Mixpanel Trends: What’s really happening with iOS in China?

alt text After months of anticipatory rumors and speculation, the iPhone finally launched on China Mobile this past January, and reception has been mixed. When Tim Cook announced at the end of February that both the iPhone 5s and the iPhone 5c had outsold all earlier iPhone models, we thought this deserved some interrogation, especially when it came to China. China is a huge opportunity for Apple – provided everyone plays their cards right. We dug into iOS data coming out of China to get to the bottom of some key observations.

alt text First and foremost, location is everything: about one third of the data comes from people on the country’s Eastern coast, and it’s clear that activity increases with proximity to major metropolitan areas. There’s also an extremely long tail of Chinese cities where iOS devices are used -- 40% of our data is generated by people in 783 cities. The concentration of data in major cities is likely most correlated to income, but may also be linked to Apple and China Mobile’s strategy for 4G: the 4G network is slowly being rolled out across the country, starting with urban areas. It’s already accessible to people living in Shanghai, Beijing, Guangzhou and Chongqing, who may have more incentive to move over to iOS. Chongqing is notably absent from the list of top iOS-generative cities. There are a few reasons why this may be the case: this is probably a geo-locational issue with our IP lookup database. Because Chongqing is so close to other major metropolitan areas, users may be identified as belonging to neighboring cities and regions.

Beijing, Shanghai, and Guangzhou are frontrunners in iOS activity; 4G accessibility may play into this, but it’s more likely due to the fact that these cities are the commercial, financial, and manufacturing centers of mainland China. The people living in and around these cities are wealthier than most and more likely to purchase high-end technology (for both utilitarian and status purposes) than their counterparts in smaller cities, towns, and rural areas.

alt text On a regional level, Guangdong sticks out as one of the most prominent and active areas for iOS users. Guangdong contains both Guangzhou and Shenzhen; the region is one of the most prosperous in China, with a robust manufacturing industry. The area is the closest region to Hong Kong, a city with a ton of money, high-tech, and many foreigners.Shenzhen is also home to Foxconn, an electronics manufacturing company that makes many of Apple’s mobile devices. While Foxconn employs somewhere around 500,000 people, many of whom live on-site at the factory, it’s hard to say whether this is the same population that’s generating iOS data. But it seems safe to assume that Foxconn’s presence can only serve to raise awareness of Apple and its latest offerings within Shenzhen as a whole.

Notably minor on the list of iOS-generative regions is Tianjin, which has the greatest GDP per capita in China but only sends about 2.5% of China’s iOS data. This may be more reflective of its modest population than purchasing power or interest in iOS devices; Tianjin has only about 13 million residents, which is close to 1/8 the population of Guangdong and less than half the population of Shanghai or Beijing.

That said, population and data usage don’t always correlate. Shanghai and Beijing are almost equal in population and have nearly parallel PPP – logically, their iOS usage is also neck-and-neck. About three times as many people live in Guangdong than do in Beijing or Shanghai, and the region’s iOS usage is comparable. This may be due to the fact that Guangdong is a less wealthy city, with GDP per person about 60% of what it is in Beijing or Shanghai (to say nothing of Tianjin).

alt text When it comes to the iOS devices themselves, people in China primarily use older iPhone models, with the iPhone 5 taking up significant real estate at 19%. All of the iPhones represented above are CDMA/GSM models, and are specific to a small number of networks within China; China-CDMA phones are usually not functional outside of their origin country. Despite the China Mobile release and marketing efforts for the newest iPhone models (more on that momentarily), preferences clearly lean toward older models such as the iPhone 4S (2011) and iPhone 5 (2012). There’s also a long tail of devices running iOS that don’t fall into the above categories; 22% of iOS data is generated by 25 different types of devices, none of which are represented in the chart above.

alt text When the iPhone 5s and 5c were released in December 2013, there was a jump in China’s iOS usage, but the devices pushing that increase weren’t the usual suspects. Even when Apple had a huge marketing campaign for the 5c and 5s, the greatest growth actually came from the older iPhone 5 -- which of course became cheaper as soon as the newer models came on the market. A similar trend was seen around the January China Mobile release: volume grew, but the devices driving the data spike still weren’t the 5s or the 5c. People still were drawn to the newly-cheapened iPhone 5, which drove most of the growth seen around that time. One notable area of growth was in iPad usage: this does suggest that when Apple calls attention to itself, people pay attention – interest in (and usage of) Apple products grows, even if that doesn’t translate directly into sales.

