China pays about half the American price for my app, and China outsold the United States by 1.59 to 1.
That’s the single strongest argument for charging less in poorer countries that I own, and I own it because it’s my own Apple report, not somebody’s case study. May 2026, one app, iOS only. Indexed to the United States at 100, China came in at 159 — at ¥21.80, roughly $3.22, against $5.99 in the States.
So that’s the answer, then — cut your prices in poor countries and go collect the money? No. The same report holds two numbers pointing the other way, and one of them is embarrassing.
Saudi Arabia pays about 37% more than America for the same app in real terms — SAR 24.99, roughly $6.66, which works out to 1.32 Big Macs against the 0.96 an American pays. It didn’t punish me for it. Refunds there were 0.0%, and it finished tied for sixth on the sales index. Meanwhile the four Nordic countries — the richest customers in the world, by the folk wisdom every indie developer repeats — came in at Norway 1, Finland 2, Sweden 0, Denmark 0 on that same index where America is 100.
One poor market bought more at a discount. One market with half America’s income per head bought fine at a premium. Four rich markets didn’t show up. If you’re holding the question should I charge less in developing countries, that’s three answers, and they don’t agree.
I think the question is built wrong. Not “is this country poor” — “what does this price mean in this market.” Those aren’t the same question, and only one of them has an answer you can act on.
A note on the numbers before anything else. Sales are indexed to the United States at 100 on net units; the absolute counts aren’t mine alone to publish. Refund rates are a share of gross sales in the same month, and I give the denominator every time, because most of them are tiny. Prices are the public App Store prices anyone can look up. Big Mac prices are the July 2026 edition. The mechanics of actually setting these prices — Apple’s tier table, the storefront list, the arithmetic — are in the first piece in this series, how to price an app by country, and the same arithmetic applied to ebooks, courses and SaaS seats is in the second, PPP pricing for digital products. This one is about whether you should bother at all.
Signals that contradict each other
Here’s the part of the report that matters. Price is what a customer pays divided by that country’s Big Mac in the July 2026 edition. Sales index is May 2026, United States at 100. A blank refund cell means I don’t have a rate I’d stand behind for that row.
| Country | App in Big Macs | Sales index | Refund rate |
|---|---|---|---|
| China | 0.82 | 159 | 6.9% |
| United States | 0.96 | 100 | 0.0% (0 of 118) |
| Taiwan | 1.47 | 19 | 4.5% |
| Japan | 1.58 | 14 | 6.2% |
| Mexico | 0.72 | 9 | — |
| Britain | 1.09 | 8 | 20.0% (2 of 10) |
| Saudi Arabia | 1.32 | 8 | 0.0% |
| Australia | 0.76 | 6 | — |
| Thailand | 0.96 | 6 | 14.3% |
| Germany | 0.97 | 5 | 16.7% (1 of 6) |
| Turkey | 0.40 | 4 | — |
Read the first two columns together and the discount theory doesn’t survive the contact — it fails in the wrong direction. Turkey has the cheapest app on my list in real terms — 0.40 Big Macs, less than half what an American pays — and it sits at the bottom of the sales table with Korea and the Philippines. Japan pays 1.58 Big Macs, nearly four times Turkey’s real price, and outsold it more than three to one.
Cheap didn’t sell. Expensive didn’t fail. The two columns barely know each other.
The income story doesn’t rescue it either. In the income-per-person column of the same dataset, the United States sits near $90,000, Saudi Arabia near $43,500, China near $23,400 — and my best market is the poorest of the three while my zero-refund market is the middle one. I’ll be honest about that column: it’s a blend I didn’t build, it disagrees with both nominal and PPP figures you’ll find elsewhere, and I use it here as an ordering, not a measurement. The methodology page says where each number in the dataset comes from and how far to trust it.
Germany’s row carries a second caveat. There is no German Big Mac price in this dataset — there’s a euro-area one, and I’ve argued at length that the euro-area Big Mac is a statistical fiction covering twenty economies that don’t share a price level. So treat the 0.97 as a regional approximation, not a German fact.
Why developers leave defaults alone
When I searched for whether anyone had answered this with real sales data, the two best-ranked results were Reddit threads — why don’t app developers adjust prices for different countries, asked and re-asked by users who’d noticed the same thing. Google putting forum posts at the top of a commercial query usually means it couldn’t find anything better to put there.
The threads answer the question, though — and their answer is roughly right. Apple gives you one checkbox that converts a US price into about 175 storefronts. Changing it means opening a table with 175 rows and having an opinion about each one. Nobody has an opinion about 175 things.
The second reason is that the feedback loop is broken. You change a price in Indonesia, and three months later you have a number that moved — and no way to know whether it moved because of the price, the season, a feature update, or nothing. My own report is exactly this problem in miniature: one month, one app, no control. I can show you what happened. I can’t show you what caused it.
The third reason is that the downside is vivid and the upside isn’t. Everyone’s heard of storefront arbitrage — buy through a cheap region, use it everywhere. On Apple’s platform that’s mostly bounded by billing address and payment method, so the leak is small, but “small leak” is a worse story than “someone is stealing from me,” and the worse story wins arguments.
None of which is a reason not to look. It’s a reason nobody does.
A price is a filter
Here’s the mental model I’d defend — and it isn’t mine.
The best-evidenced thing I’ve found on price and customer quality is a study from ACM EC 2012 that matched 16,692 Groupon deals against 7.13 million Yelp reviews. Deal-buyers do leave worse ratings at the businesses whose coupons they bought. The interesting part is the control: the same deal-buyers, reviewing other businesses where they had no coupon, were less likely to give one star and less likely to give five. Not harsher. Flatter. The authors explicitly rule out “discount customers are meaner” as the mechanism.
