Team Cherry never published a budget for Hollow Knight, so anyone quoting an exact dollar figure is guessing. Here is what is actually on the record: the 2014 Kickstarter campaign raised A$57,138 from 2,158 backers, the core team was two people ( Ari Gibson and William Pellen ) plus a small handful of contributors, development ran from a 2013 game jam to the February 24, 2017 PC release, and personal savings covered the gap once the Kickstarter money ran out. The game has since sold more than 15 million copies off a $15 launch price.
The rest of this post separates what Team Cherry disclosed from what has to be estimated, runs the revenue math the way you would run it for your own game, and labels every estimate as an estimate. It is a spoke of our famous indie game development budgets hub, so if you want the same treatment for other titles, start there.
The honest answer: what is actually on the record
Four facts are disclosed and checkable. Everything past them is inference.
- The Kickstarter. It ran from November 19 to December 19, 2014, with a goal of A$35,000. It closed at A$57,138 from 2,158 backers, roughly US$43,000 at that year’s exchange rate. That is the only hard funding number Team Cherry has ever put on the table.
- The team. Team Cherry is Ari Gibson and William Pellen. Depending on how you count, the credited team was about five people: Gibson and Pellen, programmer David Kazi, composer Christopher Larkin, and Matthew Griffin handling marketing and business. So “a team of two to three” is fair for the core and understates the full credits a little. Larkin and Griffin read more as contractors than staff.
- The timeline. The game started as a 2013 game jam concept, hit a beta state around September 2015, and shipped on PC on February 24, 2017 ( dates per Wikipedia’s Hollow Knight page ). Call it about two and a half years from the Kickstarter to launch, closer to three and a half from the original jam.
- The gap funding. The Kickstarter money did not cover a multi-year build for even a tiny team. In interviews around the 15-million milestone, the founders described keeping costs low with leftover lunches from a neighbouring office, and one of them recalled his dad occasionally popping by to hand him $20. Personal savings did the rest. The amount was never disclosed.
That last point is the load-bearing one. A$57,138 is what Kickstarter paid out. It is not the budget. The budget was that number plus an unknown pile of the founders’ own money and, mostly, their unpaid time.
What Hollow Knight probably cost to make
The data says: A$57,138 in crowdfunding, plus undisclosed savings, plus two to three people for roughly three years. That is the whole factual base. Any single dollar figure past it is a model, not a fact.
You will see write-ups floating a multi-million-dollar “budget” for Hollow Knight. Treat those with suspicion. A two-to-three-person team eating leftover sandwiches did not spend millions in cash. What those numbers are really trying to price is the value of the team’s time, and that is a different question with a much softer answer.
There are two honest ways to think about cost, and they give very different figures.
Cash out of pocket. Kickstarter money plus personal savings. Genuinely unknown, but bounded by common sense: tens of thousands of dollars in crowdfunding, and personal savings that a two-person team can plausibly burn over three years lands you somewhere in the low-to-mid five figures, maybe into six if they were paying Kazi and Larkin properly. Cash cost of Hollow Knight was small. That part is not really in dispute.
True economic cost. If you paid this same team market indie salaries instead of letting them work for savings and sandwiches, what would three years cost? At roughly US$50,000 to US$80,000 per person per year, for two to three full-time-equivalent people over about three years, you land around US$300,000 to US$700,000. That is my estimate, not a Team Cherry number, and it is an order-of-magnitude figure. Do not quote it as fact.
Here is the same split in a table, so the line between disclosed and estimated stays visible:
| Item | Figure | Basis |
|---|---|---|
| Kickstarter raised | A$57,138 (~US$43K) | Disclosed (Kickstarter, 2014) |
| Personal savings added | Undisclosed | Stated by founders, amount never given |
| Core team | 2 (Gibson, Pellen) | Disclosed |
| Full credited team | ~5, incl. Kazi, Larkin, Griffin | Game credits |
| Development span | ~2.5 yrs Kickstarter to launch | Disclosed |
| True labor cost | ~US$300K-700K | Estimate (2-3 FTE x ~3 yrs at indie salary) |
The reason this matters for your own game: the cash budget of a Hollow Knight is not the interesting number. The interesting number is the opportunity cost of the founders’ time, and that cost is only bearable if your burn rate is near zero. Our indie game development costs breakdown walks through where that money actually goes for teams that are paying themselves.
The revenue side: 15 million copies at $15
Now the fun half. Team Cherry has confirmed the sales figure directly, so for once we do not have to estimate units.
Normally, to size an unknown game’s sales, you would count its Steam reviews and multiply by a sales-per-review factor. That is the Boxleiter method, and the current review to sales multiplier sits around 20 to 60x depending on genre and era. For Hollow Knight you can skip all of that, because the studio handed us the number: more than 15 million copies across PC, Switch, PlayStation, and Xbox, as of an August 2025 Bloomberg interview.
The only real unknown left is the average price people actually paid. The list price was $15, but Hollow Knight has spent eight years on deep discounts, often 50% off or lower during Steam sales ( you can see the pattern on its SteamDB price history ). Nobody pays list across 15 million copies over eight years. So the honest move is a range, not a point.
Read that chart as an estimate with the label on it. At a plausible average realized price of $7 to $11, lifetime consumer spend on Hollow Knight lands somewhere around US$105M to US$165M. If every copy had sold at full $15 the ceiling would be $225M, but that number never happened and never could, so I left it off the chart.
