be predicted, for instance, the prevalence of male accounts which only have references from attractive women. However, often it is too hard to predict given the limited information that Couchsurfing™ provides you. There are even individuals and groups that advocate for this kind of predatory behavior. There are countless stories of users who have been hit on by their hosts or surfers in a way that makes them feel uncomfortable. Users, especially men, sometimes do not recognize the power imbalance of those situations. Guests who feel uncomfortable are not able to leave as they are reliant on the host for accommodation. On the other hand, hosts who feel uncomfortable may be too reluctant to kick the surfer out as they feel an obligation to the surfer who is reliant on them. Experiences range between situations and the severity of the incident which results in a spectrum of outcomes; at one end, an individual may be comfortable handling the situation and continuing the interaction, to the other end where the situation may lead to sexual assault. Most cases of course fall somewhere in the middle. In some cases, people will just be uncomfortable and stick it out. In others, surfers will sleep with their hosts out of a sense of obligation. Although this might be technically consensual, it will likely be a bad experience for the member, make them less trusting, and turn them away from couch surfing. This kind of situation is widespread, and it is not uncommon for women to be hit on by the majority of their male hosts/surfers.

This is exacerbated by the fact that most negative experiences are murky at best, and the review system is poor for indicating either the type or severity of a negative (or even positive) experience. Members often feel disinclined to leave a negative review, out of fear of reprisal. Similarly, if the experience wasn't extremely bad, the weight of a negative review may seem unwarranted, as it might heavily impact on the other member's ability to use the platform in the future. For example, if a member found themselves slightly uncomfortable or felt that a surfer was just in it to freeload, such an experience may not warrant a full negative review. There ought to be a way for a user to leave this type of non-positive review on the platform.

Women being turned away from Couchsurfing™ has caused various negative feedback loops. Women may for safety reasons only want to surf with female hosts. Fewer women on the platform mean less female hosts, and this will lead to less female surfers. Due to the number of male hosts who only host women compounded with the super-host effect has meant that in many places the only hosts available are these predatory hosts, further limiting the options of women, and causing a side-effect where men can't find hosts.

The issue again comes back to the degrading of communities. Strong communities not only give a wider variety of hosts, but they also have accountability. When local forums existed, visitors were warned away from predatory hosts. People's actions in the community were broadcast by their community. Users expressed opinions in public, and you could stay with people based on those opinions. Their actions did not exist inside a bubble. While it may have not been the optimal way to hold people accountable, a system existed. Then it was destroyed and we were left only with a review system that does not incentivize users to leave negative reviews.

The freeloader problem, although not comparable in the weight of negative experiences for members, is similar. Many people sign up purely as a way to get free accommodation. A good member of couch surfing will surf because they want personal, local and community experiences in a non-transactional setting. Free accommodation is part of that, or at most a way to remove the monetary barrier to entry. The classic freeloader is only in it for the free accommodation and is uninterested by the other elements such as interacting with their host. Similarly, profit motives for Couchsurfing™ want them to attract this large market as freeloaders often do not host, and so are more inclined to buy verification. There is also limited accountability here as many hosts will be inclined to leave positive or no references for these people, even if they've had a bad experience.

Our plan to fix it: Member accountability