Over the last year, OpenSignal has published a few blog posts on some of our more interesting findings in the Indian mobile scene, but our plan has always been to do a more thorough analysis of one of the largest and most complex telecom markets in the world. Well, today we’ve accomplished that goal. We’re proud to debut our State of Mobile Networks: India report, the first of what will be many detailed looks at the mobile data experience in the subcontinent.
India proved to be a very challenging market to analyze because it’s so fractured. Nearly a dozen operators offer service in India, but not all of them compete head to head in every region. India is divided into 22 telecom circles, and each operator runs networks in a different set of circles. Normally that would make any national comparison between operators impossible, but in recent years a handful of operators have emerged as national players in either 3G or 4G services, and it’s those operators on which OpenSignal has focused its national analysis. We didn’t stop at a national comparison though. We selected four significant telecom circles and conducted a regional LTE analysis of each one. You can read our detailed analysis and see our interactive charts in the full report here.
Our report debuts at a rather controversial time for the Indian telecom sector. The launch of Reliance Jio and its inexpensive and virtually unlimited data plans have set off a new wave of competition and a war of words among India’s top operators. Reports such as this one are therefore drawing a lot of attention and generating a lot of arguments. As our report goes live, we felt we needed to be transparent about our methodology and to explain how we’ve used it to measure the mobile experience of smartphone users in countries across the world.
First, OpenSignal uses crowdsourcing to collect data. Users download our app, which allows them to perform speed and signal tests from their smartphones both actively and in the background. All of that data is sent to OpenSignal stripped of any personal information (we protect our users’ anonymity) where it’s collected in our databases and used as the jumping off point for our reports and detailed online and in-app coverage maps. For this report alone we gathered more than 1.3 billion measurements collected by 93,464 smartphone users across India.
In short, we have access to a lot of data. But collecting a huge volume of data isn’t everything. Painting an accurate picture of the mobile data experience in a country also depends on how you interpret that data. OpenSignal has adopted a methodology for collecting and analyzing data that provides a holistic look at mobile data services and best represents what the typical consumer sees on a day-to-day basis.
One example of this approach is how we aggregate our speed test results. In India, there is a small subset of users that conduct a massive volume of speed tests (thousands a month compared to a handful for the typical user). Those tests are often run when the network is at optimal performance. If our methodology were to count each speed test equally, we would have a small handful of users skewing our results, often in favor of a particular operator. OpenSignal’s methodology avoids that issue by averaging the speed tests of each user, and then counting each average equally. That ultimately fits well with our philosophy. We want to examine the experience of all subscribers on a particular operator, not the experience of a few very active testers.
Another example is how we treat dual-SIM data, which is especially important in India where many people subscribe to more than one operator’s service. The problem here is assigning the proper test to the proper operator, since the same phone is connecting to different services. It’s a problem that OpenSignal has solved through the careful parsing of our data. We’re able to identify the individual network our users connect to whenever conducting a test and then assign that measurement to the correct operator.
If you want to read more about how we collect and analyze our data, we encourage you to read the methodology section of our website. And be sure and let us know your thoughts on our India report in the comments section below.