Minimum Viable Product (MVP)

A place to discuss things that aren't specific to any one creator or game.
Forum rules
Ren'Py specific questions should be posted in the Ren'Py Questions and Annoucements forum, not here.
Post Reply
User avatar
Posts: 60
Joined: Tue Jul 08, 2014 3:28 am

Minimum Viable Product (MVP)

#1 Post by inkbrush » Fri Jun 10, 2016 12:58 am

Hello Lemmasoftians!

Today I ran across a video talking about how new game developers should start by making something the size of a minimum viable product--let's call it a MVP from here on.

I'd highly suggest watching the video first

After watching the video, I decided that I wanted to make a MVP (as one of the many creators who struggles to finish their ideas, of course). But, trying to figure out what the MVP of a visual novel would be is definitely stumping me. A visual novel is built around story and choices--it's not the traditional gameplay of the popular games out there right now. (Of course, there's RPG's with battle systems. But I'm more thinking about story and option focused visual novels.)

Do you think that a MVP would entail a story simple enough with a single choice? Or would it entail something more?

I'd love to hear your thoughts,
Thank you!

User avatar
Posts: 153
Joined: Sun Jun 19, 2011 1:16 am
Completed: Cursed Lands, Trial by Fire
Projects: Necrobarista, Idol Manager
Organization: Route 59 Games
Tumblr: kuiperblog

Re: Minimum Viable Product (MVP)

#2 Post by Kuiper » Fri Jun 10, 2016 2:29 am

In the world of software, the idea of MVP is to test/validate something. The Extra Credits video you linked discusses using an MVP to test/validate core gameplay mechanics (like running/jumping in Mario). If you're making an app, you want to validate the market (are people going to actually download my app and use it) and whether your app actually solves the problem that it set out to solve.

What are you trying to validate?

Maybe you're trying to validate mechanics, which is what the EC video talks about, but there usually isn't much of a need for that in visual novels since a lot of them tend to be pretty rudimentary and trivial to implement. (For example, the whole "click to advance" thing is pretty much handed to you if you're using Ren'py, and doing things like having menus, and storing player choices in variables to refer to later, are also very simple to implement.) However, if you're doing anything mechanically interesting beyond the "basics," you should probably code it to validate. Things like: point-and-click adventure style "object hunts," or trying to do something like mapping menu items onto icons that are part of an image (like a navigation map where someone can click on a location to trigger an event at that location). Often this includes things like providing feedback to the user (like giving some kind of visual indication if they try to make an invalid selection).

Odds are you're trying to validate something that lies outside of the mechanics (odds are that those have already been validated by countless other existing visual novels, and the thing that sets your visual novel apart from the others will be its content, which includes images, writing, sound, and all the other assets that go into the visual novel).

If you're trying to validate an aesthetic, then a single frame might be enough. Just give me a background, two character sprites, a UI, and a line of dialog. You don't need multiple poses for the characters, you don't need more than one line of dialog, you don't need multiple backgrounds. Show me something that looks like a screenshot, and I can tell just from that single frame whether or not I like the art style.

If you're trying to validate the writing, then just a script may be enough. In fact, we may not even need an entire script. If we're just trying to judge writing quality, most of us can tell within a few pages whether the writing is good or not. Now, if you're trying to validate multiple parts of what falls under the broad umbrella of writing (things like plot, character development, and so on), we might need something longer.

Sometimes you're trying to validate something a bit more intangible. For example, you can examine and evaluate things like writing, images, and music all independently, but oftentimes what you're concerned with is the collective impact that all of these elements have when experienced together. This is especially true of sound: a lot of times, "sound design" is as much about how the sound "meshes" with the other elements as it is about the sound itself. For example, a horror stinger heard in a vacuum might just feel weird, but heard at exactly the right moment, it could give you goosebumps. Ditto for other "small" sound effects like footsteps. In this case, you'll likely want to produce a "vertical slice," which is basically a small piece of the game that is designed to give a production-quality experience, but for a very short duration. To do this, you'll want to go high on quality, low on quantity. This could be something as simple as a brief dialog sequence with two characters talking to each other, or a short "action" sequence that is just long enough to produce a certain desired emotional state. This could be as brief as a few minutes.

