I’m writing this as a conclusion to and reflection on my time building EmojiSwap, an app I worked on for ~1 year. All startups are unusual, but this one felt even more so - for a while I struggled to articulate the experience to others, let alone make sense of it myself. This post is meant to explain the evolution of EmojiSwap from beginning to end, and everything I learned along the way.
EmojiSwap was originally conceived as a silly project, as a distraction from another app I was working on that didn’t have any traction. I created it in the vein of other whimsical, performance-art-esque social apps that I was obsessed with at the time. MSCHF was a big inspiration, specifically their Finger on the App and White Whale works. I was also inspired by Yo, Who Paid 99 Cents, and I am Rich. I just wanted to build something viral; I was tired of having no users.
My idea was a chat app where you could only chat with with emojis - specifically 5 random emojis that you were assigned when you signed up. If you didn’t like your emojis, you could trade your friends for the ones you wanted. That was it.
After some development, I threw up a demo video on Twitter, which got a lot of unexpected attention. I decided I needed to finish the V1 as quickly as possible to capitalize on the hype.
In February 2022, I launched version 1.0 of EmojiSwap, sharing it with friends and a few dozen people who reached out to me on Twitter. I realized a few things quickly. Chatting was amusing, but it was gimmicky. It got annoying after a bit. Everyone wanted to own their favorite emoji though, and they were actually seeking out trades. Trading was the most interesting part!
Soon after launch, I decided to double down on trading - I wanted to make every emoji a unique, collectible item. I wanted to create a public trade ledger, rarity tiers, and traceable ownership histories. As I imagined it, I wanted to build “the world’s first emoji exchange”.
This is when EmojiSwap took a new primary inspiration, as a parody of NFTs. 2021-2022 was peak NFT mania, and I was tired of it. I liked the idea of digital collectibles, but I never understood the crypto angle. The art was tasteless and the technical reasoning was dubious, but everyone was making money, so no one batted an eye. I realized I could simulate all the best bits of digital collectibles without using crypto, and I could launch a native iOS app, something that NFTs couldn’t do. (Good luck getting people to use your mobile Safari PWA).
EmojiSwap has many accoutrements of an NFT market, just without the underlying decentralized technology. Much of the app’s terminology, like “swap”, “airdrop”, and “mint”, was borrowed from crypto. The name itself is a parody of decentralized exchanges like UniSwap and PancakeSwap.
The premise of NFTs is that you provably own digital art. But this is only true in the context in which you bought it - outside that context your ownership is meaningless. The problem with NFTs is that the context, the blockchain, has no strong meaning beyond as a transaction ledger. Projects promised “utility” but rarely delivered; the majority of NFTs are simply jpegs stored in AWS S3. [1] I teased at this fragile idea of ownership. Of course you can’t “own” an emoji. It’s right there on your keyboard. You only own it within the context of my app, and you have to buy into the pretend world that this app represents for it to mean anything. So my goal was simply to get people to buy into the world.
I launched the “emoji exchange” update in April 2022, and everything became clearer as I pushed down this direction. EmojiSwap turned from a side project into a real product, and I started focusing on it full-time.
Telling the Story
From the beginning, I experimented with marketing EmojiSwap on TikTok. I was not naturally a social media person, I thought it was cringe (still do) [2], but my previous failures had me so desperate that I was willing to try anything for downloads.
I struggled for a long time to figure out what worked. Early videos didn’t get many views, but they did earn me my first early adopters. The first breakthrough post was one in which I teased that the 🦤 emoji was going “extinct” within the app - no more would be minted. I got thousands of ironic comments to the tune of, “wdym it’s extinct I have one right here 🦤”, which I actually liked, for two reasons: one because that fake scarcity was part of the parody, and two because the TikTok algorithm loves comments.
This video’s success helped me realize I could build an immersive virtual world for EmojiSwap, and I could create events within this world to change market dynamics. The 🦤 extinction led to more in-app events like banning the 🔫 (with a buyback program!), and the “Federal Reserve” ceasing minting 💸 due to “inflation”.
I tried a lot of different video formats. What ended up working best was showcasing interesting trades of rare emojis. These videos worked because they set forth a crazy premise - trading rare emojis??? - and actually followed through on it.
Another lesson from TikTok: If you don’t explicitly plug the product, your video means nothing. Other videos of mine got millions of views but didn’t include a CTA to download the app, and converted poorly. You’re on TikTok to sell a product, not to build a following, so don’t be afraid to be explicit at the expense of virality.