So price isn’t a dial that makes people more or less demanding. Price is a filter that decides who walks through the door. Cut it, and you don’t get the same audience in a worse mood — you get a different audience, with different reasons for being there, and some of those reasons work in your favour.
That reframes the whole developing-markets question. Charging less in Brazil isn’t a concession you make to poorer customers. It’s a decision about which Brazilians end up in your user base — a product question, not a charity one. And it’s why the first piece in this series argues you should reprice rather than discount: a permanent local price is just the price, while a markdown creates a reference point that the next buyer measures you against forever.
The expectation problem
Now the part I want to get right — because it’s the part where I could most easily fool you, and myself first.
There’s a piece of folk wisdom in this industry that cheap and free users complain the most. My data says the opposite. My two most expensive storefronts in real dollars — Britain at £5.99, about $8.07, and Germany at €5.99, about $6.85 — are the two with the highest refund rates. China, at $3.22, refunded 6.9%.
Britain’s 20.0% is two refunds out of ten. Germany’s 16.7% is one out of six. Write those denominators down before you quote the percentages at anybody — including me. One additional British refund would have made that number 30%, and one fewer would have made it 10%, and nothing about my app would have changed.
The only refund figure in the whole report with any statistical weight runs the other way and says nothing at all: zero refunds out of 118 in the United States, at $5.99, which is 0.96 Big Macs. That’s a real base and a clean result — and what it tells you is that the American price is fine. It doesn’t tell you anything about anywhere else.
There is peer-reviewed work pointing in the same direction as Britain. Luca and Reshef, in Management Science, tracked daily menu prices against ratings on a food-delivery platform and found that a 1% price increase produced a 3–5% drop in average rating (doi:10.1287/mnsc.2021.4049). Their reading isn’t that customers retaliate — it’s that price sets expectations, and the product then has to clear the bar the price just raised.
I want to be blunt about the distance between that paper and my spreadsheet. Restaurants aren’t software. A rating isn’t a refund. Their design compares a restaurant to itself over time; mine compares countries that differ in a dozen ways at once. And their sample is a platform’s worth of orders while mine, for Britain, is ten. My numbers do not support that conclusion. They’re consistent with it, which is a much weaker sentence, and the reason I went looking for the paper rather than the other way round.
Saudi Arabia is the row that keeps me from settling on any of this. Third-priciest storefront I have in plain dollars, 37% above America in Big Macs, and not one refund. If price alone raised the bar, that market should have been near my worst.
The better question
Drop “is this country poor” and ask three things instead.
What does this price equal locally? Not converted — equal. My app costs 0.40 Big Macs in Turkey and 1.70 in Vietnam, a 4.25x gap I never chose and never would have. That’s the check the App Store pricing calculator exists to run, and you can do the same comparison for any purchase in the purchasing power calculator. It takes about ten seconds and it catches the pricing you didn’t decide.
What does the price signal against local alternatives? A price is a position in a lineup, and the lineup is different in every storefront. Saudi Arabia at 1.32 Big Macs isn’t priced “too high” — it’s priced where it apparently belongs, and the zero refund rate is the closest thing to evidence I have.
Can the storefront convert at all? This is the one the Nordics taught me, and they taught it by humiliating a belief I held for free. Norway 1, Finland 2, Sweden 0, Denmark 0. High income, high card penetration, high App Store spend per head. It didn’t matter. Three points of index between the four of them, against America’s 100 — and no price change on earth was going to move that, because there was nothing there to move.
Which is also the case against getting too clever with any of this. The Big Mac is a burger; it carries local rent, local wages, and local beef tariffs that have nothing to do with software, and I’ve written up where Big Mac PPP breaks down at more length than most people want. It’s a smoke alarm — not a thermostat. It won’t tell you what to charge in Jakarta. It will tell you that your Jakarta price is four times your Istanbul price in real terms, which you didn’t know, and which you didn’t decide.
The market I priced best is the one I never thought about for a second. Saudi Arabia got its number from Apple’s rounding table — and Apple’s rounding table doesn’t have a theory of the Gulf. 无心插柳.
FAQ
Should I lower my app price in developing countries? My May 2026 report doesn’t settle it, and I’d be suspicious of anyone whose data does. China bought at 159 against a US index of 100 while paying 0.82 Big Macs against the American 0.96, which supports lowering. Saudi Arabia bought at 8 while paying 1.32 Big Macs with a 0.0% refund rate, which doesn’t. One app, one month, no control group.
Do cheaper markets refund more often? Not in my report. The highest refund rates were Britain at 20.0% and Germany at 16.7% — my two most expensive storefronts in real dollars — while China at 6.9% was among the cheapest. But Britain’s rate is two refunds out of ten and Germany’s is one out of six. Those bases are far too small to conclude anything, in either direction.
What’s a reliable refund number in this data? Exactly one: the United States, 0 refunds out of 118 gross sales in May 2026, at $5.99. Every other refund rate on the list sits on a base small enough that a single order moves it by several percentage points.
Why don’t more developers adjust prices by country? Apple’s default converts one US price into roughly 175 storefronts with a single checkbox, and overriding it means forming an opinion about each market by hand. There’s also almost no published outcome data — which is why the top search results for this question are forum threads rather than case studies.
Does high national income predict high app sales? Not in this sample. On an index where the United States is 100, Norway came in at 1, Finland 2, Sweden 0, Denmark 0 — four of the highest-income storefronts I sell in. That’s one month of one app, and it isn’t enough to generalise from. It was enough to stop me repeating that Nordic users are the world’s best customers.