Shaping Words With Polygonle’s Competitive Creator

Published Apr 23, 2024

Hello! Can you tell us who you are and about your game?

Hi there! I’m Grant Warman, also known as “blorppppp”, stylized with five p’s, and I’m the creator of Polygonle and Quadragonle. I primarily work as a software engineer, and the technical skills I gained from my previous jobs helped me create Polygonle and its off-shoot, Quadragonle.

Polygonle is a 4x4 word game with shape-based clues. It works similarly to Wordle, but is made more accessible with polygons acting as hints to what the word is. Each polygon or shape corresponds to a specific letter, and these shapes become more useful whenever there is a word where a polygon appears more than once. So think of Wordle’s word-guessing mechanic that’s made more understandable with shapes, that’s Polygonle.

What games do you like to play?

I was waiting for this question! Unfortunately, it’s not a word game, but I play competitive Super Smash Brothers Melee, the old game for GameCube. That's my main gaming passion these days. And, actually, the idea for Polygonle was spawned in a Discord group for that specific game.

I play the GameCube version for several reasons. It has unique game mechanics that were sort of removed in later versions, and it has this really fanatical grassroots community that just loves the game. Because the game hasn't changed since its release, I think it's been 22 years or so, it has developed this really remarkable meta that is evolving over time where characters that were considered very bad when the game was released, you're seeing people use those at the highest level. It's almost like chess in the sense that there are truths about the game that are discovered and people are ridiculously good at it now. So it's really fun.

I have always been very competitive, and I think that is the central theme in my life with games, I hated losing and I loved competing. I played games like Smash when I was younger just for fun and to compete against our friends. And then as I grew older, I really enjoyed this process of improvement, of trying to identify the areas of weakness or your strengths and doubling down on your strengths. Really, just the art of improvement in general is something that is interesting to me, and games are kind of a vehicle for that.

Experience in competitive gaming

Yeah, so the competitive nature of it is attending tournaments in person. There is an online ranked format that some people have built on the side as a passion project, but the main thing is traveling and attending tournaments. So here in Denver, there's a tournament basically every day of the week in one of the cities nearby. So there's one in Denver, another in Fort Collins, there's one occasionally in Colorado Springs, and there's one in Boulder.

There are events all over the place, and there are also these larger events that you'll go and travel to that have, like, 1-2000 attendees. And those are also a really fun experience because when you compete, then get knocked out, you can just go watch the pros. It's like a sporting event, and it's really exciting to watch your favorite player compete.

Currently, I’m not officially ranked, but I was so close to being in the top ten in Colorado. I started working on some other projects, so yeah, not at the moment. However, I can still very much give everyone a run for their money, and I’ve taken some games off of some ranked players as well, so it’s pretty wild.

Prizes in competitive local tournaments

Most tournaments do have a prize pool. Personally, I have made a very tiny amount of money out of the tournaments, and even for the very best players, it's not very financially lucrative. So when you enter, you normally pay some entrance fee that goes into a pot, and then, depending on the size of the tournament, the top three players or maybe the top eight will get a payout. So yeah, that's how the payment bit works.

On other games

Chess is the other big passion I should have mentioned, and I think it and Super Smash Brothers are related. This is more noticeable if you play faster formats of chess, like Bullet chess or Blitz, on a faster time control. So in chess, you'll have a tactic, a series of moves where someone will gain an advantage. And that exists also in Smash, where if you are aware of what's going on and you hit the right opening, there's a forced sequence of moves where you can respond to everything your opponent does and get a large advantage. There is a specific point where both games can enter a pattern that you can distinguish, and that aspect is similar.

The other thing is people discover, like, new techniques. So what happens in chess is there are these high-level principles that people try to follow, and then the best players learn when you can break those principles. The same thing applies in Smash, where there are, like, “good options” in certain positions. And the very best players know when to mix in what would typically be a bad option since it catches their opponent off guard, and then they gain some large advantage. 

So, yeah, there are definitely some similarities and a lot of differences, but the cool thing is just there are truths about the game to be discovered, and they're still discovering things in chess and Smash, and I love it.