Eventually I developed a coherent pitch of what this app was, and where I wanted it to go. It definitely did not start that way. Many ideas had been loosely floating in my head, but it took a lot of work to forge them into a coherent narrative that I could share with people.
The pitch had 3 main pieces:
Trading is a fun game mechanic
It’s exhilarating to negotiate with friends, make a trade, and build up a collection. EmojiSwap captures the thrill of trading in a way that other games haven’t before.
Emojis as IP
Intellectual property is a big component in mobile games. Studios need to spend big money to license IP or create their own. Emojis are sort of an IP arbitrage - they’re ubiquitous and nonproprietary, but also universally loved. Instead of convincing players to buy into new IP, EmojiSwap taps into the existing enthusiasm for emojis, the world’s largest open art collection.
And a simple medium isn't necessarily limiting. You can express just about any idea or emotion with emojis; simple primitives can can create complex worlds.
Make it real
EmojiSwap is an absurd premise - taking text characters and turning them into rare collectors items. That draws people in, but the key for it to work is to make the experience feel real - not just through marketing, but through the product itself. The more depth built into the product (trading, analysis, collection tracking tools), the more real it seems.
For this reason, I spent a lot of time crafting the narrative. I was careful not to refer to EmojiSwap as a “game” in public messaging, because I didn’t want to break the illusion. It’s not a game, it’s a real trading market!
At first, I neglected the need to clarify my pitch because the app was growing without one. I felt the product was self-evident and anyone who didn't get it simply wasn’t a believer. That was an ignorant lie. Writing forces you to clarify your thinking, and it’s an uncomfortable feeling when you confront an idea you thought you had a grasp on, but can’t express coherently. I won't make that mistake again.
Building a Business
I remember speaking to a friend early on, when he admitted bluntly that he didn’t understand the business behind EmojiSwap. “What’s your model?”, he asked me. I didn’t have a great answer, because I didn’t have a real idea. In retrospect, the answer is I was building a game, and mobile games have clear monetization models that I could work off of, but I didn’t realize that I was building a game. To that point my entire experience and way of thinking was in mobile apps, not games.
EmojiSwap was originally motivated as a parody (an art piece?), but it became a real product, and I had to confront the realities of building a real business. I felt partially conflicted by departing from this “pure” vision of what the app was supposed to be, but I knew that I wanted to be an entrepreneur, not an artist.
The game mindset further clarified my roadmap. I was still building something new, but now had many reference points to draw inspiration from. I added an in-game currency (EmojiCoin), an item shop, a mini-game arcade where your emojis become your characters, and more. I had new benchmarks and frameworks to think about retention, monetization, and progression. All of these concepts were borrowed from other successful mobile games.
I was hesitant to monetize the app because I always wanted to focus on creating value, and was wary of making the game pay-to-win. I probably could have made 10x the amount that I did, but I opted for tame monetization methods like rewarded ads and a flat monthly subscription.
Discord
I started a Discord server when I launched the app. I hadn’t used Discord much before this, but I saw a few friends create servers for their own apps, and decided to give it a try. In the earliest version of EmojiSwap you could only trade with people you knew; there were limited ways to find friends. People wanted more trades, so I invited them to my server to matchmake with each other. It became sort of a missing liquidity layer for the immature product.
I quickly began to see Discord as an extension of the app itself. I aggressively pushed users to join the server from within the app. I used it for just about everything - announcements, user support, feedback, forums, events, and analytics. I could only update the app so quickly, and Discord’s flexibility was great for filling in the gaps. EmojiSwap is overwhelming at first glance, and I tried many things to improve this, but I suspect none worked better than funneling users into my server and showing them the best ways to get started.
At its peak, my server had over 10,000 members - I never anticipated it getting so big. All the most engaged users were there. At times I had to remind myself that the user base was larger than the server (90% of users weren’t in it), because it felt so engrossing. Discord was my most legible view into how people were using EmojiSwap, better than any analytics could tell me.
I spent a lot of time observing. People would send me their bugs and suggestions, but most interesting was reading how users were hacking around the edges to accomplish what they wanted. (Not literal hacking, although I dealt with my share of that too). I took those behaviors and turned them into features for the app. For example, I noticed many users were posting photoshopped images of all the emojis they had ever owned, and using them to seek out the emojis they had yet to find. I built an in-app feature to do that automatically.
Scaling was tough, and in retrospect I probably spent too much time worrying about the server that I could’ve spent working on the app. I dealt with plenty of wild issues, but luckily I found some great people to help moderate. I liken the whole experience to running a virtual summer camp.
Reflections
I worked on EmojiSwap primarily from January 2022-January 2023. It is not dead - I intend to keep it running for as long as I can. I don’t actively update the game, but still fix bugs and occasionally run events.
