The layers of Trust

What makes one person trust another?

The Stormship Ventures team recently announced that our beta program for iTrustYou will soon be available.

Over the last few months I’ve explained our mission to several interested people, but I’ve never written an explanation in the same way as I’d give it in person. It’s not that I haven’t been writing lots about iTrustYou, it’s that 99% of this writing has been for the UI, FAQs, press releases, landing pages, online ads and so forth. It’s all been very specifically targeted with a very specific purpose in each instance. This is my attempt to explain it like I’d explain it to someone over a beer (or two). Cheers!

So, some time ago, we started thinking about everyday problems to tackle with our system skills. Instead of just building systems for large businesses like we do at bSOLVe (my day job for the last 15 years), I challenged myself to identify a societal problem and try to figure out a way to solve it. One problem in particular drew my attention: The difficulty many people have with knowing whether they can trust a stranger. One very visible manifestation of this is people getting scammed while buying stuff through online classifieds sites.

We initially thought that we might be able to solve the trust problem by simply giving people a tool with which to easily and quickly verify other people’s identities, but we soon realised that it’s not just about identity. Identity is a good starting point, for sure, but it’s only one part of trust. To be able to trust someone at a basic level, you need at least two things:

  1. Certainty about who the person is.
  2. Some form of insight into their past actions, into their reputation.

Earlier this week, the brilliant game designer Nicky Case released The Evolution of Trust. It’s probably the best explanation of game theory I’ve seen and it cuts right to the core of why we trust (or don’t trust) others. Take 30 minutes to play it, I guarantee it’ll be the best 30 minutes you spend this week! One of my favourite quotes is this:

Miscommunication is such an interesting barrier to trust: a little bit of it leads to forgiveness, but too much and it leads to widespread distrust.

The miscommunication he’s referring to is mostly a result of people not being perfect. People make mistakes and always will, but it’s how they handle those mistakes which defines their character. You’re more likely to trust someone who has made a mistake, but then sincerely apologised and fixed it than someone who has never made a mistake. This is the foundation of reputation.

What does this have to do with building a tool which helps people trust others?

Simply put: to make it easy to demonstrate trustworthiness, you need a trail of interpersonal interaction which is easy to share.

OK, but reputation is built into many of today’s sharing economy applications, isn’t it? You can see a host’s ratings on Airbnb or a driver’s ratings on Uber, can’t you? You most definitely can, and it’s great that you can. The thing is, these reputation tracking tools are highly contextual to the interactions facilitated by their apps. They’re housed within walled gardens. The fact that someone has been really diligent at putting fresh linen and flowers in a room they rent to you doesn’t mean much when that person is buying a camera from you and you’re unaware of the fact that they rent rooms. Someone might be really friendly to ride-sharing drivers, but they may be terrible at paying rent on time. There are many examples like these.

With iTrustYou, we’re trying to help people prove trustworthiness across interaction channels. To do this, we’re putting in place two interlinked building blocks for trust:

  • Proof of identity: When you’re dealing with someone, you can be sure that they are who they say they are because the token they’ve sent you confirms their identity.
  • Reputation: Each time you interact with someone who has sent you a token, you provide feedback about the interaction, whether it’s positive or negative. You can either just give a thumbs up or down, or you can add a single short comment. The sender then has the opportunity to provide a single, short response. No fluff.

With both these building blocks, we’re freeing reputation from being bound to specific types of interactions. We’re not calculating some sort of reputation score based on number of ratings, we’re leaving it up to the person looking at the ratings and comments to use their judgement about what a rater said and how the person being rated responded. This makes it possible for each piece of the reputation trail to form within its own context, instead of being confined to a walled garden limited to how good someone is at taking rides or renting rooms.

Very importantly, we’re making it as safe and secure as possible for both parties. The person wanting to prove their identity shares it only with the person they need to prove their trustworthiness to, and the recipient can only give reputation input if they themselves have a verified identity in the system. It’s a one-to-one transaction where someone says “Hey, I’d like to privately prove to you that I’m trustworthy, and I give you permission to attest to my trustworthiness if you feel comfortable doing so.”

This is our mission. To help people build a more trusting world.

PS: If this resonates with you and you’re a South African (we’re starting small for now), my team and I would love it if you signed up for the beta program. Thanks for reading!

Show your support

Clapping shows how much you appreciated Martin Hattingh’s story.