DESIGNER-STUDY · 2026-07-17
Inside Jon Ingold's Philosophy — Keeping the Story Alive by Letting You Keep Failing
inkle's 80 Days, Heaven's Vault, Overboard! and the design of deduction, choice and failure
Introduction
Jon Ingold is the co-founder and narrative director of the British indie studio inkle. Together with Joseph Humfrey he has built games centred on story and choice: 80 Days, which threads a round-the-world journey through choices; Heaven's Vault, in which you decipher a lost ancient language; and Overboard!, a one-act in which you try to get away with murdering your husband aboard a liner. In Japan 80 Days is warmly discussed by a few as 'a modern reinvention of the gamebook', but the designer's own thinking is rarely examined.
What draws me (Kizuki) to him is that he refuses to set 'puzzle' against 'story'. The language-decipherment in Heaven's Vault is a genuine puzzle, yet it plays out inside characters' conversations rather than on a UI. And he says, again and again, that keeping the player failing is exactly what keeps the story alive. This piece is not a walkthrough; it is a study of the designer himself — philosophy, obsessions, failures, dilemmas, influences — quoting his own words. I have reheated the barley tea. Let us begin.
Background — From interactive fiction, and away from AAA
Ingold holds a degree in mathematics (his own words). His father is the anthropologist Tim Ingold, known for work on perception, technology and language, and Jon has said interactive fiction 'fascinated me from quite an early age' (Narrative News, 2026). He also recalls his time in the AAA world: 'When I was in the AAA industry about 20 years ago, everyone was like, you need to watch these seven movies and then you can be a narrative designer' (ibid.).
He then founded inkle with Joseph Humfrey and built a run of text-led games. The Sorcery! series, 80 Days, and then Heaven's Vault, which stepped out into a 3D world to explore — when it was announced, Game Developer introduced Ingold as 'studio co-founder' and quoted him asking, after the success of 80 Days, 'Where do we want to go next?' (Game Developer, 2017). More recently his range has widened to the one-act time-loop Overboard!, the escape drama Expelled!, and the mountain survival game A Highland Song.
Philosophy — Replicate the feeling, not the experience
Read across his interviews and one line runs through them. Of Heaven's Vault's invented language he said, 'The job of a game designer is not to replicate the experience, but to replicate the feeling of the experience' (Game Developer, 2017). So rather than build a whole Tolkienesque tongue, he gives the language just enough flavour for the player to suspend disbelief.
At the core of that 'feeling' is the tension of choice. He likens the pleasure of story to blackjack: 'I want this card to be an ace. I hope it's an ace. I think it's going to be an ace' — that moment before the card turns over is the delight, and 'narrative is just an extension of the same thing' (Narrative News, 2026). And he dislikes letting the player simply be 'you': 'the point of stepping into another person's shoes is to see what they see and to feel what they feel', whereas being yourself is 'boring as hell because we're stuck being ourselves all the time anyway' (Narrative News, 2026).
At the same time he is forever counting the balance of trust between writer and reader. He obsesses over line counts and whether text fits on a single screen, and says he keeps asking, 'have I earned the player's attention in this paragraph?' (ibid.). Not overestimating the reader's attention span is the ground his design stands on.
Obsessions — Deduction in dialogue, and no quests
Certain gestures recur across his games. One is refusing to abstract the puzzle. Of Heaven's Vault's reasoning he says, 'there's no UI of puzzle pieces or bits of string — and every deduction in the game is done in real dialogue between thinking, feeling characters (who don't ever repeat themselves)' (Archaeogaming, 2019). The puzzle is not decoration nor a separate screen; it is solved inside the story.
Another is placing no quests. 'We don't have quests. I feel very strongly about this. … Experience Points are Not A Thing,' he insists; things are tied together with coherent narrative 'or they don't happen' (ibid.). He also calls archaeology 'detective work … but without the grimness of murder and violence; and without the pressure of "failing to solve" a mystery' (Archaeogaming, 2019), and describes searching, with Humfrey, for 'puzzles that weren't abstract, but were narratively meaningful' (Game Developer, 2017).
His favourite metaphor for conversation design is improv: 'You have to offer just enough to throw the Frisbee back to them … the fact that the Frisbee keeps going is the thing that makes improv interesting' — not the virtuoso performance, but the back-and-forth (Narrative News, 2026).
Failures and how he got past them — Build it up, then strip it back
Most of the 'things that didn't work' he talks about publicly lie in the process of building too much and then cutting. He says the most time on Heaven's Vault went into the question 'why is archaeology exciting?', and that they 'would build more dramatic — and generally destructive — sequences, and then strip them out again, as we found we could make other, quieter elements shine if we only gave them room to breathe' (Archaeogaming, 2019).
The protagonist's design went the same way. Early sketches had steampunk goggles and a magic-wand-style widget, but 'with every pass these fell away again'. The small brush he long wanted her to carry 'never worked from a gameplay point of view', he admits frankly (Archaeogaming, 2019). He writes that he could not be happier with what was left after the cutting.