So what does this mean for the future of iOS in China? What’s clear is that extra awareness in the marketplace begets returns: this was seen during both the iPhone 5s/5c release, and during China Mobile’s rollout. Though Chinese consumers didn’t show an immediate interest in Apple’s latest iPhone models, interest in Apple devices overall did seem to grow with each new release. It will be fascinating to watch iOS activity in China as 2014 continues, and to see whether iOS usage increases as 4G becomes more widely available. We predict that the release of the iPhone 6 -- rumored to be shortly on its way -- will catalyze sales of the models it’s intended to replace, further growing Apple’s Chinese customer base.

Community Tip: All About Time

In this Community Tip we discuss all aspects of time in Mixpanel, including how to set your project timezone, send time properties, and handle timestamps when exporting or importing data. Using these directions, Mixpanel can help standardize how you handle time to ensure your data reflects accurate timestamps. In today’s connected world, a startup can be headquartered in Berlin, Germany, have a core user-base in Beijing, China, and be analyzing their data through an analytics company in San Francisco, California, USA. With all of these different locations in play, seamlessly handling time data being received from around the world can be intimidating. Luckily, with a few of the following best practices, you can ensure your Mixpanel data is time agnostic to the physical location of your users. Setting Your Mixpanel Timezone When you create a new Mixpanel project, your project timez...

Community Tip: Implement Mixpanel in Swift Apps

This Community Tip will describe how to implement the Mixpanel iOS SDK, written in Objective-C, within your Swift app. We will walk through integration using CocoaPods, library set-up, provide code samples, and ultimately save you development time. Requirements: CocoaPods This guide assumes you are using CocoaPods since that is the recommended way to integrate your app with the Mixpanel iOS library . For assistance integrating Mixpanel into Swift projects without CocoaPods, please follow the instructions to build the Mixpanel iOS library from source . Please note the below instructions for integration will only work for Swift apps targeting iOS 8.0 or higher — if you are targeting any versions below iOS 8.0 please jump to the final section of this guide. Integration with CocoaPods With any version of CocoaPods greater than 0.36, integration of Objective-C libraries within...

How HubSpot grows a product into something you can't live without

Did they even open my email? Did they check out my link? Marketers have had tools to answer these questions for a long time. Sidekick , a project from HubSpot, brings this kind of email tracking to one-on-one email communication. For HubSpot, it's a little outside their wheelhouse. Over the last nine years they’ve developed some of the largest and most popular software for marketers. Sidekick is part of a decision to broaden their scope and serve non-marketers. And it's built on the freemium model. "The whole idea from the beginning was to go for really big scale that millions of people could use for free," says Dan Wolchonok. "And then to have a seamless path for those who want to pay for power features." Dan is a senior product manager at HubSpot. It's his job to get those millions of users. Which is less about finding them and more about keeping them around. Being active ...

Mobile A/B Testing: Walkthrough & Best Practices

Using Mobile A/B testing, you can test changes to your app without writing code. You can measure the impact on any conversion, like signups or purchases. In this Community Tip, we'll provide a walkthrough and best practices for running A/B tests in your iOS or Android app. Setup To get started with Mobile A/B testing, all you need is the Mixpanel SDK installed and initialized in your app. If you haven’t done this step yet, you can jump over to our iOS Integration Docs or Android Integration Docs for instructions on installing Mixpanel. To work with complex A/B tests which go beyond modifying UI elements, such as changing view flows, modifying defined constants, and more, you will need to implement developer tweaks ( iOS / Android ) in your app. Tweaks allow you to adjust actual variables in your code from the comfort of your Mixpanel dashboard. Once you add a Tweak to your...