In the games industry, a "vertical slice" is often the proof of concept for stakeholders outside of the development team. The "prototype" (or "minimum viable product," as the EC video describes) is what you show to the people on the development team to get the idea for the mechanics across. The vertical slice is often what you show to producers or investors to get them to fund your project. It's the kind of thing that you show off as a 10-minute E3 demo when the actual game (which will be a 10+ hour experience) is still years from being finished and released.

There are a lot of game jam games (including VN game jams like NaNoRenO) that serve as good examples of trying to validate a concept.
Necrobarista - serve coffee to the living and the dead
Idol Manager - experience the glamour and dangers of the pop idol industry
Cursed Lands - a mix of high fantasy and gothic horror

User avatar
Posts: 119
Joined: Thu Jul 30, 2015 5:11 pm
Projects: Shadow City, Scribbles!
Tumblr: asktheprismguard

Re: Minimum Viable Product (MVP)

#3 Post by Parataxis » Fri Jun 10, 2016 8:30 am

Funny story I just watch that video the other day apropos of nothing.

By the video's definition, I think for your average visual novel especially ones made with Ren'Py, the minimum viable product already exists. The minimum viable product is basically the Question or the Ren'Py tutorial. Each one covers every type of "game play" in your standard visual novel stripped of things like specific character design or Gui or music choice.

But! Sometimes people add structures to their games to facilitate further gameplay like adding puzzles or RPG elements and this I would presume would change the minimum viable product for testing. I myself made a MVP to test out a battle system for my game focused on choice timing.

User avatar
Miko-Class Veteran
Posts: 540
Joined: Sat Oct 25, 2014 6:20 pm
Completed: The Unknowns Saga series
Projects: The Unknown Saga series
Tumblr: bigattck
Deviantart: bigattck
Skype: bigattck firecat
Soundcloud: bigattck-firecat

Re: Minimum Viable Product (MVP)

#4 Post by firecat » Fri Jun 10, 2016 8:45 am

for vn to be so basic you will only need text and background, thats all you need to build a vn. take "Narcisu" as an example, you only see background and text thats what a vn is about. music and voice are extra stuff that is not needed but helps, text and background is the most important thing for a vn. you can even build a vn with just a blog and it will still be a vn.

no the Question story has menu system, characters, music and engine. remove all this to only have text and background, which is again the most important thing.


special thanks to and iichan_lolbot

User avatar
Posts: 85
Joined: Sun May 15, 2016 4:47 am
Projects: The Lost Sheep Of Purgatory 「煉獄の亡羊」
Organization: House Of Duematrix
Tumblr: HouseOfDuematrix
Location: A Dark Cliff Overlooking The Moon

Re: Minimum Viable Product (MVP)

#5 Post by hoihoisoi » Fri Jun 10, 2016 12:08 pm

From the video, he had cut out click and point adventure games since it was mostly content and not much mechanics, I guess Visual Novels fit into that category as well. Unless your game has gameplay elements such as mini games then I don't really think you need to develop an MVP.

However, something which may need an MVP is the script to your visual novel. Coming up with a complicated script with multiple branching paths from the get go is simply not easy to do at all, with so many variables to keep track of, a writer would quite often get confused if not frustrated that they see little progress in their writing. I'd say, create an MVP with only one story line first, read through it, ascertain the strengths and weaknesses, correct it and make it better (From a writing and plot standpoint) and then slowly add in the multiple branching paths until you've reach the level of complexity you're trying to achieve. I guess, this lets you test the strength of the plot and the idea behind the whole script in a way. That's just my two cents on the matter though X(
Current Active Project:
[Mystery, Comedy, Drama]
[Kinetic Novel]

Check it out by clicking on the siggy above!

User avatar
Posts: 99
Joined: Thu Oct 03, 2013 12:23 am

Re: Minimum Viable Product (MVP)

#6 Post by Suika » Sun Jun 12, 2016 12:55 am

I agree with everything said so far about MVP in a general sense, but one aspect of the video that stuck out to me was the definition of MVP. The video defined it as "the smallest thing you could possibly make that will still give you useful data once released." I think there's a huge disconnect here between the different communities of game development, and the different scales of each. I also think that disconnect echoes in the range of definitions of MVP that are out there. I've also heard it defined as:
  • The smallest releasable product
  • The smallest thing that can still be recognized as your game
  • The smallest thing that could serve as a representative demo of your game
Personally I like the first definition. Whereas the other definitions seem to focus on data that will be useful for future projects, the first feels like it hones in on the most important goal for a small indie developer (or at least for me): release a game. I don't want my MVP to be a mediocre version of my game, I want it to be my actual game. So instead of taking out the cool features that separate your game from everything else (like blocks and Bowser in SMB), focus on reducing the scale of the game.