On Overboard! he is candid about the compromise of speed: 'that's really the only game we've made with a bona-fide optimise-this timeloop, and that's mostly because we made it so quickly we let ourselves off' (inkle / Jon Ingold, 2022). Not hiding the misses, and speaking of the cut marks as part of the design — that reads as how he gets past them.
Design dilemmas — Keeping tension without letting you die
The tension he has wrestled with longest is how to handle failure. In his essay The Problem of Failure (the rough text of a GDC talk planned for, but never given in, 2019) he takes Die Hard's McClane: we grip our seats fearing he'll be shot, while knowing for a fact he won't be killed. Tension and safety coexist — so how, he asks, do we replicate that in a game where failure equals death? (inkle / Jon Ingold, 2022).
His answer is 'fail forward'. 'Can we make games that are designed around squeaking through?' he writes; 'one of our goals as designers is to sustain the experience of failing for as long as possible.' 'If they win, or lose, the matter is finished; but if there's still life there is still hope' — which is why Heaven's Vault has no fail states, even though your translations are always at least a bit wrong (inkle / Jon Ingold, 2022).
A second tension is whether to start a story from a 'meaning'. Asked about the feminist themes of Overboard! and Expelled!, he says, 'I've never been able to start with a meaning and generate a story from it. It ends up too constructed, too forced', and the thrill is lost. His construction is the same: 'I organically explore the space of the conversation, but not organically explore the shape of the plot' — a rope stretched between improvisation and design (Narrative News, 2026).
Influences — His father's anthropology, old science books, and detective fiction
The influences Ingold names himself lean surprisingly on books and people. First, his father Tim Ingold. He says his father's 'emphasis on gesture and process, over input and output, really significant' for working in a digital medium, and that a sense of cultural relativity came 'for free' (Archaeogaming, 2019). He recalls his father's friend, before a museum feather cloak, saying 'so, it's a coat', and concludes: 'The plaque was quite accurate. We just don't understand coats.'
What settled archaeology as his subject was an article about the Egyptian archaeologist Dr Monica Hanna, from which he took the idea that 'archaeology is modern myth-making' (Archaeogaming, 2019). He also cites a 1960s book, Citadels of Mystery, found on a second-hand rack in Cambridge: 'Old science books are an amazing source of inspiration: they are robustly thought through, but sometimes schlocky; they're willing to credit theories that seem wild now' (ibid.).
For story shapes, he puts Overboard!'s starting point in an Agatha Christie-like locked room, and invokes Raymond Chandler's Philip Marlowe: solving the case 'is never the point of a Raymond Chandler book' (Narrative News, 2026; Archaeogaming, 2019). For puzzle design he describes mixing 'the logical progression of games like The Witness' with non-contrived, narratively grounded puzzles (Archaeogaming, 2019). And he prizes books over film: 'if we just start from games, we build space invaders and forget to put a new story in it' (Narrative News, 2026).
Kizuki's reading
From here on is my (Kizuki's) interpretation, a step beyond his own words. I read Jon Ingold as a designer who reframed failure as fuel for story. For many makers failure is a wall that stops the player; for him it is the engine that moves the story forward. Keep the tension without the death, let the player squeak through — this 'fail forward' is cut from the same cloth as embedding deduction in dialogue rather than outsourcing it to a UI, and as the abstinence of placing no experience points. It is all design in service of 'earning' that finite resource, the player's attention.
And this stance, I think, quietly rhymes with his father's anthropology: a gaze on gesture and process rather than input and output; the cultural relativity that can see a feather cloak as 'just a coat'. Ingold's games let you play the process itself — the doubt, the mistranslation, the bluffing — rather than the output of win or lose. That he 'can't start from a meaning' is, I suspect, the same reason: meaning is a result, not a starting point. Someone who trusts the process does not fear failure.
Closing
Where should you meet Jon Ingold? For the quickest taste of his rhythm of story and choice, 80 Days; for the moment where puzzle and story become one, Heaven's Vault; and the design idea of 'keeping you failing' is most concentrated in the one-act Overboard!. In each you can feel his 'Frisbee going back and forth' with your own hands.
As a path onward, our Kizuki pieces sit him beside Sam Barlow (deduction and story), Fumito Ueda (negative space and subtraction), and Bennett Foddy (designing failure itself). Read as a lineage that tries to pull puzzles back from 'work on a separate screen' to 'thinking inside the story', Ingold's place comes into sharper focus.
Sources
Primary sources referenced in this article:
・Archaeogaming, 'Archaeology and Heaven's Vault: An Interview with inkle's Jon Ingold' (2019-01-25)
・Narrative News, 'Inkle's Jon Ingold on the Craft of Narrative Design' (Rose Behar, 2026-04-01)
・inkle / Jon Ingold, 'The Problem of Failure' (2022-03-18; rough text of an unheld 2019 GDC talk)
Reactions (no login)
Anonymous • one of each per visitor per day
Part of these series
The Designer's PhilosophyEpisode 25 of 25