(For the other parts of my “Lessons Learned”-Series, click here)

Sometimes, the best ideas come to you at the unlikeliest times. The idea for Letteral came to me at 1:30 a.m. being very tired and yet unable to sleep…

As long as I can remember, I love board games (Being German, that’s probably in my genes…), especially the strategy heavy ones, like Agricola or Tzolk’in or Puerto Rico. And ever since I owned my first Palm PDA (a Clié PEG-SJ33) and discovered Popcap’s Bookworm, I’m also a big fan of word games. So it shouldn’t come up as no surprise that games which combine elements of these two genres, like Letterpress or Nimblebit’s fantastic Capitals, are right up my alley.

The game that I imagined at 1:30 a.m. was quite different from the one that you can now download from the App Store, but it was a a start. How it grew into the Letteral you know and (hopefully) love, you can find out here.

The Premise

Two years ago, I was looking for an app project that I could work on in my spare time and that would be the creative outlet besides my regular job as a mobile developer. This came with a couple of conditions:

  • The app had to be good enough so that I could put it on the App Store one day
  • Having worked on a Tumblr client for my last project, this time I wanted to make a game
  • The project was to be small enough so that I had a chance of actually completing it
  • Although I’m a fairly experienced with UI-Design, most three-years-olds are better at drawing than me, so the game’s visuals had to be simple enough

Combine all of these stipulations with my aforementioned love for word- and strategy-games and out came the idea of creating a two-player game which combines word knowledge and strategy. My 1:30 a.m. idea wasn’t Letteral though, it was something much more rectangular:

The idea was the following: Players take turns finding words in the 25 letters. You were able to use all letters, no matter if they belonged to you, your opponent, or if they are still neutral. When you find a word, you’ll take ownership of all its letters. In order to score, you’ll need to own all five letters in any of the five rows or columns. After scoring, all tiles in this row/column would be filled with new letters and become neutral again.

Refining the idea

This game, which I tentatively called 25Letters, already sounds a lot like Letteral. However, after taking a Saturday to create a quick prototype, I quickly realized that it had two major issues:

  • You had ten opportunities to score (5 rows & 5 columns) and each one required five letters. It turns out that players almost always scored in their turn, making the game fairly benign
  • Especially with scoring happening so quickly, there was no real element of interaction. Players would play side by side and not with (or rather against) each other

I quickly found a solution for the lack of interaction: If all letters that you use for a word become blocked for the next turn, players have the opportunity to interfere with their opponent’s plans. After I implemented this change gameplay was significantly better. But the first problem remained, it was still too easy to score.

One way of solving this would have been to increase the boards size to 6×6. This makes it harder to take over an individual row/column but has problems of its own. First of all, having 36 letters available is just overwhelming. Secondly, finding words by itself should be a challenge and thus rewarding. With so many letters at your disposal, the real questions is which words you can not form. So nope, increasing the board’s size was not an option.

The solution came to me, when I remembered a game I played a lot in the very early days of the App Store. Unfortunately, I don’t remember its name but it played a bit like Bejeweled, only that the tiles where triangles and you had to create hexagons of one color in order to score (Update: @jonclayden helped me out. It was Frenzic!).

I was fairly excited about this idea. A word game based on triangular tiles felt more original anyway and I was optimistic that it would solve the problems the 5×5 approach had. So, I spent another day to create another prototype.

Now, this looks much more like Letteral, even the colors were already there. Summarizing the rules:

  • Two Players take turns finding words
  • All letters in a word turn the player’s color
  • Additionally, if a letter was used in a word, it becomes blocked for one turn. So the other player won’t be able to use it in his next turn
  • When you manage to take over all six letters in a hexagon, you’ll score a point
  • After scoring, all letters in the hexagon are replaced by new ones


I immediately noticed that the gameplay now felt pretty good and that it was more challenging to score. There are eight scoring opportunities (aka. hexagons) on the board and they now require six letters. While it is still possible to fill a whole hexagon in one turn (by now, there is even an achievement for it) it is much more difficult and subsequently really feels rewarding. Another positive side effect of the new board is the fact, that some letters are more important than others now: While the letters on the outer borders of the board are only part of one hexagon, the letters in the center belong to up to three hexagons, making them strategically more valuable.

Thus, the basic concept was in place and it stayed unchanged until today. The gameplay wasn’t final though: Only after a lot of testing and many subtle changes it became what is is today. I’ll talk more about the fine tuning next week!