« Back to Blog

How do you know you can trust someone?

trust

Source: kylesteed on Flickr via photopin

Trust in the offline world is very different from online encounters. So what information should you base your decisions to trust another individual on the Web on?

In the real world, trust comes from a combination of time and context. The longer you know someone, the more you’re able to predict his or her behavior in certain situations. You know, for instance, that even if friend X is a complete slob in his own home, he takes perfect care of hotel rooms and rented apartments. You have the necessary information to trust friend X.

Online relationships generally don’t have the luxury of being built on either time or context. As the sharing economy grows, we need a way to make sure that online trust can be as meaningful as offline trust. To do that, we need to think about how we measure trust and reliability.

Insurance companies give us one model. Actuarial science works by predicting risk based on statistical analysis. Basically, that means that insurers make predictions based on patterns. For example, women are charged less for car insurance because they have tended in the past to get fewer accidents and speeding tickets.

Of course, that doesn’t mean that every woman drives more safely than most men, but for insurance companies, there’s an acceptable risk. They stay in business by making enough money from other income that they can offset loss from exceptions to the rule.

When you’re participating in a sharing economy, it’s not enough to make predictions based on patterns. Being the exception to the rule can be disastrous both financially and emotionally, especially if you’re sharing your home or your car.

A similar example is your credit rating. Credit scores are calculated by analyzing someone’s past behavior in order to predict future behavior. While this may be relevant in many cases, the analysis doesn’t allow for outlier events such as sudden health problems, unemployment or other unpredictable events. These events can compel people to act less responsibly than they might otherwise.

An effective system for measuring online trust doesn’t just attempt to predict risk, it looks closely at the time and context of relationships. Even without the luxury of context, there are ways to make sure that people are worthy of trust. But we can’t do that by relying solely on past behavior.

In the Press: