I have never once decided what my app costs in Vietnam.
I picked a US price — $5.99, about ninety seconds of thought — ticked the box that fills in every other country, and shipped. Apple did the rest. It does this for roughly 175 storefronts, quietly, for almost every indie developer alive, and almost none of us ever look at what came out the other end.
I run a site about the Big Mac index, so I eventually looked using the only measuring stick I own.
Measured in Big Macs, the same app costs 0.40 in Turkey and 1.70 in Vietnam. Same file. Same download. A Vietnamese customer pays four and a quarter times what a Turkish one does, in the one unit that is physically the same object in both countries.
Then it got worse, in the useful way. The market that outsold every English-speaking country I sell in — at roughly half the American price — was China. The market where I charge more than America, and where nobody asked for a refund, was Saudi Arabia. And the region I’d spent two years believing was full of the world’s best customers came in at a combined index of three, against America’s hundred.
It’s a drawing toy — you draw something, it wiggles, you share it. Not a serious piece of software, and not a serious business. That turns out to be an advantage for this particular exercise: nobody buys it because they need it, so what’s left in the numbers is closer to pure willingness to pay. The next version, JigglyPaint, ships on 31 August 2026.
Here’s the whole table, what actually sold, and the three things I believed that the data didn’t.
A note on the numbers before we start. Sales figures are indexed to the United States at 100 — the absolute unit counts aren’t mine alone to publish. Prices are the public App Store prices, which anyone can check. Big Mac prices are the July 2026 edition. The sales month is May 2026, iOS only. One month, one app. Treat everything below as a case, not a study.
Apple’s default price tiers
Apple gives you a price tier table. You pick a US price, and it fills in every other country by converting at something close to the market exchange rate, then rounding to a locally plausible number.
Almost everyone ships that default. I did too, for the first year.
The logic seems fine until you say it out loud: the right price in Jakarta is the New York price, converted. Nobody would defend that if you phrased it as a claim about Indonesian buying power. It survives because it’s the checkbox that’s already ticked.
There’s an old phrase for this — 削足适履, cutting your foot to fit the shoe. The shoe here is a US price list, and every other country gets trimmed to it.
The Big Mac index exists because economists needed a cheap way to ask whether exchange rates reflect what money actually buys. That is the same question I’m asking about a $5.99 app. So I started pricing the app in Big Macs — and eventually built an App Store pricing calculator so I’d stop doing it by hand in a spreadsheet.
Price localization, measured
Here’s the whole thing. Price is what a customer pays; the last column divides it by that country’s Big Mac in the July 2026 edition.
| Country | Local price | ≈ USD | Big Mac | App in Big Macs |
|---|---|---|---|---|
| Turkey | ₺129 | $2.74 | $6.91 | 0.40 |
| Mexico | MX$79 | $4.54 | $6.27 | 0.72 |
| Canada | C$5.99 | $4.26 | $5.81 | 0.73 |
| Australia | A$6.49 | $4.54 | $5.95 | 0.76 |
| China | ¥21.80 | $3.22 | $3.91 | 0.82 |
| Singapore | S$6.98 | $5.41 | $5.77 | 0.94 |
| Thailand | ฿129 | $3.84 | $4.02 | 0.96 |
| United States | $5.99 | $5.99 | $6.22 | 0.96 |
| Germany | €5.99 | $6.85 | $7.08 | 0.97 |
| Britain | £5.99 | $8.07 | $7.40 | 1.09 |
| Korea | ₩6,400 | $4.31 | $3.84 | 1.12 |
| Indonesia | Rp49,000 | $2.71 | $2.38 | 1.14 |
| India | ₹299 | $3.11 | $2.45 | 1.27 |
| Saudi Arabia | SAR 24.99 | $6.66 | $5.06 | 1.32 |
| Hong Kong | HK$35 | $4.46 | $3.25 | 1.37 |
| Philippines | ₱249 | $4.04 | $2.74 | 1.47 |
| Taiwan | NT$115 | $3.57 | $2.42 | 1.47 |
| Malaysia | RM21.99 | $5.39 | $3.57 | 1.51 |
| Japan | ¥790 | $4.87 | $3.08 | 1.58 |
| Vietnam | ₫129,000 | $4.91 | $2.89 | 1.70 |
Look at the middle of that table. The US, Thailand, Germany and Singapore all land within three points of each other — 0.94 to 0.97. Four countries with nothing in common, sitting on the same number.
