On September 30, 2026, Mario Kart Tour stops working. Nintendo announced the shutdown on July 8, stopped selling Rubies, and switched off Gold Pass renewals. There is no offline version and no refunds. Seven years, 290.7 million downloads, $365.6 million in player spending, and then nothing.
Two years earlier, Animal Crossing: Pocket Camp died the same way. Nintendo replaced it with a $10 one-time purchase called Pocket Camp Complete.
Ten dollars, once. That is where Nintendo started on mobile in 2016, and after nine years, nine games and $2.27 billion in player spending, that is roughly where it ended up.
This is not a story about bad games. Nintendo’s mobile titles were built by the same people who make the best console games on earth, and several of them are good. It is a story about a company that got 816 million installs and had almost no way to earn from the people holding them.
The company that did not want to be here
Nintendo resisted smartphones for years. Satoru Iwata told investors the plan was to release something on smart devices that could communicate the value of Nintendo’s entertainment offerings, which is a careful way of saying mobile was a marketing channel. The DeNA partnership in 2015 made it real. Miitomo came first in 2016 and made $3.7 million across 20.8 million downloads before shutting down.
Then came Mario.
Super Mario Run launched December 15, 2016. Free to download, three levels playable, $9.99 to unlock the rest. It hit 25 million downloads in four days and roughly $21 million in App Store spending. Nintendo’s stock fell 11% in the same week, because the reviews were about the price. In Japan the App Store rating sat at 2.5 stars.
Sensor Tower’s read at the time was that about 8% of people who tried it bought it. By the January earnings call, with 78 million downloads on the board, Nintendo put the paid conversion at just over 5% and said it had wanted double digits.
The downloads never stopped. The revenue did. Super Mario Run now sits at 384.4 million downloads and $96.6 million in lifetime player spending. That is 25 cents per download, across the most recognisable character in games.
Why ten dollars once was the right answer to the wrong question
The $9.99 was not a mistake made in ignorance. It was the output of a coherent strategy, and the strategy is the part worth understanding.
Nintendo had two things it was protecting. The first was the console business. A free-to-play Mario that people could sink forty hours into would compete directly with a $60 Mario on a machine Nintendo also sells. Charging once, capping the experience, and keeping the whole thing a taster made mobile safe for the hardware line. The second was the brand’s position with parents. A Nintendo spokesman told Reuters the fixed price was chosen partly to reassure parents that there would be no further charges. In 2016, with loot box scrutiny building, that was not a small consideration.
Put those together and you get an accounting where downloads are the product. If mobile exists to widen the funnel into Switch, then 384 million people touching Mario on a phone is a win, and 25 cents per download is a marketing cost, not a revenue failure. Shuntaro Furukawa was still framing it that way years later, telling investors mobile’s job was to broaden the fan base for Nintendo IP and increase points of contact with customers.
The problem with that accounting is that it never has to justify itself. No line item anywhere in Nintendo’s reporting shows what the unmonetized 95% was worth. The cost of the decision does not appear as a cost. It appears as an absence.
Then Fire Emblem proved it
Six weeks after the Super Mario Run earnings call, Nintendo shipped Fire Emblem Heroes with a gacha economy. Players spend orbs to summon random characters, orbs are earned slowly or bought, and the summon rate does the rest. Deconstructor of Fun pulled the loop apart back in 2017 and the design analysis has been settled for years. What matters here is what it did to the shape of Nintendo’s business.
Heroes made $265.1 million in its first 365 days, from a fraction of Super Mario Run’s audience. Its first 30 days alone came to $51.4 million against Fire Emblem Shadows’ sub-$200,000 eight years later. Lifetime, Heroes stands at $1.193 billion on 19 million downloads. That is $62.81 per install.
Two games, released ten weeks apart, from the same publisher. One earned 25 cents a download. The other earned $62.81. The gap is not audience size or brand strength or game quality. It is the money model, and nothing else.
