Why Tinder works? Insights on how to build addictive products that your users love (Part 1/2)
Nivedan Rathi
Founder @Future & AI | 500k Subscribers | TEDx Speaker | IIT Bombay | AI Strategy & Training for Decision Makers in Top Companies | Building AI Agents for Sales, Marketing & Operations
The last half-decade has seen some brilliant product & UX innovations. All wildly different, but every single one deriving the roots of their success in human psychology. After all, no great design has ever been invented with a poor understanding of how we make decisions and what tickles our insides. I have been engaged in reverse engineering the success of many such products to find the underlying commonalities and patterns.
Introduction to Tinder
Those seamless thumb-swipes are one such product innovation that was first popularized by Tinder. The app was launched in 2012, and by 2014, it was registering a staggering 1 billion “swipes” and 2 million “matches” per day. With 50 million active monthly users, these numbers make perfect sense. The phenomenal success of Tinder changed the dating landscape overnight and for the product designers among us - new insights into how the human brain works and how its principles can lead to building wonderful products. Products that engage and addict our users - eventually resulting in a massively successful company.
Deciphering Tinder
To understand what makes Tinder a delightful experience, we’ll use the “Hook” model by Nir Eyal, a framework to decipher and design an engaging and addicting user experience. The model states that for a product to deliver an addicting user experience, it must have four elements, each of which we’ll discuss in detail below:
- Trigger
- Action
- Variable Reward
- Investment
1. Trigger
Triggers are a pointer to the next action that can be performed in anticipation of a reward. Without them, it would be difficult for your users to even start using your product.
In the case of Tinder, it’s the desire to mate - a basic instinct - an internal trigger. For some people, it may also be boredom, loneliness or the desire to socialize & make new friends. All internal triggers. In other cases, a friend’s recommendation or an ad you see on your facebook timeline can act as an external trigger to download and use the app.
Now think - What internal / external triggers are needed or can be created by you for your users to use your product?
2. Action
Once the trigger has done its magic, it’s mandatory for the action to be easy enough to perform. The tougher it is, the more intimidated will the users feel.
The action can be broken down into two parts - one, deciding upon what to do and two, being able to successfully execute it.
Tinder hits hard on both fronts.
It only shows you the bare minimum information you need to make a choice of whether you like the person or not. It builds upon our ability to snap-judge a person of being a potential mate just by having a momentary look at them or their face - an ability that humans have developed over hundreds of thousands of years as a species and a dozen years of social interaction as an individual. Of course, there are downsides of making such snap judgments, but those aren’t a concern for this post.
As for the ease of performing the action, Tinder takes it to a whole new level - a binary choice with a zero downside whatsoever, executed with just a flick of your thumb and bam! Zero cognitive effort and a flick are all it takes to successfully express your interest or the lack of it for a person. No wonder a large user base uses Tinder multiple times a day. Give them too many options like send a message and rate a person and you have the good old dating platforms which Tinder put out of business.
Now think - What can you do to make the main purpose users use your product for minimal & effortless that they can pull it off with a blindfold on?
3. Variable reward
Now this is the part that is the most fun! Ever noticed how the infinite scroll of Facebook and Pinterest keep their users hooked for hours and hours? Well, Tinder has its own equivalent version - The Infinite Swipe. There is a great reward of instant gratification with the added variability of seeing a new person - another potential mate, that is unlocked with each swipe. Interestingly, what keeps the users engaged isn’t just a hot match. It’s the anticipation of a match, satisfying their curiosity to reveal what’s next. This is what keeps users heavily engaged in swiping the infinite deck of cards that holds them ‘plenty of fish in the sea’!
Now think - How can you infuse some variability and uncertainty in the reward that users expect from your product?
4. Investment
This one is the trickiest and most often ignored.
Investment or ‘user investment’ is any activity, upon performing which your users load a trigger for next usage of the product. This ‘loading’ of a trigger is often without the user being precisely aware of the fact that they’re making an ‘investment’.
With Tinder, every time the user swipes right, they create the possibility of creating a successful match. When the match finally occurs, they receive a notification from Tinder exclaiming so - giving them a solid reason to come back and use the app - chat with their match and swipe some more! Got the hook, eh? The action of swiping is thus also an ‘investment’ or ‘user investment’.
Even a simple act of sending a message and awaiting a reply is an investment made by the user that drives engagement (think Whatsapp). Other examples include posting your status on facebook, commenting on a post, liking a page - all leading to a repeated engagement cycle.
Investments also store value and the product itself becomes better with every usage. Consider Quora - the more you use it, the better your feed generated by its algorithms would be.
These investments must also be designed in a way that also make the user work. This may sound contrary to the idea of a simple action but it isn’t. Your product needs to make your users work a bit because people like things they’ve spent their time and effort on (read cognitive dissonance). Isn’t that EXACTLY what you want?
Now think - How can you tweak your product that your users are prompted with a reason to come back to it again and again?
In Part 2/2, my colleague and co-author, Rohit Ghosh will take you through business ideas that can potentially benefit from a Tinder-like design. Most importantly, whether yours can make the cut!? Stay tuned!
Do post your thoughts, suggestions and questions in comments below. Also, I would be most happy to help with your product-specific challenges with this concept & in general.
Special credit: Rohit Ghosh / Co-author
EDIT: Do check out the Part 2/2 written by Rohit Ghosh - "Why Tinder works? Solving for X in Tinder for X"
Manager at Lenskart.com
9 年A useful info to have in mind. Waiting for the 2nd one.
Guest Service Agent at Vinpearl Resort and Villas Hotel
9 年I can not stop read your post! :)
Head of R&D @Teachmint
9 年Another way in which Tinder uses the 'Hook' model successfully is that whenever someone swipes you right (ie likes you) Tinder sends you a notification saying that "Someone has liked you, start swiping now to know who". This creates a sense of curiosity and gets you swiping more than before just to find out who liked you. And mind it, they do it without actually telling you who swiped you right. Now that is pure genius.
Builder @ CoinDCX || Building for Indian Crypto and Web3 Ecosystem || Fortune 40 under 40 || Forbes 30 under 30 || Angel investor || Hiring For Leadership Roles @CoinDCX
9 年Insightful read :) Waiting for the 2nd part :)
Conversational AI | LLM | Autonomous AI Agents | Generative AI strategy | Innovation Management
9 年While there is an obvious reason why it draws so much interaction ;), But from the product perspective, I feel its because it follows 'Preference Elicitation' policy and not 'just' an algo-generated recommendation; which works in the favor of tinder. While, traditional recommenders are known to approach the process as an "exploitation" strategy, which provides recommendation entirely based on the past information , Tinder approaches it lesser as 'exploitation' and more as an "exploration" strategy which intends to understand more about the user's preference by eliciting responses on the recommended choices. The algorithm currently doesn't seem to be performing any reinforcement learning online, but may soon start doing so. As far as the usability is concerned, Elicitation is definitely the reason why Tinder, Inc. or even TrulyMadly.com for that matter works better than something like OkCupid, which is still stuck with the 'exploitation' strategy for recommendation. Soon, the preference elicitation would make its mark in e-commerce, where rather than the trivial process of : System: I recommend products A,B,C User : I buy product B. The Elicitation would go like : System : Which one do you prefer A,B,or C? User: I like B. System: How about E? it is similarly priced, yet packs additional features you might like !