On chess

I mostly play on lichess.com, and I play on chess.com occasionally. But, yeah, I started playing in my early twenties, and chess is super fun. An easy place to burn time and improve. You could spend all your time just practicing puzzles and tactics and improve a lot. But I reached a point where I'm good enough to watch and appreciate grandmaster-level play. I think that is a good level. But I don't have any serious aspirations of trying to get a title or anything. It seems like an enormous amount of work.

Could you tell us about the creation of Polygonle?

Polygonle is the first game I've made. I've definitely been playing games for a while, and I was playing Wordle during its peak. I made Polygonle just as a fun little side project with no intention of it becoming something popular, and it kind of grew legs and became a thing.

As for the shape-based clue system, it all started in the Smash Brothers Discord group in Colorado. Someone had attended a tournament, and they posted this riddle. Someone had invented this little cryptographic puzzle, and it was this picture with a series of different shapes, and there was just one answer. So those shapes, each shape, corresponded to a letter, just like how it works in Polygonle. And there was one distinct word that fit that pattern.

If I remember correctly, it was like the word “cataract” or something. And so someone posted this puzzle, and people were trying to solve it. And for a group of video game nerds. I was surprised by how excited everyone was to solve this word puzzle. So being the programmer I am, I wrote a script that solved it, and I ended up spoiling the solution for some people, and I felt kind of bad.

At the time, I had just quit my job at Verily, so I had a lot of free time on my hands. So I just started building some tools to generate these sorts of puzzles. And the reception from this community on those puzzles was really tremendous. I was really surprised by how many people were engaged and interested in solving these. And then I was thinking I should try to figure out how to transform this into something that people can play regularly.

And that was what ultimately led me to build Polygonle.

Technical aspects

For Polygonle, I had a really relatively simple heuristic for getting the list of words. I filtered out words that weren't very common in the English language, and then I looked for words that had enough repeated characters so that the shapes would actually give you a hint.

And then the only other really big change I added is in the first few months, there were a ton of words that ended in -ing. So people would just consistently guess a word that ends in -ing, and they would be right, and then it would give them a bunch of information, and that wasn't very interesting. So I left some words that ended in -ing, but I stripped a huge number of these so that it was more diverse from day to day.

The average Polygonle experience for casual players

I would say the average player gets it in three to four guesses. It behaves very differently than Wordle, even though it's somewhat of a small twist. The really advanced players will play on what's called Expert Mode, which will only allow guesses that conform to the pattern. So if you see a shape repeating, your guess has to fit that pattern. It is far more difficult to play in that mode.

I basically never play on that mode. It's really, really hard. The upside of that is there may be only 40 different words in the whole dictionary that fit the pattern, or on some days it's just like two or three. So it might take you 30 minutes of thinking to arrive at the answer, but the likelihood of getting it in a few guesses goes up a lot. And then we also have another set of players who play it exactly like Wordle.

They just think of a seven or eight-letter word, depending on the day, and get all the information they can. They don't mind how many attempts it takes, and then, you know, they just try to reason through with the shapes and then traditional reasoning what the solution might be.

That's what I would expect from most people. I think the cool thing is seeing the different solving strategies. So a lot of folks, when they become more experienced with the puzzles, start solving before even guessing anything. They look at the pattern of the shapes they plug in, like suffixes or prefixes, and see what words could possibly fit that pattern. They use that to have a really educated first guess, and then they'll normally get it in like two or three.

Engagement time

I think the average daily engagement on Polygonle is around five minutes. On a Quadragonle, it's about three minutes per session. So I don't know if people are just solving it super fast or they're solving one and then leaving. The original intent behind Quadragonle was also to build something challenging so that people are engaged for a longer period of time.

On Quadragonle

I think we had this discussion about game design and what makes a game unique and stand out on its own versus being more heavily influenced by an existing game. And Polygonle was absolutely a game that was just a small twist on Wordle. So I was trying to build a game where you were just faced with this puzzle. There was no guessing mechanic, and your goal was to complete the puzzle similar to a Sudoku. After some iteration, I came up with this idea of a grid of words, just like Polygonle, that you had to complete in a way that there were valid four-letter words in the rows and columns, just like a crossword.

And my first iteration of this, I generated some puzzles and had some people try to play them, and it was just way, way too hard. Like, no one could solve them. But the few who did solve them actually really enjoyed the mechanics of the game. So I ended up iterating on this idea a little bit and coming up with different difficulty levels. So, just like with Sudoku, if you do a really easy sudoku, most people can just solve it using very simple deduction.

