Sales and analytics
Many downloads, few purchases: App Store conversion by country
Thousands of downloads from India, Brazil or Indonesia and hardly a sale? This guide shows how to measure purchases per download country by country, how to tell a real pattern from noise, and how to test whether a lower local price turns those downloads into buyers.
Why a country downloads a lot but rarely buys
When a country downloads your free app often but rarely pays, the usual suspect is price against local purchasing power. Apple converts your base price at exchange rates and adds local tax, so $4.99 costs about the same in dollars everywhere while incomes differ a lot. Measure purchases per download by country first, then test a lower price where it lags.
Price against what people earn
Unless you set prices per country, Apple equalizes your base price into all 175 storefronts: same value at Apple’s exchange rate, local tax included, snapped to a price point. That is fair in currency terms and blind to incomes. A $4.99 unlock is an impulse buy in the US and a considered purchase where incomes are a fraction of American ones. Downloading is free everywhere, so the gap shows up at the paywall. That is the case purchasing power parity pricing is meant for.
Payment methods and currency
To buy anything, a customer needs a valid payment method on their Apple Account, and what is accepted varies by country. Apple publishes the list in Payment methods that you can use with your Apple Account: in India, for example, it includes UPI and net banking alongside cards; in Türkiye, mobile billing with supported carriers. If many users have no payment method set up, part of the gap has nothing to do with your price.
Sometimes it isn’t the price at all
- Localization. A paywall in English only, or a value proposition that assumes a US context, converts worse anywhere English is a second language.
- Low-intent traffic. A press mention, a featuring or a broad ad campaign in one country can bring many curious downloads that were never going to buy.
- Family Sharing. Apple counts Family Sharing downloads in App Units for free apps (see Sales and Trends metrics), which slightly inflates downloads in markets where families share accounts.
- Product fit. A feature that matters in one market, such as a local currency or calendar format, may be missing.
That is why the number to watch is not raw purchases but purchases per download, compared against your own app’s average. It takes the size of the country out of the picture: of the people who tried the app here, how many paid?
How to measure purchases per download by country
Everything you need is in Apple’s daily Summary Sales report, one of the Sales and Trends reports. Each row is a product, a country and a transaction type, with Units, Product Type Identifier, Country Code, Parent Identifier (the app SKU an in-app purchase belongs to) and, for auto-renewable subscriptions, Subscription (new or renewal). You can download it in App Store Connect under Sales and Trends, or through the API with `GET /v1/salesReports` (report type SALES, subtype SUMMARY, frequency DAILY) and your vendor number.
- Download the daily summary reports for the same 28 days (or 90) for every country. Apple says daily reports are available the following day and kept for one year (report availability).
- Downloads: sum
Unitsof your app’s rows whose product type is a first download —1,1F,1Ton iOS,F1on the Mac. Leave out re-downloads (3,3F) and updates (7,7F,7T,F7). - Purchases: sum
Unitsof in-app purchase rows (IA1,IA9,IAYand their Mac variants) whoseParent Identifieris your app’s SKU. For subscriptions keep only rows whereSubscriptionis new; renewals are the same buyer paying again. - Keep negative units. Apple reports refunds as negative
Units, so a refunded purchase cancels itself out. - Group by
Country Codeand divide purchases by downloads. Then divide total purchases by total downloads for the app’s average. - Put every country next to the average. The ratio — this country’s rate divided by the average — is the number you act on.
A worked example
A free app with a $4.99 “Pro” unlock, 28 days of summary reports, prices equalized by Apple everywhere:
| Country | Downloads | Purchases | Purchases per download | Against average (2.3 %) |
|---|---|---|---|---|
| United States | 4,200 | 168 | 4.0 % | 1.7× |
| Germany | 1,100 | 41 | 3.7 % | 1.6× |
| Japan | 450 | 25 | 5.6 % | 2.4× — buys more often |
| India | 3,600 | 18 | 0.5 % | 0.2× — many downloads, few purchases |
| Brazil | 1,900 | 19 | 1.0 % | 0.4× — many downloads, few purchases |
| Indonesia | 1,300 | 9 | 0.7 % | 0.3× — many downloads, few purchases |
| Türkiye | 700 | 10 | 1.4 % | 0.6× — watch |
| Portugal | 25 | 0 | 0 % | too few downloads to read |
| All other countries | 2,725 | 78 | 2.9 % | 1.2× |
| Total | 16,000 | 368 | 2.3 % | — |
India, Brazil and Indonesia together bring 6,800 downloads — 43 % of the total — and 46 purchases. Had they converted at even half the app’s average (1.15 %), that would be about 78 purchases, so roughly 32 sales a month are being lost in those three markets alone. Japan is the opposite case: fewer downloads, but buyers there pay well over twice as often, a hint that the price may have room to go up. Portugal’s zero means nothing: 25 downloads cannot tell you anything.
Thresholds that keep the comparison meaningful
- A minimum of downloads. Ignore any country with fewer than about 30 downloads in the period; with a higher-priced product, wait for more.
- A minimum share. Also skip countries that bring less than about 2 % of the app’s downloads; a price change there cannot move your revenue anyway.
- Compare against your own app. Your app’s overall purchases per download is the baseline, not a number from another app or a blog post.
- Clear gaps only. Under half the average is a signal that the price may be in the way; more than twice the average, with at least 10 purchases, is a signal there may be room upward. Anything in between is normal variation between markets.
- Same window everywhere. Every country should cover the same days, ideally whole weeks, so weekends and paydays fall equally.