That’s not because I was careful. It’s because those four are places where the market exchange rate happens to sit close to purchasing power, so Apple’s conversion and the Big Mac agree. Where they disagree, my price went wherever the exchange rate took it.
Turkey is the extreme. A Turkish Big Mac costs $6.91 — more than an American one — while my app there converts to $2.74. I never decided that a Turkish customer should pay a third of what a Vietnamese customer pays in burger terms. The lira decided.
China outsold the US
Indexed to the United States at 100, China came in at 159.
More than half again as many sales as my largest English-speaking market, at a price that’s 54% of the US price. Taiwan was 19, Japan 14, Mexico 9, Saudi Arabia 8, Britain 8, Australia 6.
I want to be careful about what this does and doesn’t show. It’s one app in one month, and a drawing toy travels differently than a productivity tool. It does not prove China is anyone’s biggest market.
But it does something else, and this is the part that stuck with me. For years I sorted the world into developed and developing, first world and third — and priced accordingly, low tier for the second group, without ever checking. That framing is wearing a monocle. It bundles income, willingness to pay, payment infrastructure, and app-store maturity into one word, and then the word does the thinking for you.
China’s Big Mac costs $3.91 against America’s $6.22. By raw price, cheap. Adjusted for income, the Chinese Big Mac is roughly at fair value — you can run it yourself in the purchasing power calculator, which is exactly the correction the adjusted index is built to make. The naive framing and the corrected one disagree, and my sales followed the corrected one.
Saudi pays more than America
Saudi Arabia ties Britain for sixth place by units, and I charge more there than in America: SAR 24.99, about $6.66 against $5.99. Britain and Germany are higher still in dollar terms — but those two are expensive by accident, because the pound and the euro converted that way. Saudi is the only one where the tier I picked lands above the US and the market didn’t flinch.
In Big Macs that’s 1.32 — meaningfully more than the 0.96 an American pays. I didn’t set that deliberately either. It fell out of Apple’s tier rounding.
And it sold anyway — level with Britain, at 1.32 Big Macs against Britain’s 1.09.
Did I plan this? No. I found it in a spreadsheet three months after the fact.
What I take from it is narrow: the Gulf storefronts are not a low-tier market, and treating them as “emerging” — pricing them down with Egypt and Pakistan — would have left money on the table. Sensor Tower’s regional numbers point the same direction, with Gulf in-app revenue up 41% over two years while download growth stayed in single digits. An index score of 8 doesn’t confirm that. It just stopped me from arguing with it.
Britain sent a fifth back
Refunds, as a share of gross sales in the same month:
| Country | Real price | Refund rate |
|---|---|---|
| Britain | $8.07 | 20.0% |
| Germany | $6.85 | 16.7% |
| Thailand | $3.84 | 14.3% |
| China | $3.22 | 6.9% |
| Japan | $4.87 | 6.2% |
| Taiwan | $3.57 | 4.5% |
| United States | $5.99 | 0.0% |
| Saudi Arabia | $6.66 | 0.0% |
Britain and Germany are my two most expensive storefronts in real terms — £5.99 converts to about $8.07, €5.99 to about $6.85 — and they’re the two that sent product back. That’s a tidy story, and I don’t fully trust it.
Britain’s 20% is two refunds out of ten. Germany’s is one out of six. You can’t do anything with numbers that small except notice them.
