Testing Scenario for Mobile Business Intelligence

Overview
Mobile devices have  evolved over the years affecting online access to the point where a majority of Internet access is being conducted via mobile handhelds today. To maximize this large-scale market penetration, industry segments have made mobile apps an integral part of their marketing strategy.

This accelerated activity has resulted in the growth of Mobile Business Intelligence (Mobile BI) which has transformed the business landscape from a ‘wired’ to a wireless world. Mobile BI is far more versatile than other forms of intelligence as it can be woven closely into people’s movements, work, conversations, meetings, discussions and fun time.   Leveraging this versatility, Mobile BI is a package that uses existing BI applications to make informed decisions in real time.

What does Mobile BI mean?
Mobile BI is the ability to access BI-related data such as KPIs, business metrics, and dashboards on mobile devices.  Information delivery suitable for Mobile user interface has been made possible by the various applications provided by Mobile BI.

Basic Workflow:
The below diagram illustrates the flow of a Mobile BI architecture

Comparison of Mobile BI Apps with other options:
The table below depicts the difference in effort, interactivity and the ability of various channels in comparison with Mobile BI apps

Important points to analyse before opting for Mobile BI Strategy:-

1. See if mobility works for you from a business perspective.
2. See if you have the right IT infrastructure to support Mobile business?
3. Do you have the ability to leverage hybrid (Native + Web-browser) applications to cater to different requirements?
4. Can you account for security based-unique considerations required before its implementation?

Best practices for designing the interface in Mobile BI Apps:

• Avoid dashboard burgeoning – A dashboard should be designed in a way that it provides the necessary information for the decision-maker. It should give easy access to information in a standardized format. The dashboard should be business driven and not purely technology driven.

“More is not better”- An abundance of KPIs are not better and leads to overcrowding of the dashboard.

• Refrain from writing in a smaller form factor – The font size should be in a readable size to ensure that the user does not have to strain his eyes to see the font. The placeholders can be used on the small text or common form inputs like login forms or search boxes.The headings should be kept short to ensure that it does not push the content down the page or out of the frame for users.

Strategy for developing and testing Mobile BI Apps:

• Build once, deploy anywhere strategy

With a variety of development languages and approaches, it is apparent that there are varied ways to build regardless of the language. The key is to avoid building the code in a way that it has to be modified for each specific environment. To achieve a `build once, deploy anywhere’ status, it is best to exclude environment specific resources in the final application build so that it is interoperable with the environment.

The environment specific artefact can be deployed to the container separately from the WAR and at any time.

There are some server-side platforms available for running such applications. E.g.: Oracle WebLogic.

• Account for New mobile scenarios

The evolving mobile solutions landscape has led the field of mobile testing to a challenging level and it has become important for  QA managers to understand the unique testing needs of the Mobile BI segment. The QA needs to identify the needs/establish requirements of the various users and design scenarios, taking into account, aspects like interoperability, security and reliability of the mobile app.

Pros and Cons of Mobile BI:
Pros:

1. User can pinch, swipe, and tap to easily interact with and analyse company data.
2. Cloud based mobile solutions increase collaborations.
3. Allows the user to send notifications in various forms like email or text messages.
4. Provides sales and field support representatives with the data they need to answer customer questions on the spot.

Cons:

1. Too much reliance on mobile devices and tablets increases the risk for mobile computing.
2. Devices are expensive and replacement costs are steep.
3. Mobile BI apps in general are not very interactive and restrict users from drilling down into data.

Example Snapshots of some BI Apps:

Widely used products for Mobile BI:

1. Oracle business Intelligence Mobile:- http://www.oracle.com/us/solutions/business-analytics/business-intelligence/mobile/overview/index.html
2. Tableau Mobile Business Intelligence :- http://www.tableausoftware.com/solutions/mobile-business-intelligence

Mobile BI QA
The aim of testing BI applications for mobile is to achieve credible data and a good design display with a user-friendly interface. It is also important to ensure that the back-end of a mobile BI system is able to handle the processing load to display data in a timely manner.

How to Devise a Mobile BI Test Process:

1. Validate the data required and identify data sources
2. Identify and decide the category of Mobile BI App to be implemented, i.e.:

i.    Mobile Browser Rendered App
ii.    Customized App
iii.    Mobile Client App

1. Understand data, unearth related problems early and identify boundary value conditions for test scenarios.
2. Set up the acceptance criteria as per data accuracy / consistency and benchmark the performance time for rendering of reports.
3. Test plan to identify the scope of testing and prepare test data and testing techniques.

Tools which can be used for Testing BI Applications on Mobile:
BI Application testing encompasses a lot of systems like data mining, statistical analysis and graphically-rich dashboards.  To test all these components tools are available:

1. Eggplant: This is a tool by Testplant. It uses image recognition technology to instantaneously test multiple aspects of Business Intelligence systems. It is also flexible to operate across multiple applications which helps for better testing.
2. Sikuli: Sikuli uses image recognition technology to identify and control GUI components. URL: http://www.sikuli.org/

References:
http://searchcio.techtarget.com/essentialguide/Strategic-business-intelligence-for-a-mobile-future#guideSection1
http://www.tableausoftware.com/learn/whitepapers/5-best-practices-mobile-business-intelligence
http://en.wikipedia.org/wiki/Mobile_business_intelligence
http://wiki.scn.sap.com/wiki/display/BOBJ/Architecture+and+Workflow+Diagrams