The Future of Dating

Can you engineer love with token incentives?

User avatar


I like to do deep dives on products or problems that caught my interest. Today's article is inspired by a friend who sent me this amusing thread on "Web3 dating app".

Tokenomic designers are not experts at love but they are experts at market. And dating is a market.

In this essay I will go through 1) Problems with dating apps, 2) Current landscape, and 3) What the future of dating might look like with tokenomics.

Problems with dating apps

I've been on a few dating apps, and I was never really satisfied. The main reason I like them is that you can meet people outside of your social circle, and you can do that in the comfort of your own bed, whenever you have time.

But dating apps have a lot of downsides:

  • low conversion rate: incentives to meet in person is to hookup. For people that don't want it, getting a date off the app is difficult.

  • geographical limits: not enough users you can match with if you live outside of extremely large cities. The reality is online dating is a numbers game, and you need volume to get to the 1 or 2 people who are actually good matches.

  • skewed ratio: way more men than women on the app, which aggravates the market inefficiency- the most attractive ones are flooded with requests while the rest sits at the bottom pile. To be fair, it's not that society doesn't have single women. It's just a) they aren't using these apps, b) we need better matching algorithms.

  • the botcalypse: dating pool diluted with fake profiles & people who are there to promote businesses. Creating a negative user experience, many users simply give up.

  • other human factors: it's boring. Profiles are all the same and it feels like browsing a catalog; Managing all the conversations you started is akin to a full-time job; flaky matches, etc.

Current Landscape

During my research, I found out that in the last two or three years, new dating apps catering toward Gen Zs have emerged, many of which target growing behaviors around

  • preferred date destination (Desti)

And some new product features that seem counterproductive to dating:

  • Hiding profile pictures, such as blurred pictures (S’More, Jigsaw) and voice-first dating (SwoonMe). Tinder and Bumble also experimented with weekly "blind matches". In my opinion, physical attraction is a big component in matchmaking. Hiding profile pictures won't reduce market inefficiency as it's against dating psychology.

  • Orienting the product around live video. Meta tested this briefly with Sparked. This help reduces fake profiles, but video call is high effort. Most people probably use dating apps in bed, disheveled, wearing an old t-shirt. This feature is fighting against human laziness.

  • Adding countdown timers (Match). This is supposed to help keep conversations going and prevents ghosting behavior. But dating app is an asynchronous network— you initiate and respond to interactions when you are available. Adding an expiry time decreases the benefits of an asynchronous network.

Features that are actually helpful:

  • Sort by dating preference (Fruitz). Users will only match with people that shared the same intention

    • As an interesting side note, the Stable Marriage Problem asserts that lying creates inefficiency in the matchmaking process. The best strategy is for everyone to tell the truth, to be as honest as possible.

  • Segregate scammers into their own pool with bots that talk like real people (Filter Off)

  • Focus on IRL dates and specific local events (Inner Circle). It's obvious dating apps should be about maximizing fun dates.

  • Letting your friends be the matchmaker. Ship have tried this, but the app is in graveyard now. Your friends are probably not altruistic enough to keep doing the matching for you.

    • Someone should build a Pallet but for dating. A "singles collective" that lets community leaders introduce you to a curated group of singles from their network.

What the future might look like


  1. People simply opt out of dating. Interestingly 40% of Japanese men in their 20s reported having never gone on a date.

  2. Sexbots become widely accepted. If you don't want kids and don't want the troubles that comes with human interactions, why not just have robots as partners?

  3. Achieve 100% compatibility with algorithm matching. The most efficient way of finding love might be training AI algorithms to matchmake people. I imagine a GPT-214 like the one in Black Mirror.

Pessimistic but with tokenomics

Since dating app requires network effect, tokens create a financial incentive for people to be early participants.

Where does the money come from? Partnerships with governments, hotels, restaurants, bars, amusement parks, etc.

  • During bootstrapping phase you give out rewards $DATE (utility token) for joining, interacting with other members, meeting IRL, posting about the app on social media

  • Since liquidity depends on the number of connections available for a user, liquidity provider (ie. introducing a friend to sign up) gets rewarded with extra $DATE

  • After a match happened, users need to stake $DATE to start a conversation

  • Conditional Burn: Both parties should have the right to force another one to burn $DATE (or donate it to charity) for ghosting or fake accounts. But if the parties decide to split amicably both gets their portion back

  • Users can use the $DATE token to upgrade membership (tradeable NFT passes) or redeem experiences

Limits of token incentives

  • Dating app requires high engagement and new sign ups. When the financial upside is gone, so will new signups and engagement

  • Tokens might attract the wrong type of users— those driven by financial incentives instead of those looking to find love. So a web3 dating app might attract people who actually are in a relationship, or just there to earn and speculate on the tokens, instead of engaging with the network.

  • At a high level, tokenomics designers need to ensure only users that add value to the network can receive token incentives

On the topic of love, a study on ten thousand couples suggests that the person we choose is not nearly as important as the relationship we build.

Relationship-related characteristics are likely to account for about 45 percent of the differences in relationship satisfaction. By contrast, a partner's personality may only account for about 5 percent of that relationship satisfaction.

So, we should probably spend less time worrying about finding the perfect partner and more about building that relationship. Okay, how do we add a token to solve that?

  • Loading comments...