The one figure with weight behind it runs the other way: zero refunds out of 118 in the United States, at $5.99. That’s a real sample, and it says nothing bad about the American price.
Saudi Arabia breaks the pattern too — $6.66, the highest absolute price on my list, and nobody asked for their money back.
There is published work pointing the same direction as Britain. Luca and Reshef, in Management Science, tracked daily menu prices against online ratings at a food-delivery platform and found that a 1% price increase produced a 3–5% drop in average rating — a clean within-restaurant design, not a comparison across restaurants. Their explanation isn’t that customers retaliate. It’s that price raises what you expect, and the product then has to clear a higher bar.
If that carries over to software, my British price is buying me a harder audience. Maybe. It’s a restaurant study, my sample is ten units, and I’d be embarrassed to present that as a finding. I’m presenting it as a thing I noticed and then went looking for company on.
The Nordics never showed up
Now the part where I was wrong.
I’ve believed for a while that Nordic users are the best customers in the world. High incomes, good payment habits, comfortable with paying for quality — the IKEA theory, roughly. People used to a high cost of living and high standards will pay a premium if the thing is good.
My May numbers: Norway 1 unit. Finland 2. Sweden zero. Denmark zero.
The first thing to say is that this is a sample of three, and three is not a number you get to reason from. If I’d pulled a different month it might read differently.
The second thing is worse for me. I went looking for evidence and found a study measuring how harshly users in different countries rate the same apps, controlling for the app itself. Nordic countries came out at roughly −0.05 stars against the global average — Sweden −0.018, Norway −0.019, which is indistinguishable from zero. The genuinely harsh storefronts were Saudi Arabia at −0.25, Egypt at −0.20, Kuwait at −0.18.
So the half of my theory that said “Nordics demand more” has no support. The other half might survive: Denmark and Norway do rank fifth and sixth worldwide in App Store spending per capita, which is real. But per-capita spending in a country with that income level is close to a tautology. It doesn’t tell you anyone is willing to pay a premium. It tells you they can.
And here’s the part I can’t resolve. Our own Big Mac data says the Nordic premium is real after you adjust for income — Switzerland +48.5%, Norway +33.6%, Sweden +26.6%, Denmark +19.7% above fair value. That’s not “they’re rich, so things cost more.” The richness is already divided out. Something is still there. I just no longer think “they’re pickier” is what it is.
I’d rather leave that open than patch it. Judge it yourself — the numbers are all above.
What I’d do differently
Four things, and only the first one is confident.
Reprice, never discount. Regional pricing is a long-term anchor: a Vietnamese customer sees ₫129,000 and that’s simply the price. A global $9.99 marked down to ₫129,000 is a sale, and a sale creates a reference point. The best-evidenced finding I ran into on any of this is that discount-acquired users leave much worse ratings — the Groupon study on Yelp found a 0.46-star gap — and that it isn’t because those people are meaner. They rated other businesses more moderately than average. Price doesn’t change how harsh someone is. It changes who walks in.
Stop reading the absolute complaint count. I used to think mid-priced tiers attracted the fussiest users. Middle tiers also have the most users, and I was counting complaints, not rates. Anyone doing this should divide by active users first and see whether the pattern survives. Mine may not.
Don’t treat $0.99 as almost-free. Free is its own category, not a very low price. The moment you charge anything, you inherit the full expectations of a paying customer.
Check the table in Big Macs before shipping a price change. Not because the Big Mac index is precise — it’s a burger, it carries local rent and wages and beef tariffs that have nothing to do with software. But it’s the only widely published number that’s the same object everywhere, and it catches the 0.40-versus-1.70 problem in about ten seconds. I built an App Store pricing calculator that does this arithmetic for every storefront, and the methodology page explains where the underlying prices come from and how far to trust each one.
None of this required a theory of world markets. It required noticing that I’d been quietly using an exchange rate as a theory of world markets, which is the kind of thing you can do for two years without ever saying it out loud.
I still don’t know why Sweden bought nothing.