I mostly spoke above about things I did well, but if everything went well, then I wouldn’t be writing a retrospective today. So I’d like to reflect on things that I did poorly.
Mainly I did not build beyond myself. I tried to do everything, to a fault. I couldn’t find other people to work on other parts of the app. I needed to get better at selling and recruiting people, whether it was engineers, users, investors, or advisors. I shunned the idea of raising VC funding, even when presented opportunities to do so - my reasoning was that EmojiSwap could be efficiently built bootstrapped, but I think I also felt uncomfortable with the new operational scope and pressure that funding would bring. Although I ended up becoming pretty good at making TikTok videos, it still felt unnatural. It would take me hours just to make a 1 minute video, and as an introvert, it was emotionally draining. In retrospect, I should have outsourced this piece to someone else after I figured out a video format that worked.
I also simply didn’t execute well enough or fast enough. I had no formal software experience and incurred a lot of bad technical debt due to my impatient desire to just ship stuff. While that impulse is good, it definitely held me back from making a top-tier mobile game. Some features I built were cool, but others were sloppy or impractical. I shied away from some complex features because I didn’t know how to build them. Games are some of the most intricate products in the world. The bar is high. I was at an advantage because I built something very different from most mobile games (and marketed it differently as well) [3], but that is not sufficient to succeed.
In February 2023, I decided to join Airchat, where I got to work with amazing people and build cool interfaces for producing and consuming audio content.
I also felt creatively stuck at that time. Growth was ok, but not accelerating. I was highly dependent on TikTok. I had a lot of ideas on where to take the product next, but didn’t have clarity on how. I got conflicted about monetization, because I wasn’t big on building out more loot-boxes or catering to whales. I knew that I hadn't maxed out EmojiSwap’s potential, but lost confidence I could get it to the next level.
Many early product constraints actually came back to hamstring me. Emojis are a wonderfully simple medium, but there are only so many interesting ones. 5 emojis at a time forces you to pick carefully, but prevents amassing a big collection. I’m reminded of this quote from Kevin Systrom, founder of Instagram:
…“Wow, every time I see a startup stick to their original, what I’ll call their sacred cow or their 140 characters, that never works out in the long run.” Our version was square images, our version was just photos, our version was Feed. But I think we got lucky enough, and I’ll say lucky because I think we went kicking and screaming against some of those sacred cows and we were lucky enough to evolve and people still loved us as we grew. But you have to change what you’re doing drastically almost every three years, to stay alive. [link]
I do sometimes wonder what I could have done if I didn’t step away. I simultaneously feel proud at what I built and embarrassed at what I didn’t. Although I didn’t reach the promised land, I know that there was something real and resonant at the core of EmojiSwap.
I still think trading is an under-appreciated game mechanic, and a very fun one. A few apps have taken this idea to whole new levels.
One is Real, which is primarily a sports data app. I spoke to John and Louis, the founders of Real, when they first wanted to add a collecting meta layer into their product. Real cards are sort of like digital trading cards, but for specific plays instead of players. They’ve built incredible depth around this feature, but the complexity can be overwhelming - I don’t even understand all the pieces. I trust the Real team will find a balance between simplicity and depth, turning it into the most exciting collecting experience for sports fans.
Another is Soundmap, which is kind of like EmojiSwap for music mixed with Pokemon Go. Music is a much deeper medium to pull from, and even more beloved. Everyone wants to collect their favorite songs and artists. I’ll admit I’m skeptical of the Pokemon Go mechanic, I think it doesn’t make sense here and they can do just fine without it. But the team has done an amazing job building around the collecting experience, and beyond making a fun game, they have the potential to become a real artist-fan engagement platform (something that many web3 companies unsuccessfully tried to do).
I have some new trading ideas as well, if I do ever return to making a game like this. But my time on EmojiSwap is complete. ✌️
Footnotes
[1] One of the few interesting developments here was twitter launching NFT-linked profile pictures, expanding the context in which an NFT means something. But they killed it.
[2] Sometimes you have to embrace cringe to be authentic.
[3] The mobile game industry, despite being only ~15 years old, is pretty mature. The competition is intense, and dominated by top studios spending big money on social media ads to buy downloads. Most of the value here is actually captured by the ad networks (Meta, TikTok, Google), so any distribution method which can circumvent this is a big advantage. I was lucky to discover organic TikTok as a channel early on, and I never seriously experimented with running ads. I also built the product around sharing and social (because I originally thought I was building a social app), whereas most mobile games are single-player.
留言