Mobile analytics has become a necessity now with the boom in mobile web usage. In the last quarter, more smart phones were shipped to India than ever before. Mobile analytics studies user behavior, content views, downloads, and content shared on social media. Data collected as part of mobile analytics typically include page views, visits, visitors, and content tracking.
Another area mobile analytics studies is the mobile device itself. People use various handsets, and each of these handsets uses a different platform. Thus device-specific information such as the model, manufacturer, screen resolution, device capabilities, service provider, and preferred user language is also important. You can imagine how diverse the generated data can get.
Most of the big BI players offer mobile analytics as a minimal feature with their social analytics products. Most companies know how to make sense of structured data, and are comfortable slicing and dicing treated data.
However mobile analytics works with unstructured data and analyzes unorganized text and fragmented data sources. Unstructured data is said to have a short ‘analytic life’. This data needs quick, almost real-time analysis for it to be useful, and quick reaction as well. Some of the data fields may be missing, and must be compensated for by the analytical model. Mobile analytics is not stringent in this aspect.
What mobile analytics will do for you will depend on whether you are an operator or a publisher.
Mobile operators want to know which customers can be moved to a higher-value subscription plan. Or which customers (depending on what they are accessing on their mobile devices) can be targeted for 3G realtime streaming plans. Niche customers can thus be made the focus of marketing initiatives with mobile analytics.
For publishers, mobile analytics can help decide how the website should be structured to optimize content views. The positioning of an ad on the screen has to be optimized to increase click-throughs on all handsets. If the ad is not clearly visible on a particular screen resolution, and the targeted handset is a crucial part of the marketing campaign, there will be a problem. Mobile analytics can help decipher which links get more hits, and the website can be optimized to take advantage of that.
For shopping websites, mobile analytics can draw attention to the pattern by which shopping carts get dropped if the payment button is not located within a certain radius of eye focus. Mobile analytics can thus help reduce shopping cart abandonment.
Three categories of analytics
Mobile analytics dashboards can help figure out what you are already looking for, what is called the ‘known-known’. You have a KPI, and you use mobile analytics to meet it.
The next one is ‘known-unknown’, which answers the question ‘why?’ For example, you know that an area shows a churn after 30 days, and you need to know the factor behind it. Mobile analytics will help you do that.
The third category of analysis is the ‘unknown-unknown’, which is discovery analytics. The system will throw up insights such as teenagers disconnecting from a certain plan after a certain period when the benefits (through some marketing promotion) expire. The operators will not want this age group.
Mobile analytics and the cloud
Cloud computing is an excellent way for companies to start working with mobile analytics and see what it can do for them. Mobile analytics is independent of larger BI systems and does not need complex analytics to support its functions. Cloud mobile analytics depends on the size of the company and the maturity of their IT systems.
Smaller companies prefer the cloud, as they lack the expertise to manage it themselves. Huge investments in recruiting new people to work on it can be avoided with the use of mobile analytics. Larger organizations with a bigger IT staff and infrastructure find it easier to incorporate mobile analytics into their web of BI systems.
We have had cases where larger companies wanted mobile analytics hosted on the cloud. This is generally for a test run. The company first tries mobile analytics on not-so-crucial data. When satisfied with the results, larger investments are made. And personally I think mobile analytics will have a higher level of adoption on the cloud.
About the author: Anandan Jayaraman has worldwide responsibility for Connectiva's corporate and product strategy, product management, marketing and sales consulting. He has extensive management experience in leading change, managing global business development, transforming go-to-market models and driving revenue growth in a number of enterprise software companies including SAP, Epiphany, and Siebel Systems. Jayaraman holds an MBA from the Indian Institute of Management, Calcutta and is a graduate of the Stanford Executive Program for senior executives at the Graduate School of Business, Stanford University.
(As told to Sharon D'Souza)
This was first published in September 2011