I'm working on a puzzle game. I thought of a core mechanic, implemented it, tuned it and now I have something I'm happy with.

I'd really like to build a few levels before I work on stuff like UI and sound, however, I'm not sure if I should sort out some other things first:


The game has a story (as opposed to many other puzzle games). The thing is, I just have a basic idea of what it should be, nothing detailed and nothing's written down.

Should I work out the story in some detail first? Figure out what arc I want to tell in the first 5 levels before I build them? Or is the story somewhat orthogonal in puzzle games and I can just go ahead and build some levels?


I've thought of a rather minimal set of gameplay elements, i.e. objects you can interact with, to evaluate the core mechanic.

I'd really like to keep things minimal and see how far I get with just those, but is that sane? Should I think long and hard about what elements I need upfront or should I add them when I feel that something's missing in a level I'm building?

Bottom line: What should I do before designing the first few levels?

  • 1
    \$\begingroup\$ This is really opinion based. I'd say whatever you have the most fun doing. \$\endgroup\$ Aug 29 '13 at 15:55
  • \$\begingroup\$ As there is no best answer to this question, it will probably get closed soon. Is there any reason why you're putting off level design? It's an iterative process after all. You need practice to make good levels. \$\endgroup\$
    – ThatOneGuy
    Aug 29 '13 at 15:55
  • \$\begingroup\$ @user1895420 I think I'm trying to avoid it. I'm a programmer by trade and while I enjoy high level game design, detailed design freaks me out a bit. So I think I might have to push myself a little here... \$\endgroup\$
    – futlib
    Aug 29 '13 at 18:07
  • \$\begingroup\$ @futlib Remember that you don't have to use everything that you make (it's actually better to discard your worst levels). Maybe realising that takes the edge off a bit. If all else fails you can always make a level editor and pay someone else to do it. \$\endgroup\$
    – ThatOneGuy
    Aug 29 '13 at 18:57
  • \$\begingroup\$ @user1895420 I did make a level editor (super low hanging procrastination fruit!) but I don't really want to outsource this. I want to make at least the key levels myself, even if I have to step way out of my comfort zone. What I'm trying right now is to make small just-for-fun throwaway levels to get things going. That works kinda well. \$\endgroup\$
    – futlib
    Aug 29 '13 at 21:32

As soon as possible. Making levels will not only give you much needed practice at making levels, it will give you a good idea of if/how your game play mechanic might need further changing.

A rough storyline and the core mechanics implemented should be all you need to begin making your levels. Creating the levels will be a great interactive way to further develop the story and learn what you want from the levels.

Typically, when creating levels, designers will start somewhere in the middle. You don't want the rough "still learning" levels to be the first thing the player sees, nor do you want it to be the last thing the player sees. So your rough storyline should be somewhere in the middle of your story. Create the levels for that part first, then expand out from there.

  • \$\begingroup\$ I'd say make some practice levels first, your first levels are always worse, because you're still getting used to the tools. Or, if story is no problem, discard your worst levels. \$\endgroup\$
    – ThatOneGuy
    Aug 29 '13 at 18:56
  • 2
    \$\begingroup\$ If you don't want to make explicit throwaway levels for practice I'd suggest spreading your first levels out instead of creating a weak middle. I think the latter would be more likely to make players get annoyed and bail than an occasional poor level. \$\endgroup\$ Aug 29 '13 at 19:46

I recall Extra Credits covering some of this in an episode on indie games.

I'm not sure if I'd agree that you should start on levels as soon as possible. Maybe a core few - what you feel are going to be your most fun levels. But as long as you feel you can vary your puzzles in enough imaginative ways, I'd say it's not a bad thing to have 70% of your levels un-made some time into development.

The obvious issue you're trying to avoid (and nevertheless, will fall into by definition of "iterative process") is that the work you do on levels will need to be re-done when your mechanics change. If you have a core few levels, that's not a big task. If you have 90% of your levels done when you realize your character doesn't jump high enough, then you've got a lot of work ahead. To try to put it simply, game design is an iterative process that can take a long time to "find the fun". Once you have found the fun, you can being the arduous task of nailing the details in place, and making a suite of levels that make use of the mechanics.

Minimal gameplay elements sounds like a great idea - simplicity makes these things very easy to understand. However, around the time that testers would begin to get bored, you should probably think about "what interesting twist might breathe new life into this?"

It's hard for me to give much advice on the story without knowing what it is, but I will say you should try to be careful about letting it influence your game mechanics too much. The end goal, easier said than done, is story elements that fit well with their accompanied twist to the gameplay (you found a lost, distressed girl - maybe the two of you can work together through the next 18 puzzles to find your way home!) If that becomes too difficult of a goal, I don't think too many people would blame you for amounting your story to simple cutscenes between level-blocks.


Not the answer you're looking for? Browse other questions tagged or ask your own question.