Hey Everyone, I’m Amy, author of The Witch Girls.
I was incredibly surprised to come 8th in the comp—my aim had been 42nd, just to get into the top half—so just thank you so much to everyone who ranked the game.
I didn’t respond to any of the reviews during the judging period as I wasn’t sure about the etiquette regarding interacting with them (in other writing spaces reviews are for readers, not authors) but based on what I’ve seen, I assumed wrongly there.
So, thank you so much to everyone who took the time to play and reviewed the game on this forum: @Sequitur, @Kozelek, @lpsmith. @DeusIrae, @silvercyborg, @alyshkalia, @n-n, @mathbrush. Your comments were really appreciated.
With that, here’s a bit of an essay on some of the things I was thinking about when I made The Witch Girls.
Truth and Patience: A The Witch Girls Post-Mortem
On the Fifteenth of August, I wrote in my writing journal: “Minus bugs, The Witch Girls is done. Like all projects, it limped over the finish line with little love afforded to it.” I have a propensity towards the dramatic and pessimistic in these diary entries, but honestly, this wasn’t true at all! The creation of The Witch Girls was an exercise in truth and patience, and what are those if not love?
The game is truthful in its dealings with the very real social landscape of teenagers in noughties Scotland (magic notwithstanding), and the less-than-palatable topics that that raises, and patient in that, unless I am paid, I have no desire to work on a singular project for more than three months at a time, and the overall development time for The Witch Girls was eight months.
Being Mariana Enriquez
When I began writing The Witch Girls, I was on a real Latin-American horror kick. I’d picked up The Dangers of Smoking in Bed on a whim at my local bookshop, and it opened a new world for me. I have never been to Argentina, yet Mariana Enriquez made me feel like I was walking the terror-tinged streets of Buenos Aires.
I wanted to write something similar, but I couldn’t invoke a country I’d never been to or a culture I’d never participated in. Instead, I thought about what Enriquez was doing with her sense of place and tried to mimic that with what I do know: rural Scotland and gift-shop Wicca.
The feel of Enriquez’s writing is conversational and simple, and this is what makes it effective. It represents the kind of ordinary people she writes about.
“Juancho was drunk that day. He was getting belligerent as he walked up and down the sidewalk, although by that point no one in the neighbourhood felt threatened, or even unsettled, by his drunken antics.” (The Dangers of Smoking In Bed, The Cart, P27)
“You asked the witch girls for help, first. They’d built a hut, or commandeered it—you never really knew—and hung out there every Wednesday after school, at least when they didn’t have choir or orchestra or detention.” (The Witch Girls, Screen 1)
This style of writing wasn’t alien to me—“simple but evocative” is a phrase that has been applied to my work in the past—but I felt like there was a deliberate simplification here that I struggled with.
“I’m feeling unsure about its prose style. It’s detached, simple, and at a distance in an attempt to write like Mariana Enriquez, but I worry it looks like lack of sophistication over deliberate choice.” (My writing journal, 13/02/25)
So, I was very pleased to see that reviewers understood what I was attempting here!
@Sequitur in his review described my writing as “confident, straightforward-but-still-evocative horror”.
On top of her style being representative of her characters, Enriquez uses specific detail to bring Buenos Aires to life. There’s frequent reference to drinking mate, to Catholic shrines, to markets and late-night ice-cream parlours in her work, and I wanted to do the same. In the first screen alone, I reference Golddigga, N64, Green Day and the North Sea.
References are a tricky one. I sometimes find them cringey, especially in contemporary works—will this feel dated in ten years? Here, I was making references to things that were relevant almost twenty years ago, and the curation was personal—much of The Witch Girls is autobiographical—so I hope that it feels like an attempt to bring a time and place to life, and not a carousel of “hey, remember emo music?”.
I’m confident the sense of place was achieved: @n-n in their review wrote that The Witch Girls has “a great sense of place and lots of fun pop culture references”, and @Sequitur mentioned that the game is “evidently drawn from authentic experiences growing up in that time and place”, but given Kinetic Mouse Car wrote that “the author captures a sense of nostalgia by name-dropping music, clothing brands, and pop culture”, I wonder if I romanticise the era too much. My aim was to be honest about that time in my life, not to hark back to how cool things were before smartphones.