Nintendo saw the result and half-committed. Animal Crossing: Pocket Camp got a subscription. Mario Kart Tour launched as a gacha in 2019, ran that way for three years, then dropped the mechanic in 2022 after player backlash. Dragalia Lost was built gacha-first with Cygames and made $167.1 million on 4.3 million downloads, a strong per-install number that was never going to scale on a new IP with no console business behind it.
None of it moved the concentration. Heroes has carried more than half of Nintendo’s total mobile revenue for most of a decade, and it still does.
The years the one line item shrank
Gacha revenue decays. Everyone in mobile knows this. What Nintendo’s numbers show is how quietly it happens.
Heroes fell 12% in year two to $233.7 million, then 34% in year three to $155.1 million. Year four ticked up marginally to $155.4 million, helped by the FEH Pass subscription and lockdown-era spending. After that it went down every single year. Year nine, ending February 2026, brought in $47.9 million. That is 18% of what the game made in its first twelve months.
Look at the annual steps between year four and year nine and the average works out at roughly 21% a year. Spread across months, that is a decline of about 2% a month, every month, for five years. No single month looks like a problem. No alert fires at 2%.
Meanwhile the rest of the portfolio was doing worse. Dr. Mario World managed $13.8 million across its whole life. Mario Kart Tour, with 290 million downloads, never cracked $1.30 per install.
And Nintendo’s revenue base was geographically thin. Japan supplies about 8% of Nintendo’s mobile installs and 46% of its mobile money. For Heroes specifically, half of lifetime spending came from Japan alone. One game, one country, one mechanic.
Closing time
The wind-down was slow enough that no single announcement felt like a decision.
Dr. Mario World closed in November 2021. Dragalia Lost, Nintendo’s only original mobile IP, ended service in November 2022. Mario Kart Tour stopped receiving new courses and characters at the end of 2023 and sat in maintenance mode for two and a half years. Pocket Camp’s online service ended in November 2024 and came back as the $10 offline purchase.
Then Fire Emblem Shadows arrived in September 2025 with no announcement, no gacha, and a battle pass. It has made $516,000 in its first four months. Appfigures measured its first ten days at $0.21 per download against Heroes’ $2.31 per download over the same window in 2017.
Five weeks from now, Mario Kart Tour goes off. Every driver, kart and glider that 290 million people collected stops existing. Unspent Rubies are worthless, and Nintendo has said there is no refund programme and no offline client.
Where it stands now
Nintendo’s nine mobile games have generated 816 million downloads and $2.27 billion in player spending since September 2016. Average revenue per download across the whole portfolio: $2.79. Strip out Fire Emblem Heroes and the remaining 797 million downloads produced about $1.35 each.
For scale, Furukawa took over in 2018 telling Nikkei he wanted to build mobile into a ¥100 billion business, then roughly $910 million a year. Nine years of gross player spending across every title Nintendo has ever shipped on phones comes to about two and a half times that, in total, before Apple and Google take their cut and before DeNA takes theirs.
Nintendo is fine. Switch 2 sold 19.86 million units in the fiscal year ended March 2026, the movie business works, the parks work. The mobile chapter closed because it never mattered enough to fix. That is the honest version, and it is also what makes it useful to look at.
The uncomfortable version for your studio
Here is the portfolio in one sentence. Nintendo’s biggest install driver, Super Mario Run, accounts for 47% of every download the company has ever had on mobile and 4.3% of the money. Its smallest install driver in the top five, Fire Emblem Heroes, accounts for 2.3% of downloads and 52.6% of the money.
That inversion is not a Nintendo problem. It is the default shape of any business where installs are free and revenue depends on a single rail. Three to five percent of a free-to-play base ever makes a purchase. The remaining 95% either has a second rail attached to it or produces nothing. Nintendo chose to attach nothing, for reasons that were defensible in 2016, and the result is nine years of the largest free audience in mobile gaming generating $1.35 a head.
The second half is the part that should worry an ad ops team more.
