When Selling Virtual Products Abroad, Don’t Put Prices On Autopilot

If you have a physical product that you want to sell in more than one country, determining the price in different markets can be challenging. You might have to open an office in each country, or at least hire a consultant to assess local demand and analyze the competition.

But if you have a virtual product — say an app for a mobile phone — setting the price for it in different countries is easy. Using the individual exchange rate, the app store instantly will convert the price from your home country to any of the world’s many currencies.

This is, very likely, how prices are set for most smartphone applications sold in different countries. As developers prefer to spend time solving technical challenges, it is all too convenient to leave the responsibility of currency calculations and pricing to Apple or Google or some other virtual marketplace.

But is this the best approach when selling internationally? Is there a more profitable way to price virtual products sold in different currencies?

We explored these questions in an experiment that was both a real-world business trial and an academic exercise. We wanted to see whether we could boost revenue for a virtual product, Root Checker Pro, an app that helps Android users customize their phones. The app is sold through Google Play — the app store for Android devices — in more than 130 countries.

For our experiment, we selected six different currencies — Australian dollar, Canadian dollar, British pound, Mexican peso, Malaysian ringgit and Saudi riyal. Over six months, we charged various prices for the app in each of the currencies to see how sales and revenue would respond.

We questioned whether the common practice in the U.S. of setting prices with endings of 99 cents would resonate internationally. Whatever psychological advantage comes from this pricing strategy is lost in automated currency conversions. We moved prices for the app up and down from 99-cent endings to see what would happen. We saw no correlation between price endings and sales in other currencies. It’s likely that people in other countries have become accustomed to seeing unusual endings on prices as a result of automated currency conversions, especially in app stores, thus removing the common U.S.-based psychological associations with 99 cents.

We also questioned whether there might be payday spikes on days when people get their paychecks, which we assumed would follow the U.S.-based model of being around the middle and end of the month.

Again there was no pattern of increased purchases or reduced price sensitivity at the middle or end of the month across currencies. One possible explanation is that pay periods differ across markets because of differences in national banking systems and business practices.

Another question we hoped to answer is whether there is a universal threshold above which people would stop purchasing the app. Again, across currencies, there was no uniform pattern showing a significant decrease in purchasing at specific price points.

We did identify two statistically significant patterns in the experiment. For all currencies in aggregate, purchasing increased on weekends. On closer examination, we found that the phenomenon was mainly driven by two of the six currencies — the British pound and Canadian dollar. In these two markets, we found that on weekends, customers were less price sensitive.

In most of the currencies we tested, purchasing fell as prices increased. There was one important exception — the Saudi riyal. The average number of purchases stayed about the same, even as we increased the price three times its base level. During the experiment, overall revenue from sales in Saudi Arabia quadrupled.

We don’t know for certain why Saudi users were willing to pay so much more. We suspect that Saudi Android users are relatively wealthier, which could make them less price sensitive. Also, we understand that Android phones available in Saudi Arabia can be of lower quality, and thus there would be more interest in an app that improves the devices. Regardless of the reasons for Saudi users’ purchasing behavior, it was clear that they were willing to pay more.

App developers and others who sell virtual products in different countries may be making a big mistake when they let app stores and other online marketplaces set prices through currency conversions.

Putting pricing on autopilot means you are assuming the rest of the world is just like your home country. Our experiment demonstrates that in different countries, consumers do, indeed, behave differently. Learning about other markets and experimenting with a range of prices can boost sales and revenue.