Truth and Danger
With honesty in mind, I didn’t pull my punches when representing the casual homophobia and sexism of the time. I’ve got older boys asking thirteen-year-old girls about sexual acts, fears about future children turning out gay, and underage sex in there.
I once took a playwrighting class, and our teacher told me after reading my first, pulled back and tame, play, that he didn’t know me very well, but he thought I had more “danger” in me. That led to me writing a play that opened with a fourteen-year-old girl monologuing about a severed penis, and I think The Witch Girls is a spiritual successor to that play given its similar themes of teenagehood, magic and the sea.
This one was for a competition, however, not to please a teacher who was trying to push me, and I was concerned that readers would believe I condoned the subjects I was writing about.
The original draft of the game was written from an adult perspective, reflecting on the events of the protagonist’s teenage years, but this pulled the player out of the moment and ran against what I was trying to do with time and place, and I eventually realised that this writing style was self-defence. On the 19th of June, I wrote in my journal: “I fear the modern reflections are attempts to say ‘but I don’t think this’”.
I pulled that perspective out entirely and focused the game on the moment-to-moment, and I think it feels so much better for that.
And then, I guess I just stopped feeling fear at all because I wrote the jellyfish scene. I felt sick while I was writing it. I wondered if I should be writing it, if it was too gross, but what the aforementioned play, and now The Witch Girls has taught me is that the things I think I shouldn’t be writing are the exact things I should be writing. Those are the things that are the most truthful, the things that could only come from me.
“The project was a little controversial, and I’m proud of myself for really pushing the danger and refusing to self-censor. The best part is the part I made myself feel sick writing.” (My writing journal, 09/08/25)
Multiple reviews mentioned that I’d taken a subject—teenage witchcraft—that can be cliche, but built a unique take on it. I think this working into danger is how I managed that. I went with what was in my head, no matter how strange it seemed.
“Young girl’s burgeoning sexuality” is well-trodden ground in the genre, and can sometimes veer into trite territory, but Witch Girls never falls into that trap. It feels incredibly grounded in a time and stage of life, and the multiple endings lend a creepy ambiguity to the proceedings.” (@silvercyborg’s 2025 IFComp Thoughts)
In the end, I did get shocked reviews, but they were all positive, and I revelled in them! My favourite review comes from @DeusIrae: “I will be shocked if, at any point between now and mid-October, any game makes me mutter “what the fuck” under my breath half as many times as The Witch Girls”. As a horror writer, this is exactly the kind of response I should be after, and I’m happy to try and lean into this.
Patience and Cutting
There’s an oft-quoted phrase “writing is rewriting”. I hate rewriting! I hate working on anything for an extended period of time! I get to about the twenty-five percent mark, and the interesting questions are answered and all I’m left with is writing and ending and endlessly polishing. I’m impatient to get on to the next thing, to throw this out into the world and do something else.
I resisted this urge with The Witch Girls and the piece is so much better for it. I wouldn’t have taken out the point of view that wasn’t working if I had just called the first draft done.
After my first round of beta testing I took a month’s break from the project. The aim here was to come back to it with fresh eyes, and it really did help. I felt less fatigued on the project and was more prepared to make larger rewrites. Taking a break was difficult for me, as I tend to focus pretty heavily on one thing at a time, but I think I need to build up tolerance for letting projects take as long as they need to take.
There were a couple of concessions I had to make after the second round of beta-testing that didn’t feel good to implement but I know made the game better. The original version of the game had all of the links a different shade of brown. They looked lovely and brought to mind the Jane Nor bags that were all the rage at the time, but players had no way of distinguishing between cycling links and links that moved the story on. I don’t really like the new pink and green links, but they make the design clearer and that is ultimately more important than aesthetics.
I made another design change to the choices at this stage in order to fix a major bug. Players complained that despite choosing to take the roses, they were ending up on the jellyfish route. It turned out that players were cycling the link after the text reveal about the roses, which meant the variable for having collected the them was being unset despite the text telling them they had collected them.