- One app at a time. Each app has its own average; mixing a cheap utility and an expensive pro tool averages away exactly what you are looking for.
A rule of thumb for counts: random variation on a count of n events is roughly √n. Nine purchases could easily have been six or twelve; 168 could have been 155 or 181. That is why a gap has to be big and the counts reasonably large before it means something.
What not to conclude from small numbers
- “Nobody in this country pays.” Zero purchases from 40 downloads is weak evidence; zero from 4,000 is strong.
- “The price cut worked” after three days. Daily sales swing with weekdays, holidays and featuring; give any change whole weeks.
- “Conversion is low, so the price is wrong.” Rule out localization, a broken purchase flow in that country and low-intent traffic first.
- “More units means more money.” A cheaper price that sells 30 % more units can still earn less. Proceeds decide, not units.
What to do: lower the price where it’s in the way
The usual fix is not a country-by-country guess but a regional tier: a group of countries that pays a percentage of your base price, say 50 % or 40 %. Before you cut, know how many more purchases the lower price needs just to earn the same. Proceeds are a share of the price, so the break-even is simple:
| New price (share of today’s) | Purchases needed to earn the same | In the example, India (18 today) |
|---|---|---|
| 75 % | 1.33× | 24 |
| 60 % | 1.67× | 30 |
| 50 % | 2× | 36 |
| 40 % | 2.5× | 45 |
| 30 % | 3.33× | 60 |
Every purchase beyond that line is extra money. India at 0.5 % needs to reach about 1.25 % at a 40 % price to break even — still well under the app’s 2.3 % average, so the bet is plausible. Actual results depend on Apple’s price points: the new local price snaps to the nearest valid point, so the real share may be a little above or below your target.
Then test before you roll it out widely. Change the price in a few of the lagging countries, leave the rest alone as a comparison group, and compare per-day sales before and after. The method is described step by step in testing App Store prices by country. Judge the result on proceeds and on proceeds per download, not just units.
A step-by-step plan
- Pull 28 days of daily summary sales reports (90 days if your app is small) and compute downloads, purchases and purchases per download for every country, as above.
- Drop countries under about 30 downloads or 2 % of the app’s downloads.
- Mark countries under half the app’s average as candidates for a lower price, and countries above twice the average with 10+ purchases as candidates for a higher one.
- Check the non-price causes for each candidate: localization, payment, traffic source, a recent featuring.
- Pick two to five similar candidate countries and work out the break-even for the price you have in mind.
- Run a country-split experiment for at least two full weeks, ideally four, with the other countries as the comparison group.
- Compare proceeds and proceeds per download per day, before against after, adjusted for what the comparison group did.
- If it pays, move those countries — and similar ones — into a lower regional tier. If it doesn’t, roll back.
- Re-run the comparison every quarter. Exchange rates drift, Apple adjusts prices, and your audience changes.
Finding these countries with RepriceCat
RepriceCat’s Sales page does the counting above for you. Add your vendor number and a key with the Admin, Finance or Sales role, and it reads Apple’s daily summary sales reports (Apple publishes each day about two days later) and shows, per country, downloads, purchases, units, proceeds, share and trend over 7, 28 or 90 days, as a sortable table and a world map. On top of that it writes pricing hints, using the thresholds from this guide:
- “Many downloads, few purchases”: a country with at least 30 downloads and 2 % of the app’s downloads, whose purchases per download are under half the app’s average. A lower price there may help.
- “Buys N× more often”: purchases per download over twice the average, with at least 10 purchases. There may be room for a higher price; an experiment can tell.
- Countries growing (+50 %) or slowing (−40 %) between the two halves of the period, and a warning when one country carries 60 % or more of the revenue.
- Both pricing hints link to the app’s prices, where you can move the lagging countries into a cheaper regional tier — prices snap to Apple’s price points and you review every country before anything is sent — or run a price experiment in just those countries first.
The hints are signals from the shape of your sales, not certainties; the experiment is what tells you whether a new price pays. See how tiers work on the regional pricing page, or start the free month and look at your own countries today.
Frequently asked
Why do I get lots of downloads but no purchases from some countries?
Most often because the price, converted from your base price, is high against local incomes. Missing payment methods, an untranslated paywall and low-intent traffic from a featuring or ad campaign can also play a part, so rule those out before cutting the price.
How do I see App Store conversion rate by country?
Download the daily Summary Sales reports from Sales and Trends, sum first-time app downloads and in-app purchase units per country code, and divide purchases by downloads. App Store Connect Analytics also has a Download to Paid cohort view you can filter by territory.
Is the App Analytics conversion rate the same as purchase conversion?
No. Apple’s Conversion Rate in App Analytics is downloads divided by unique device impressions, so it measures the store page, not purchases inside the app.
Should I lower my app price in some countries?
Only where many people download but few buy, the counts are large enough to trust, and the break-even looks reachable. Test the lower price in a few countries first and keep it only if proceeds rise against the countries you left alone.
How many downloads do I need before the numbers mean anything?
As a rule of thumb, at least about 30 downloads and 2 % of the app’s downloads in the period for a country to be compared at all, and more for an expensive product. Random variation on a count of n is roughly the square root of n, so small counts swing a lot.
Does lowering a subscription price affect existing subscribers?
Yes. Apple renews existing subscriptions at the lower price automatically and does not let you keep the higher price for them, so include that in your calculation.
Try it
Reprice every territory from one base price
Tiers, price-point snapping, one review table, one confirmation — through the App Store Connect API. First month free, then $1 per app per month.