On the Matter of Words Per Playthrough

This is an expansion of a discussion that started in the thread for my current WIP, Quite App-Parent. Currently it has a number of scenes that are randomly selected, especially in the early years. For example, in the child’s first year you see two scenes out of five available, and then two out of three or four in many of the others. Some are mandatory, and others are determined by your stats or choices, but at least a dozen won’t be seen on a playthrough (and that’s just so far). A lot of my (completely awesome, I do want to point out) readers and testers have expressed interest in having all the random scenes made available each time you go through. This would beef up playthrough length considerably. But I know the sweet spot in word percentage seen per playthrough is normally considered 20-40%, and this would shoot that to pieces (it’s 31% right now, but would likely double or more with the addition of all randomized scenes).

My inclination upon hearing this was to create a New Story/Game+ option, where you could choose to go through it this way once you’d completed it one time (or choose to go through again as it had been before). Other suggestions include just having that option available at the very beginning instead of only after completing the story, or just avoiding all randomness entirely and unlocking all of those scenes no matter what. I’m trying to determine the value of offering more content versus insuring replay allure.

All that to say this: is it a big deal to stay in that 20-40% sweet spot? Commercially, critically, how much does any of this matter? I want input from everyone, but would especially appreciate some posts from authors who have firsthand reports about how this has positively or negatively impacted you (especially if you have multiple stories to compare and contrast), or even CoG staff members. I’ve also included a poll about my situation in particular as well as the more general question regarding the validity of higher wordcounts. Each potential solution has a lot of valid criticisms and opportunities, and I wanted to get answers from a larger sample size than just my own story’s thread if possible.

  • The sweet spot is sweet for a reason: keep in the Golden Range no matter what
  • They’re more of guidelines than rules, so say sod it all and lengthen away!
  • Pie

0 voters

  • Keep random scenes random!
  • Derandomize everything, make all non-choice-determined scenes available!
  • Include the New Game + option regarding randomness, and make 'em work for it!
  • Give the choice between random or not at the beginning of the game
  • Pie

0 voters


As an author of 4 hosted games and 4 WIPs, I can confidently say that most readers don’t give a solitary shit about replay ability. In fact, more often than not, they seem to be annoyed by it because it drastically decreases the all-important playthrough length.

What most readers seem to want is to be fooled into thinking their choices matter with varied, interesting, and well written fake choices. Here’s why I say all of this:

  • My game, The Magician’s Burden, is 230k words with a playthrough length of 60k. That means the replay value is very high, and I considered this to be an asset. In the markets, this turned out to be the opposite. I’ve had a very significant amount of bad reviews saying that certain ROs are bland, lifeless, and not there, while others are well developed and engaging.

This is because of the replay value: depending on your choices, you could see all 4 ROs an equal amount, or you might see certain ones very little in one playthrough and then a lot in another. But instead of viewing this as a positive thing, lots of readers simply assumed that various characters had little screentime at all. So all of that extra work just amounted to a big drop in sales and ratings.

I will not be doing that design again, where entire mini scenes with characters are hidden behind choices that cannot all be seen in one playthrough, or even a few playthroughs. There’s a reason why CoG has recommended against this system. The only exception I will make is for romance routes, like date and sex scenes.

  • My infamous WIP, Mass Mother Murderer will be roughly 220k when finished. With a projected playthrough length of around 90k, it will be my most linear work yet, aside from Foundation of Nightmares (which happened to be skewered for its blatant linearity.)

Despite this, it has recieved no complaints that I can remember for being linear, and it’s my most popular story by huge margins. also consider that this MC is one who has a defined personality, defined motivations, and so much agency that they make countless important decisions of their own accord, with the reader either just deciding how these things are done, or not giving input at all. But, because of the engaging writing, plot, and characters, the reception for MMM has still been overwhelmingly positive. And this isn’t just me blowing smoke up my own ass, either. I’m going off of my beta testers’ feedback and comments.

All that said, I am certainly not condemning huge amounts of replay ability with various mini scenes. Whatever the author wants to do is fine, and several games have had huge success with this approach, like Zombie Exodus.

All I’m saying is that, in general, I think it pays more dividends to both the author and the readers if the story is much longer at the price of being a bit more linear.

Not TOO linear, of course. I still enjoy giving enough options so that playthroughs can have noticeable differences and the reader can play as several different kinds of MCs, but I just think that, at a certain point, replay ability starts to become less and less worth the very considerable effort it takes the author.


Mmm… Pie. :pie:

Anyways, another alternative is to have the random scenes… Well… Not be random anymore. Like, instead of having a scene that appears randomly, you could include a choice that would determine which of the scenes the player sees. This would keep the replay value, since players can just go back and make other choices if they wanted to see the other scenes, and there wouldn’t be the added annoyance of having the same random scenes pop up over and over again.