And if you try to solve an expert Sudoku, many of us are not even equipped to solve them. So that was kind of my goal with the difficulty levels in Quadragonle, where the easiest puzzles could be solved by most people without too much outside help. But the expert puzzles, a very small set of people are going to be able to solve those puzzles. 

The nice thing about this is that a more casual word game fanatic can go and solve the easy puzzle and maybe push themselves and try to solve a medium puzzle, but the real expert players, they'll solve all of the puzzles, and they can solve the expert puzzle in under a minute sometimes.

And they're just really skilled at doing it. So it's nice to be able to build a game that appeals to both a casual user and a more serious and skilled player. And I think we struck that balance with Quadrigonle.

Has Polygonle changed drastically since its launch?

There are some notable changes. Before, you couldn’t freely place letters on the boxes, you had to fill them in sequence. I think the first few social media folks I saw that tackled the puzzle, they would use like, “x” and “s” as fillers. And I very quickly was like, this is not going to scale. People do not want to play this if they have to do that every single time. So now you can place the letters wherever you want, regardless if the preceding cell is blank or not.

So as for the actual game mechanics, there haven't been super large changes. I think the cursor changes and being able to select an individual tile and fill it in, that’s some of the changes. We've had a couple of iterations of that, so there's also a version where you can have it automatically fill in the matching shape. And then the Expert Mode was something that we added after the fact.

I think most of our diehard players use that mode almost exclusively. They only want to solve it in that mode. The only other pretty large change we added was an achievements system. So there are some pretty lofty achievements that people can earn for, you know, solving the game 500 times or a thousand times or something, and also solving on a different difficulty mode. So there's a whole host of achievements that people can try to earn, and that's a feature that's been pretty well received.

Feedback system

Yeah, so we have a couple of different communities. We have everything from Facebook, Reddit, Twitter and Discord. And I think our most vocal community is on Discord. We have a feedback section there where people can propose new features for Polygonle or just give feedback on game mechanics. And we had a number of iterations on Quadragonle, for example, where the first iterations were just too difficult, too frustrating.

And when you hear one person give a piece of feedback, sometimes it's hard to know if it's just them or if it applies to a larger group. But the nice thing is that our communities are large enough that you'll, if something is really important, you will hear it more than once. Multiple people will share a similar sentiment. And that is normally what drives the things that I will actually incorporate. Same thing in Polygonle with the ability to select individual cells.

That's a feature that very quickly we had numerous people saying, “Hey, this would be really helpful if I didn't have to use these fillers.” And it was a very straightforward feature to add, and people were really receptive to that improvement.

Have you collaborated with other games and creators before?

I have, and one of them is Thirdle. These collaborations have been really helpful in growing our game, and as a result, I felt the need to kind of get back to some other newer games in the community. I forget when exactly Thirdle was created, but they launched after us. And the idea of just having a crossover puzzle is great for both communities. It introduces a new game to a population that maybe has never heard about it.

There's normally some kind of theme that spans both of the games. So there's a fun aspect to that. We got the idea of doing this from Squaredle, I don't know if you're familiar with them. So Squaredle and Waffle did a really big crossover a long time ago, and the creator of Squaredle attributes a huge amount of their success to that initial crossover. And similarly, for us, we did a crossover with Squaredle.

There are so many Wordle clones now, so what do you think is the fine line between inspiration and plagiarism?

Yeah, that's a great question. I think if the foundation of the game mechanics is identical and the way someone plays the game is the same, then it's not a sincere attempt at generating a spin-off type of game. I think you can have a remarkably small twist, and the way the game is played and the strategy can be completely different. That's how I felt with Polygonle, where the gameplay aspect is very similar to Wordle but still has that different flavor that sets it apart.

There's still a hidden clue when you make a guess, it gives you some information, but the way you play the game and the way you problem-solve in it is completely different. And that is this kind of emergent property that I didn't design necessarily, but just the nature of this small twist of giving an additional piece of information at the start just completely changes the way your thought process works.

How does your monetization scheme work?