Geolocation Error on 7/9/2015

Incident Summary: No geolocation was being performed on events and people requests between 5:00PM PST 7/8/2015 to 1:30AM PST 7/9/2015. This affected all customers. Timeline: On 7/8/2015 at approximately 5:00 PM PST, a change was deployed to consumers that migrated our geolocation system from a Python module to a C replacement as part of our systems optimization plan. After this point, all geolocation requests made to the Maxmind DB began failing. Around 7/9/2015 1:00AM PST, Mixpanel began reverting the change and by 1:30AM, all consumers were reverted back to the Python Maxmind DB reader, which resolved the issue. Root Cause: The migration from a Python module to a C replacement proved not to be a direct drop-in replacement and generated errors (TypeError) when fed a value other than a string. We typically send IPs in the format we receive them, which can sometimes be a long...

A/B testing comes to Android

So you've got an idea. "We have all these awesome features most users just haven’t tried yet. I bet more people would use them if they couldn't skip the tutorial. " "I know more people would create an account if our button just said 'Get Started'." "Our game's too easy. People want a challenge. If it were a bit faster, it would keep users coming back." Great, but a good idea alone isn’t gonna cut it. Especially when opinions differ. What you really need is cold, hard data. You need to put that hypothesis to the test and see how it fares with real users in the real world. That's where A/B testing comes in. Previously, you've been able to use Mixpanel to experiment and improve your iOS app. And we've seen how useful it has been to our customers. Today, we're expanding Mixpanel A/B Testing to Android. All the capabilities that enable you to experiment and improve your iOS a...

Community Tip: Benefits & Best Practices for Cross-platform Apps

In this Community Tip, we highlight the benefits and best practices for accommodating all of your varied products, environments, and platforms under a single Mixpanel project. Using this guide, your team can decide how to best track your cross-platform app data within Mixpanel. If you're one of the many users working with cross-platform apps, you may have wondered whether it's best to combine everything into one Mixpanel project or split everything up and analyze it separately. We've found that creating separate Development and Production projects to avoid cluttering your Production data is a great best practice, but the question of what to do with cross-platform Production apps is a matter of preference. The good news: Mixpanel is flexible enough to handle cross-platform scenarios in any configuration you'd like to implement. The choice is yours, and each approach comes with its...

Introducing Codeless Mobile Analytics

Tracking how users are interacting with your app just got easier. Now you can get Mixpanel's advanced mobile analytics without writing a line of code. Let's face it, everyone's dealing with limited developer resources. If you've got to add a tracking request to the growing queue of dev needs, it'll probably take some time. With Codeless Mobile Analytics , you can focus your development time on what matters - making your app better. With a simple point and click setup, Codeless makes tracking events a breeze. Want to know if people are using a new feature? Skipping that tutorial? Changing a setting? After you've installed the Mixpanel SDK you can just fire up Codeless Mobile Analytics (it's that icon on the bottom of your sidebar). Select your platform - iOS or Android. Then connect to your app using either a phone or an emulator. That's it. From right there in your bro...

How Cozi got a 38% increase in signup completion

Tara Pugh, product owner of Cozi, stopped by our June Office Hours to explain the process for improving their UX through many small, data-driven steps. Cozi helps bring order to the chaos of modern family life with an app that gets the entire family on the same page with shared calendars, to-do lists, and shopping lists. Digging into their conversion funnel data with Mixpanel, they identified opportunities to improve user experience by streamlining the account creation flow. Testing hypothesis after hypothesis, Tara and the team at Cozi were able to incorporate bits of learning into the flow: lighter background, different calls to action, pre-populating name and email forms. No single change resulted in a huge increase in conversions, but all together, the improvements raised the signup completion rate from 55% to 76%—a 38% increase. Check out Tara's talk for all th...