Also, I didn’t vote in the second poll (well, I voted for pie) because, ultimately, you should just pick whichever choice works best for you. Some people will want it one way and some people will want it another way, but at the end of the day, you’re the author, so you should just make the game however you want it to be. :grin:


I just wanted to comment here: ZE and ZE:SH have major breadth but they really do not gate content for replayability.

As an example: There is a huge fight between the zombies and survivors in ZE:SH and there is a lot of breadth involved in it. Leaders of the survivors may be different characters, groups that the survivors are split into can vary and how the actual fight plays out changes with almost every decision made during the scene.

None of this is gated - this scene happens for everyone. The replayability is in the breadth within the choice structure.

There is a major difference between using gates and one-way traps to differentiate content and using breadth to differentiate content.

That isn’t to say breadth is the bestest thing since sliced bread because if your breadth isn’t structured well, it can lead to major issues as well.


From a personal standpoint, I want to say replayability is ideal, from a ratings viewpoint longer is better… but if there’s no replayability at all, some reviews will go back into negative. Don’t make it highly branching if you want good reviews unless you’re also planning on it being very long as well. Golden rule seems to be minimum ave playthrough length of 20k for a medium sized game, 30k is better if you have the words for it.

(Too linear is usually considered the lesser sin compared to less replayability if you have to choose. I do agree with Samuel on that one and will probably never write a really highly branching game like my first one again. In saying that, if there is no replayability, you may hit a few complaints from those that actually do replay. I’m all for having a greyed out choice at the start for choosing scenes that says it’s accessable once played through once to make it clear there’s replayability as long as your wordcount is otherwise long enough. Otherwise lengthen away and reduce replayability.)


My first two games, Trial of the Demon Hunter and Captive of Fortune, had playthrough lengths of 15/85k and 25/140k respectively, and they were both understandably railed on for being short.

I’ve had practically no complaints about length with my other two HGs, Foundation of Nightmares and The Magician’s Burden, which had playthrough lengths of 40/100k and 60/230k respectively. On the other hand, I haven’t had too many people praise the lengths, either. This leads me to believe that 40k+ is likely the length that people start to think is not short, though this is also surely affected by how much the total wordcount is advertised as.

Mass Mother Murderer will be around 90/220k per playthrough, so I anticipate lots of good “long game” reviews on that one.


Speaking purely for myself, I don’t tend to replay games very differently from how I played them the first time. I play through a game once to get an idea of everything, and then play through again perfecting my ideal version of events with the fore knowledge play through one gave me. (and then, usually instead of playing it more, I just code dive and read all the words written in the game to see the other content)

That being said, I do like if it seems like there could have been other outcomes if I chose differently. I definitely prefer longer reads, so I sometimes do pass on buying if something is listed at 250k words but only an average of 40k per play through for example. (I don’t leave reviews unless its a five star game and I really loved it, so no negative reviews from me for that though)

In the case you’re asking about in the OP, again, personal preference here, but I don’t like random events in choice script games, so for me de-randomizing everything and making all non-choice-determined scenes available would be ideal.


I’ll throw something different in the mix, rather than the actual wordcount of playthroughs, consider one’s progress through a playthrough.

We’ve talked for a while about what it would take/mean/look like to implement a progress bar on a game. Now one easy way to do that is simply to locate a player chapter by chapter. A little piece of code that tells the display you’re on dot number 6 out of 10 dots, representing chapters. One thing this might do is stave off the player’s complaint of “And then it was just …over!” Well, no, it developed over at least 8 and maybe more like 12 or 14 chapters as games are getting longer and longer.

Now, this would be impossible for games with random chapters, but fairly straightforward for CoG titles.


What do you think about having a progress meter and/or a “paths” stat in the Stats menu?
Something like…

Story percent: 15/100
Paths completed: 27/64

1 Like

Paths would be hard to track wouldn’t they? I mean you could have a variable trigger for each one, but then you run into what is a scene difference and what is a path? I mean if you have a scene where there can be two very different experiences depending on your prior actions… I don’t know, so me it seems grey in what’s what.

Story percent is also problematic (as is the dot points in Mary’s example) for HG’s that have variably length’ed storylines.

Number of endings as a progress bar could work though?

1 Like

The idea is very prototype. It would have to be custom built to a particular story. I do like the idea of bonking the reader over the head with “there is more replay value, don’t leave a 1 star because you think it’s short”


I really like that idea, and that’s why I always put “Chapter One of Thirteen, Chapter Two of Thirteen,” etc, so players know roughly how quickly the game is progressing towards the end. This way, at the very least, they won’t be blindsided and get upset when the finale shows up.