So the first nine months of Polygonle, we had no ads. Then we reached a point where with the amount of traffic we were getting and some of the new features I was interested in building, we needed some kind of income stream. So I built a premium offering for no ads. Then most of our users are playing the free version, which does have the ads. And yeah, it's been pretty remarkable. Frankly, it's a lot more profitable than I ever would have imagined when building this game as just a fun project.

And now that I’ve left Verily a while back, I can confidently say that this is now my primary source of income. Though I am working on a new project now that's still in the game space. It's coincidentally related to my Super Smash Brothers passion. But yeah, this is my primary form of income. And then I was, I was working at Google and Verily, I don't know if you're familiar with them, but it's under the Alphabet companies.

So it was originally one of these other Google bets that spun off into its own company. So, yeah, I was fortunate enough to save a decent amount and work on fun side projects like Polygonal, and it's awesome.

How did you market Polygonle?

So I think when I first created the game, I emailed some people who had blog posts about Wordle games and I emailed some creators on YouTube and TikTok. If I remember correctly, I think it was in the first week one of the TikTok creators made a post. It was the first post ever about our game, and it had 300,000 views or something. And that TikTok growth was all organic.

I mean, after my first email, I did basically no additional outreach, and the game became really popular among a set of TikTokers who solve wordle games every day. Like, that's their whole platform. They're just really skilled at word games. They play them every day, they do all the New York Times puzzles, and then they'll play some of the offshoots. And that was all organic. I mean, there was nothing that I did other than just tell people that the game existed. And we have tried some other ad spends, which were very unsuccessful.

So I would say the overarching trend was people like to discover a new game when they see someone play it, especially within their own little communities on TikTok. So when you have a creator who's playing the game every couple of weeks, that creates a following of other people who want to participate too. So they want to play the same game and compare how they did versus how the creator did. And yeah, that's been a huge driving factor behind our growth as a game and our decline as a game.

Is there any advice you can give to aspiring game creators?

Yeah, that's a great question. I think it depends on what your goals are as a game creator. And I have a few I’d like to share:

1. Adapt elements of popular trends and games

If your goal is, “I want to create a really popular game”, and you don't care about being artistic and creative, I think taking an idea that works well and putting a creative twist on it is a really good way to be successful. And part of that, it's similar to doing a startup company where if you're creating a whole new market that has never existed before, the upside is enormous. But most people don't have the right ingredients to succeed when they create something completely novel.

There's a very small set of people that figure out those exact correct elements, and that's the benefit of starting from something that is successful and then putting a twist on it. Hopefully, the core mechanics and the core game design components that made this inspiration successful will also benefit your game. And I've seen a number of games that have succeeded with that model. So I think taking inspiration from successful ventures without being plagiaristic is a good start.

2. Observe your community and make it easier for them to have fun

Earlier, I told you that many of our first players had to type filler letters just to progress to the next tile, right? Just seeing those minor inconveniences already rang some bells in my head that prompted me to act on it. Everybody just wants to have fun and jump directly into the game, and no one wants even these minor tedious things to get in the way of the experience. So if you want a game to be loved and stay loved by a good community, make it easier for them to have fun.

What’s the future for you and Polygonle?

While our cumulative numbers have reached over a million different users from so many countries, I think that this is a game that is not going to reach, like, Waffle levels of popularity. There is a smaller subset of word game fanatics who really love the challenge of longer words and this problem-solving approach, but I think it's a smaller population of users, so I don't foresee us growing ten times or anything over the next few years.

But I think we'll be pretty stable and continue to have a really invested fan base. I am not currently planning on putting a lot of more development time into Polygonle. And yeah, I mean, when I quit my job at Verily, my original intent was to build a larger business. And I have some experience with startups in the past, so that was my original intent. 

And then this Polygonle thing came out of nowhere, completely unintended. Learned a whole bunch in the process of building the game, iterating on the game, interacting with creators, and building that sort of network. So that's all been a great learning experience. But I reached a point where the next goal is to build a larger business that won't be a single game but will be something that's solving problems for a wider set of people.

So what I am working on now is disconnected from the word game or puzzle space. I’m not building a game, but a platform for managing and running tournaments.

Where can we go to learn more about you and your game?

I don't know if I have a real central location for an online identity, but you can find me on Twitter/X @blorppppp_. It's very convenient. But actually, you should probably go look up @PolygonleGame. That's where we post about Polygonle.

Have a game to sell?

Let’s find out if we play well together.