Fire Emblem Heroes did not crash. It slid about 2% a month for five years while Nintendo kept shipping banners, kept running Choose Your Legends, and kept the game alive. Nobody at Nintendo failed to notice. There was simply never a month where the number was ugly enough to force a change. That is the same arithmetic as a top-earning placement whose eCPM drifts down 2% a month, or a demand partner whose fill quietly softens, or a Tier 1 geo that stops clearing your floor. Over twelve months, 2% a month is a 22% hole. Over five years it is 70%. And the monthly report never once says anything is wrong.
Concentration and slow decay are the same failure viewed from two angles. If one line item carries half your revenue, you inherit its decay curve whether you are watching it or not.
What to actually check this quarter
None of this requires buying anything.
- Compute revenue per install by title, split IAP and ads. Find the title where your install share and your revenue share disagree most. That gap is where a monetization decision is costing you and nobody has priced it.
- Measure what your non-payers earn. Take the share of DAU that has never made a purchase and calculate ARPDAU on that segment alone. If the number is zero or close to it, you have Nintendo’s problem, and rewarded formats are the cheapest place to start fixing it.
- Chart your single largest line item, monthly, twelve months back. Fit the trend and annualise it. Anything sliding more than 1.5% a month is a double-digit annual hole that no threshold alert will ever catch, because no individual month breaches it.
- Write down your concentration numbers. What percentage of ad revenue comes from your largest demand partner, your largest geo, and your single best placement? Nintendo’s number was 52% from one title and roughly half from one country. Know yours before someone asks.
- Time one real drop from last month. Find when performance actually fell, find when someone changed something, and multiply the hours between them by your hourly revenue in that geo. That is what reaction delay costs you, stated as a number instead of a feeling.
FAQ
How much has Nintendo actually made from mobile games?
Roughly $2.27 billion in gross player spending across nine titles and 816 million downloads since September 2016, per Sensor Tower. That figure is player spending before platform commission, so Nintendo’s net is materially lower, and it excludes Pokémon titles, which are published by The Pokémon Company rather than Nintendo directly.
Why did Super Mario Run underperform if it had so many downloads?
Because a one-time unlock earns nothing from anyone who does not buy it. Nintendo reported conversion at just over 5% when downloads hit 78 million and said it had hoped for double digits. Downloads kept climbing to 384.4 million while lifetime spending stalled near $96.6 million, which is 25 cents per download.
Is Nintendo finished with mobile?
Not formally. Fire Emblem Heroes, Super Mario Run, Pikmin Bloom and Pocket Camp Complete are still live, and Fire Emblem Shadows launched in September 2025. But Shadows made $516,000 in four months, Mario Kart Tour shuts down on September 30, 2026, and Nintendo has released two new mobile games in six years.
Was gacha the right call for Nintendo?
The revenue answer is unambiguous: Fire Emblem Heroes has earned more than every other Nintendo mobile game combined. The brand answer is contested, and the public data cannot settle it. Nintendo dropped Mario Kart Tour’s gacha in 2022 after player pushback, and it has never put the mechanic near Mario or Zelda. Whether that restraint protected the IP or just cost money is a judgement, not a measurement.
What happens to money players spent in Mario Kart Tour?
Nothing comes back. Nintendo has confirmed no refunds for unspent Rubies or past Gold Pass subscriptions and no offline version. The terms of service sold access rather than ownership, which is standard across live-service mobile and is now getting more attention than it used to.
What is Fire Emblem Heroes earning today?
$47.9 million in year nine, ending February 2026, per AppMagic. That is 18% of its first-year total of $265.1 million, and it is still comfortably Nintendo’s best-performing mobile game.
Find out what your top line item is actually doing
Nintendo’s mobile business was decided by a number nobody had to look at every month. Most ad stacks have one of those.
UndrAds runs a revenue leak audit on your existing Google Ad Manager setup: how often performance drops, how long each drop runs before anyone reacts, what the reaction delay costs per month, and where your revenue concentration actually sits. No SDK, no app changes, no migration required to find out. Get in touch with the UndrAds team to book it.