For instance:
I wouldn't release SMB without Bowser, so he stays. I wouldn't release it without a point system, so blocks stay. But maybe coins are easier to implement than breakable blocks, so I'll replace the blocks with coins for now. I'd like to have enemies, but the game would still be enjoyable without them (bar Bowser), so I'll shelve them for now, and keep them in reserve in case I have some extra resources. Peach is the entire motivation for the story concept, so she stays, but she's only important in the lore, so I won't make a sprite for her.

You could continue doing that for a while, and when you're done, there's still probably some features that could be removed or replaced with something else that accomplishes the same goal, but reduces the cost. For instance, in a romance VN your initial concept might be a love story with 4 romance options each with their own path. The goal of MVP is to reduce that as much as possible. If you decide that you absolutely need 4 romance options, then try to combine portions of each route, or reduce the length of the routes. Cut the number of expressions required, or reduce the number of options/choices the player makes.

Ultimately, you might end up with a MVP that's still large. Some MVP philosophies would say that means you need to cut more features - cut three of the romance options even though it compromises your vision. I disagree with that though. Once you cut a feature that is critical to the game you have in mind, it becomes a different game. So instead of mutilating the core concept, take some time to decide if you can realistically pursue it or not. If you can't, then it might be better to focus on a new idea entirely, and saving your brain child for a future time when you can realize that vision. After all, just like RTS games are harder to make than 2D platformers, some VN concepts will be more challenging than others. I think MVP is most useful for VNs if we focus on the minimal product that we would release, focusing on the content/art/story over the gameplay (assuming you're making a content-focused VN).

User avatar
Posts: 388
Joined: Fri May 03, 2013 6:22 pm
Completed: EVOLVEd: Echoes of the Codex War. [ASH] The Seeds of Destruction
Projects: One Night of [SNOW], Stained with Magic
Deviantart: truefaiterman
Location: Spain, and without bullfighting!

Re: Minimum Viable Product (MVP)

#7 Post by truefaiterman » Sun Jun 12, 2016 7:13 am

As far as I see it, the MVP looks like a prototype with the core gameplay mechanics. Visual Novels per se don't have too much of those, so I think it would be making a VERY short story (well, even using Lorem Ipsum) that showcases every message box you're using, along with every kind of graphic and sound you want (A sprite, a BG, a CG, a voice clip, music track, SFX, even if they're just placeholders).

Visual novels mostly depend on their content, so... well, the closest thing to a prototype should be that stuff which makes sure nothing is broken, IMO.
Artist and voice actor, trying to actually write stuff.


ArtStation portfolio
Youtube channel

Recent finished projects:

Image Image

User avatar
Posts: 66
Joined: Mon Jun 20, 2011 9:02 pm
Location: San Francisco

Re: Minimum Viable Product (MVP)

#8 Post by psy_wombats » Sun Jun 12, 2016 9:48 am

An MVP for a VN would be testing that the mechanics of the story worked, as most of the time those are the mechanics of the game. There are some exceptions for games with novel control mechanics (like using your phone to make choices, RPG battles, branches based on the player solving puzzles, etc) that could probably be tested in isolation without a script. But as far as figuring out whether a VN concept is worth pursuing or not, nothing beats a rough first draft.
truefaiterman wrote:As far as I see it, the MVP looks like a prototype with the core gameplay mechanics. Visual Novels per se don't have too much of those, so I think it would be making a VERY short story (well, even using Lorem Ipsum) that showcases every message box you're using, along with every kind of graphic and sound you want (A sprite, a BG, a CG, a voice clip, music track, SFX, even if they're just placeholders).

Visual novels mostly depend on their content, so... well, the closest thing to a prototype should be that stuff which makes sure nothing is broken, IMO.
I'd really consider all that stuff polish. Removing the message boxes and sfx from a VN wouldn't make it unfun, provided it has a good story. And sprucing up a trash script with a bunch of CG isn't going to really help things either.
@psy_wombats - Hobby dev of VNs, RPGs, roguelikes, and other junk... professional phone game dev :/

Post Reply

Who is online

Users browsing this forum: Google [Bot]