That is gross consumer spend, not what Team Cherry banked. Platforms take their cut first: Steam’s revenue share is 30% up to $10M gross, 25% above that, 20% above $50M, and console storefronts sit near 30%. After platform cuts, the studio’s take is lower, call it very roughly US$85M to US$120M across the game’s life, still an estimate. To model the same math against your own price and unit count, the revenue calculator applies the platform tiers and refund rates in one pass.
Set the two halves side by side and the case study writes itself. A cash budget in the five figures, a true labor cost estimated in the low hundreds of thousands, against nine figures of revenue. That is a return most studios will never see, which is exactly why the “how” is worth being careful about.
The long tail is the real story
Here is the number that reframes everything. Of those 15 million copies, only 2.8 million had sold by February 2019, the point at which Silksong was announced. The other roughly 12 million came after ( split confirmed in the same August 2025 interview ).
Hollow Knight sold about 500,000 copies in its first nine months, crossed 1.2 million by mid-2018 after the Switch port, and reached 2.8 million by early 2019. Then it kept selling for six more years, mostly on word of mouth, Switch attach, and the slow drumbeat of Silksong anticipation. The launch window was a small fraction of the total.
That is the single most transferable lesson in this whole post, and it is not a Team Cherry secret. Most indie revenue arrives after launch week, not during it. A game that reviews well and keeps getting recommended compounds for years. Our how much does a Steam game make data shows the same shape across the catalog: the launch spike is real, but the tail is where the money lives for the games that earn one.
The 15-million figure is a lifetime, cross-platform number confirmed by the studio. The 2.8M-then-12M split is also from the studio. The dollar revenue attached to it is my estimate, because average price is not public.
What Silksong changes
Silksong shifts the answer to “how much did Team Cherry make,” because a studio’s take is its whole catalog, not one game.
Hollow Knight: Silksong launched on September 4, 2025 across PC, Switch, PlayStation, and Xbox, priced at $19.99. Co-director Ari Gibson framed the low price the same way he framed Hollow Knight’s: get as many people playing as possible. The launch was enormous by indie standards.
According to the Alinea Analytics newsletter by Rhys Elliott, Silksong passed 5 million players in three days, with over 3 million Steam copies and about $50 million in Steam revenue in that window alone. Later Alinea reporting put it near 5.5 million copies and close to US$100 million in revenue by late 2025. Confidence label: those are third-party analyst estimates from sales-tracking models, not audited Team Cherry figures, and they exclude whatever Microsoft paid for the simultaneous Game Pass launch. Treat them as well-sourced estimates, not accounts.
What is not an estimate: Silksong was again made by a tiny team. The same Alinea coverage describes it as the work of four people and a handful of contractors over roughly seven years. The pattern held. Two similar games, both built by almost nobody, both selling in the millions.
The factors that made this viable
Case studies are only useful if you name the mechanics, so here they are, separated into things the team controlled and things they did not.
Things they controlled:
- Near-zero burn rate. No office rent worth the name, no salaries at market rate, savings and sandwiches instead of a runway. This is what let a three-year build survive on five figures of cash. It is also the single hardest thing to copy if you have a mortgage.
- Scope discipline against the money. They built to what the Kickstarter and their savings could sustain, then expanded through free content updates ( Hidden Dreams, The Grimm Troupe, Godmaster ) after the game was already earning. Updates, not a bigger launch budget, drove the tail.
- A price set for reach, not margin. $15 at launch, $20 for the sequel, both chosen to maximise players over per-unit revenue. On 15 million units, reach won that argument by a wide margin. Our Steam pricing strategy guide covers when this trade actually pays off and when it just leaves money on the table.
Things they did not control:
- The quality spike and the word of mouth it earned. You cannot budget your way to “this is one of the best metroidvanias ever made.” Either the game lands like that or it does not, and most do not.
- The genre timing. Metroidvanias have long shelves and passionate communities. A game with the same discipline in a fad genre would not have a six-year tail.
What you can copy and what you cannot
Before any lesson, the caveat that every honest post-mortem needs: this is survivorship bias in its purest form. We are looking at Hollow Knight because it worked. Thousands of teams ran a small Kickstarter, kept their burn low, and shipped a tight game, and most of them you have never heard of. Low burn and scope discipline did not cause 15 million sales. They made it possible for the game to survive long enough to find out whether it would sell that many, and then it happened to be good enough that it did.
So here is the split, stated plainly.
Replicable, and worth copying:
- Keep your burn low enough to survive a long build. The whole model depends on being able to work for years without a salary you cannot skip. If you can arrange that, do.
- Scope to your funding, then expand from revenue. Ship what your money can build, add content once the game is earning. Free updates fed Hollow Knight’s tail for years.
- Plan for the tail, not the spike. Most of your revenue, if you get any, arrives after launch week. Budget your marketing and your patience accordingly. Before you commit to a multi-year timeline, our how long it takes to make an indie game data is a sobering read.
Not replicable, and dishonest to promise:
- The word of mouth. I don’t know if you can replicate this, and neither does anyone selling you a course that says they can. The quality that earns organic recommendation is the whole game, and it is not a tactic.
A decision rule to close on. Copy this model if you can genuinely get your monthly burn near zero, you are building in a genre with a long shelf, and you are prepared for most of the payoff to arrive years after launch. Do not copy it if you need income inside 12 months, you are counting on launch week to pay the bills, or your plan assumes your game will be the one in a thousand that goes viral. The parts of Hollow Knight you can control are the boring ones ( low burn, tight scope, patience ). The exciting part was never in the budget.
Want to pressure-test your own numbers before you commit? Run your store page through the free Steam Page Analyzer to see how your conversion elements stack up, then model your break-even against a realistic price and unit count in the revenue calculator. And read the famous indie budgets hub for the same math on the other games people keep pointing to.