Haha, yeah I tried and failed to do that with achievements last times. A better way would be nice as a bit of a neon sign for “More replay value to be found! Look here!”


You know, I still haven’t gotten any reports that someone has actually beaten my game yet (the bad end doesn’t count)… I don’t know if “words per playthrough” applies to my project. “Time invested” might be more appropriate for mine?

Honestly, achievements really don’t entice me to explore more. At least not immediately.

What I’d like to see, in regards to achievement, is something that is not “plot flags” and something more like secret contents with mysterious hinting toward it. I tell you, it’s more satisfying to even miss them rather than missing an obvious achievement because your stats don’t click, but it’s your first time playing.

I'll take example from recent CoG release Stronghold

There’s an achievement for succeeding in both arena and horse-racing tourney. The fact that you’re actually the one who hold out the tourney and can participate in them, doesn’t help me get the feeling of satisfaction when I unlocked the achievement. I feel like, “well, I got the stats. Let’s do it,” and there you go, achievement unlocked. No suspense.

Decent game, though. CoG’s take on The Great Tournament.

1 Like

Good info for the length vs. breadth debate. It’s clearly not as big a deal as I thought. And Sam’s somewhat mixed results aside, it seems not many have had low replay hit them too hard, and sometimes quite the opposite. My first story has had only positive things said about total length and none at all about playthrough length positive or negative (it’s a bit over 30k), but I think the relatively low price for a 160,000 word story helped in that regard.

As for my personal situation, a little background info I should probably have included for those unfamiliar with it: the chapters/what ChoiceScript refers to as ‘scenes’ are not randomly selected. I probably used some confusing verbiage there, so sorry about that. And all players will know pretty much exactly how long it is, because they go sequentially from Year 1 to Year 18. Well, with an intro and an epilogue, so 20 chapters/scenes in total. What’s currently random (and looking increasingly likely to change in at least some capacity) is what you see in each of those chapters. One time through Year One your baby might have low weight and you deal with too many visitors, another it focuses on their first illness and first steps. Whatever the die roller determines.

1 Like

Coincidentally, I was thinking about doing something like that on the stats screen for Creme de la Creme! I think it’s a good idea, it gives more of a sense of progress and should help with people thinking endings are rushed if they have a better idea of how far they are through.

The average playthrough length for Blood Money is around 40K words. It hasn’t had feedback saying that people thought it was too short, but then the final chapters branch a lot, so perhaps people have enjoyed the amount of replayability.


Depending on a few things, like how much branching is involved and what the actual story needs, I’d say 40k words per read is a good number to aim for. It is, afterall, the bare minimum for what’s classed as a novel if I remember correctly (unless you’re Stephen King. His novellas tend to be as long as some novels . . . )

It’s what I’d aim for anyway, if not longer, depending on the work.


Don’t expect people saying length is good at least not in Google play… Lol entitled people even complain on Tin star…

80% casuals don’t replay and Are entitled to believe due Virtual novels that there is ONLY ONE path and rest are bad.
Last year in a YouTube direct in Spanish Of a table rpg . Complaining about Several Cogs saying can’t be role playing as a game with only an ending can’t be role playing. The guy Had play multiple times to find what he believed it was the real good choice.

I had to explain that In sabres of infinity for instance you could end as a disgrace soldier and that’s a difference that carries over.
People just automatic assuming this is one of the crappier False choice virtual novel trash that are freemium and sell jewels. Probably in my Spanish comrades you have to think mostly the only stuff it is translated is that trash. Choice of dragon was a great exception but my people sadly prefers piracy that buying…

My personal perspective is do what you prefer as by numbers you will probably screwing a chunk of possible fan base.

I love random and I will still ise random events even if controlled the use and numbers.


Something else to consider is that a game may have more than sufficient words, at least when counting, but this can be easily overlooked if the pacing itself is off.

What do I mean? I will use Welcome to Moreytown as an example. The game was wonderful in the start, especially when you design which animal species you come from. The game also did a good job of introducing the NPCs, as well as why you lost your home.

Where it felt lacking was near the end. It honestly felt rushed to me. I’m not saying it didn’t have sufficient word count, but the set-up to it felt a little unsatisfying.

This was also the same gripe I had with Runt of the Litter. I still liked both games, and have gifted it to people, but at the end I didn’t notice the word count so much as the pacing.

That is why when I look at games like Choice of the Deathless or Neighborhood Necromancer, both are on the shorter side, but the pacing felt ‘right’ so I didn’t notice how much content was in there. Of course, this may not be true for other people.