A major shift has occurred in business analytics as critical decision-making has moved beyond corporate walls and corner offices to the field, the customer site and pretty much any location with a cell phone signal. Smart phones and tablet computers have been the main driver behind this shift because they give users flexibility in where and when they consume information. In response to this trend, business intelligence vendors have made a significant shift toward supporting analytics on mobile devices. In fact by 2013, Gartner predicts 33 percent of BI functionality is expected to be delivered to hand held devices.
While there is no doubt that the future of analytics will include a mobile focus, it is critical that companies take a long look at their existing data and infrastructure before diving right into a mobile venture. By taking a step back, analyzing the current state environment and carefully planning to become a mobile enabled enterprise, companies can then look to deploy solutions that will enable their key stakeholders to make educated business decisions on the go.
Data Quality on the Go
While mobile analytics can be a very useful tool to empower users on the move, the challenge is taking the massive amounts of corporate data that is available, consolidating it, cleansing it and then organizing it in a way that makes sense for business users and mobile devices. There is more information available to companies now than ever before, and exposing this data to business users is only useful if the data is clean, accurate and helps them do their jobs better.
If users spend time trying to decipher complex reports and make sense of inaccurate data, then IT is actually impeding the organization rather than empowering it. Because underlying data is the key to any successful mobile deployment, companies are advised to start any new mobile initiative by first looking at the data source and the quality of the data that will be used. If source data is not cleansed and organized, there is no point in building out a mobile infrastructure. If the first time a user opens a beautifully mobile-enabled dashboard on his or her tablet and the data is inaccurate, the likelihood of that user ever tapping into corporate data from the tablet again is highly unlikely.
Instead they’ll rely on email and other applications while on the go. While this is also true for analytics deployed to a standard computer, the stakes are even greater with mobile analytics, because they are often accessed when face to face with a customer or when the user has no ability to validate the data. This puts increased pressure on IT.
Taking that fact into consideration, a significant amount of time should be spent upfront cleaning the various data sources that will feed the mobile reports. Efforts should be centered on data cleansing, data quality and data governance to avoid the infamous “garbage in, garbage out” problem.
Data Management on the Move
The concept of ensuring data is clean and free of inaccurate or duplicate data has been around for more than 20 years. What started initially as a mailing issue (validating addresses so customers receive your catalog) has evolved into ensuring both structured and unstructured data are accurate and meeting the demands of the business. At the same time, the consumers of an organization’s data have continued to evolve.
Initially, data was primarily used for internal purposes only. Although a customer might call their sales rep to discuss an order, clients often didn’t have direct access to their own data. As this evolved into the presentation of information via the Web, customers began to have more direct interaction with their data. Although clients had access to data, it was still a restricted view of information (such as name and address only) that had limited effects on how business was conducted. From a sales standpoint, the data was still typically behind a firewall and not readily accessible by the sales rep when meeting at a client’s location.
This has continued to evolve, and as we move to the mobile presentation of data, information is available at all times. With this migration to readily accessible data, two key considerations should be examined: What does this mean from a data governance standpoint? And what are the data quality best practices that should be implemented by an organization?
First, for data governance, there are three considerations with mobile analytics:
- What is our company’s policy regarding mobile access to data? What information should be available and what are the roles and responsibilities associated with this access?
- Do we have data definitions and standards in place? A data taxonomy should be defined and disseminated throughout the organization to ensure everyone understands the data and the resulting analytics.
- What are our mobile technology standards? Which devices and operating systems do we plan to make available to our user base (iPhone, iPad, Android Tablets, Blackberry Playbook, etc)?
In parallel to addressing the mobile governance policies, the need to develop a data quality strategy should be considered. From a performance standpoint, duplicate data often causes the biggest performance degradation as multiple versions of the truth exist and could potentially be presented to clients. The following data quality checklist should be examined within your organization:
- What technologies are we leveraging for data quality?
- Is data quality addressed in a batch or real-time fashion within our organization?
- What is our overall strategy for data quality measurement ? Specifically: What rules do we have in place to determine a duplicate record? If duplicates are found, how should we present this to users in a mobile platform? And how do we ensure we aren’t showing one client’s data to another client?
- Most importantly, do we have our data quality strategy defined to ensure our mobile analytics are correct? If not, our credibility, both with our external and internal customers, is at risk. At the same time, the field sales team, which should use mobile analytics when meeting with clients, may depend on other sources of data to meet their information demands.
Not only should time be spent cleansing and organizing data, but it is also critical to examine the entire BI landscape within the organization. Mobile analytics brings with it a requirement for speedy performance and almost instant response times for retrieving data. While reports can be pre-cached and saved from within mobile devices, users often expect immediate visibility into the state of their business. Long-running queries or reports that take minutes to refresh will severely hurt user adoption, undermining the effort to mobilize analytics. Therefore, if steps can be taken within the database or architecture to increase performance, that analysis should be completed prior to rolling out a new mobile solution.
If it is not possible to increase data and reporting performance in the existing architecture, alternative databases and architectures should be considered. One option would be to explore an in-memory database that provides faster and more predictable performance than a traditional disk-based database. While this can be an expensive endeavor, it is worth making the investment up front to ensure acceptable performance at the time users actually receive the mobile analytics.
It is equally expensive (if not more so) to roll out mobile analytics to a group of users that don’t adopt it simply because it’s too slow. At that point, money has been spent on the mobile analytics roll out, and the business is standing in the exact same place it was before the rollout with less trust in the IT department. At that point, the business may start looking externally for its mobile reporting needs. Taking a holistic view of your information environment to ensure it is well-tuned prior to embarking on a mobile analytics roll out will ultimately lead to success.
Overall, mobile analytics can be an incredibly powerful tool that enables businesses to grow and achieve results previously not possible. By giving business users the data they need, when they need it, and in a readily, understandable format, it has the promise to be a true game-changer across all industries. However, if time and effort is not first spent on cleaning up the underlying data and architecture, it is likely that any mobile analytics solution will have difficulty getting off the ground and showing the return on investment and promise that was put forward at the outset of the initiative.
Register or login for access to this item and much more
All Information Management content is archived after seven days.
Community members receive:
- All recent and archived articles
- Conference offers and updates
- A full menu of enewsletter options
- Web seminars, white papers, ebooks
Already have an account? Log In
Don't have an account? Register for Free Unlimited Access