I fixed this by disabling the cycle after the text reveal, which made the on-page choices feel a little more fixed and less decorative. I guess this was a little muddy design wise—are in line choices branching choices or are they aesthetic? It’s both, which means it isn’t clear to players when their choices matter. I tend to prefer that approach, especially when one is trying to make the player think there is more branching than really exists, but I think for The Witch Girls it was important for the player to understand which choices had mechanical weight.
Structure
While there were a significant amount of rewrites on the project, the structure of the game never really changed. I didn’t put much thought into the idea of a route-based structure, it just sort of happened, but I’d wanted to write something like this for a while.
I did however have to put some thought into how the ingredient collection contributed to branching. I knew from the beginning that I wanted the ingredients to determine the player’s path. In my journal I found a design doc for a “word inventory” which involved picking words from the main prose to add to an inventory that could then be used to construct spells—very ambitious, and very interesting! I don’t remember why this idea died (probably scope), but I might pick it up again in future.
Anyway, I digress! My main consideration was how many routes I should have and how those should relate to the ingredients. I worked out the number of routes that should exist based on the ingredient permutations, but found there were too many to work well.
In the end, four combinations lead to the jellyfish ending, with only one each leading to the other two.
- Null, null: Jellyfish
- Null, rose: Jellyfish
- Null, oil: Jellyfish
- Null, juice: Jellyfish
- Juice, rose: Zombie
- Oil, rose: Cuckoo
One would presume based on this distribution that the jellyfish is the most likely route to be encountered on a first run, however, I had a couple of tricks that, I hope, made the playing field a little more even. As a designer, I wanted the player to encounter the routes in an order that felt right to them, even though personally I think Zombie, Jellyfish, Cuckoo is the most satisfying.
The first trick is nothing I’ve done and simply a presumption about player behaviour. Most players like to “win”. There might be more variance in the interactive fiction community than in the wider games space, but given the player is told to construct the spell according to a set recipe, I presume most players will attempt to do this, and therefore get the cuckoo route on their first run because they’ve followed the rules.
I think the cuckoo route is the weakest, so didn’t want it to feel like the default first route. To try and nudge players towards the zombie route, I made sure that the lemon juice option was the first in the list of cycling links when it appeared. I’d hoped that it being present on the page from the beginning would have it linger in the player’s mind, and being first it would feel default.
I consider the full story to be every route. I don’t see one path through the game as satisfying, and as DeusIrea wrote, the routes are “in dialogue with each other”. This was why I decided to create the flowchart. A lot of the reviews praised this, which was nice. I wanted players to see everything, and I didn’t want that to be difficult. I didn’t want any friction between the player and the game unless said friction was enhancing the themes of the story, which was why the flowchart was unavailable until one ending had been reached.
On that first playthrough, I wanted the player to feel the bewilderment that my protagonist feels as she staggers through the teenage social minefield, and the ability to see the structure of the narrative and control its outcomes runs contrary to that. Once they’ve had a dose of confusion, I let them off the leash and reveal the artifice.
I wondered if the flowchart would reveal that most of the choices have no branching effect on the narrative, but if it did, I didn’t receive any complains about that. I’m a firm believer in choice as a tool for player expression, as a vehicle for the player to fill-in-the-gaps between the material text and their imagined experience of it, and most of the choices in The Witch Girls were that.
What’s Next?
I’m getting married next year, so I’m not sure IFComp 2026 is on the cards, but if not I’d imagine ‘27 will be.
I’ve started working on some Sugarcube prototypes, given the answer to a lot of “how to do X in Harlowe” questions seemed to be “use Sugarcube”! I’m already loving the widgets feature and having a much better time with layout than I was in Harlowe.
The prototypes have been dialogue heavy, as I love writing dialogue and didn’t get to do much of it with The Witch Girls. I’m considering adapting the play mentioned in my danger section into an interactive fiction game, so that may be my next project.
If I did that, it would very much play into the same themes as I’ve been working with here, so could be nice as a sister work, but I’ve also been thinking a lot about cyberpunk and death games recently, so I might venture into completely different territory! Time will tell.
Thank you for reading my little essay, apologies if it was a little navel gazey, but that’s sort of what a post-mortem is for, right? If anyone has any questions about my process working on TWG I’d be happy to answer 
.
Almost forgot to add–The Witch Girls is now available on itch.io for those who might be caught by the geoblock.
