Mobile BI: 6 Key Ingredients for Success

For years, pundits have told us that mobiles would replace many business laptops – and the market is proving them right.

A glance at the news shows that devices are growing in size and capability: The new iPhone 6 Plus with its oversize 5.5-inch screen made a huge splash last week, and next month the 5.7-inch, high-def Samsung Galaxy Note 4 will debut. With these and other large-screen smartphones and tablets in users’ hands, it’s clear that our Business Intelligence and analytics applications need to be mobile.

And it’s not just evolving devices that are driving mobile BI. The workforce that uses BI is growing broader, and fewer of our users work at desks. Between new devices and new users, mobile BI is coming into the mainstream.

But as we know – and as some of you may have learned the hard way – we can’t just squeeze desktop BI onto a phone or tablet screen and hope to be successful.

This approach has at least two problems: First, it puts too much demand on the mobile device and network. And second, it doesn’t leverage the special capabilities that mobility brings to BI – capabilities that can deliver more usefulness and value.

With these facts in mind, I see six factors we must consider to make mobile BI successful. This list is predicated on the idea that an organization already has a BI application and wants to add mobile capabilities to it, but the factors are also relevant if you’re launching a brand-new, mobile-first BI initiative.

1. An app’s purpose. It sounds basic, but the mobile BI journey must start by thinking about what your BI app (or apps) should do.

A Swiss Army-style app – one that does everything for everyone, but does none of it particularly well – usually isn’t the best choice. Better to have task- and job-specific apps that are intuitive and easy to use.

Take time up front to map your standard BI processes and queries to the groups of employees that most frequently use them, and then make those functions easily accessible – thumb-size onscreen buttons work best, if possible. That way, all employees don’t have to scroll through long lists or click through multiple links to initiate BI tasks.

2. Development models. You have several choices here. Mobile BI can be built as a native app, meaning the app is completely device- and platform-specific: one version for Android, another for iOS, plus any other platforms.

It can be deployed as a mobile web app, a web-based app that runs entirely on your servers and is accessed via a mobile browser using HTML5.

Or you can use a hybrid model that combines selected native components with a WebView that displays HTML content. Each has advantages and disadvantages, many of which are spelled out here (and in many other articles online).

3. Connectivity. Even with today’s speediest networks, it’s tough to give mobile users true interactive data exploration.

Downloading large datasets takes too much time, and visualizing data of any great volume can choke even the most powerful mobile devices. And face it: If you have a Bring Your Own Device (BYOD) policy, you can’t count on everyone having the fastest network or the most powerful device, so you need to consider the lowest common denominator. Given this, it makes sense to do as much data processing as possible on the server side.

The good news is that smart developers are already architecting their desktop web-based apps this way. And that groundwork can be reused or reappropriated for mobile platforms.

4. Security.  We mentioned BYOD as part of the connectivity challenge, but those four letters strike genuine fear in the hearts of Chief Security Officers (CSO) everywhere.

BI systems hold some of your company’s most valuable data, so CSOs worry about making that data accessible via hundreds or thousands of pocket-sized, losable, stealable devices. You have to consider how an app can leverage your existing Identity Provider (IdP) and how the app handles data (including transient data). You also need to build remote wipe capabilities into your app and your BYOD policy.

5. Device features. Mobile devices enable your users to do more than just consume information.

With a camera, GPS, and other features, mobiles can also feed data into your BI systems. The camera can scan barcodes (or faces), and the GPS can help users locate nearby resources, both human and physical.

Look at where your BI processes collect data – or where you wish you had more data – and see if you can leverage your employees’ mobile devices to collect it. Automate as much of this data collection as you can, because the shorter the path between data collection and storage, the cleaner the data will be.

6. Feedback. Track who’s using your mobile BI apps and learn from what they do.

You can automate some of this process, but it’s also important to actually talk with users to hear their frustrations and success stories, collect their suggestions, and ask about new features you’re considering. That’s why your chosen development model is important: Your app should evolve and will need regular updating as use grows, your users become more sophisticated, and the mobile devices (and back-end technologies) advance.

If you can make those updates without requiring your users to install an update, you’ve made everybody’s life easier.

Of course, the specific ways you consider these six factors depends on your organization’s unique processes and the BI tools you have in place. And these factors are just a starting point; your specific circumstances and goals will shape the ways your company approaches mobile BI. If someone tells you that a prescribed, pre-packaged solution can solve all your problems, your best response is … skepticism.

How are you using mobile BI? Have these considerations or other factors not listed here made your mobile BI efforts successful? Please share your thoughts in the comments.

About the Author: Fred Sandsmark is a Marketing Content Writer employed by Actuate, The BIRT Company. Before joining Actuate, he wrote success stories and marketing materials for Oracle, Symantec, NetApp, Microsoft, and others as a freelance. Read more of his work at blogs.actuate.com, and follow him on Twitter at @FredSandsmark.