In 2013, mobile "fusion apps" will do things that most enterprise mobile apps don't do now. They will combine key data from more than one source and present that information in a meaningful way to help mobile workers make better business decisions.
Mobile fusion apps, which began emerging in 2012, will help overcome a limitation that plagues most mobile enterprise applications today, predicts Tom Nolle, CEO of technology consultancy CIMI Corp. The apps take data from one source and fuse it with data from additional sources and provide meaningful information that mobile workers can act on. "Today's mobile apps offer limited information, and the information lacks context," Nolle said. For instance, when a mobile CRM app shows nothing more than the most recent date a customer was contacted, that information raises more questions than it answers. What was the outcome of that contact? Did the customer place an order or lodge a complaint? Are there issues still to be resolved?
Those are examples of questions mobile salespeople want answers to, he said, and most mobile apps today fail to provide these answers. That's because up until now, the key way to support mobile workers has been to offer them a subset of data from existing enterprise applications. But the subset strategy has not worked out well, and the coming wave of mobile applications in 2013 will attempt something far more ambitious, Nolle said. "2012 was the first year the enterprise realized that supporting mobile workers meant giving them a different app -- not just a different view of the same application."
In this article, Nolle explains what mobile fusion applications are, and a QA professional Hamilton Gilbert weighs in with his experiences testing these multidimensional mobile apps, which he says are still works in progress. Gilbert works as a project manager at Sefas Innovation.
What are mobile fusion apps?
Mobile fusion apps fuse data from multiple sources. For instance, you could combine information on what a customer bought recently with location-based data that pinpoints where your salespeople are right now, said Nolle, elaborating on his previous example. "Who is available to call on that customer? Joe is, but is he the right person? He contacted the customer three times in last 30 days and no sale has resulted. We could send Joe's boss, but he is on the other side of the country right now. OK, then, Joe's boss's boss is going to make the call."
Trying to do something intelligent
QA professional Hamilton Gilbert refers to fusion apps as multilayered or multidimensional mobile applications. He has seen a few of them in his recent work as a crowd source tester for service provider uTest in Southborough, MA. Multidimensional mobile apps are still works in progress, he said. "When you test them, they often fail. I found that apps [that] strived for that level of complexity were really broken," said Gilbert.
He offered an example of a mobile application he recently worked with. It aimed to provide content -- pulled from multiple sources -- based on the mobile user's preferences, pulled from an additional data source. "The content the app returned was all wrong. It was completely meaningless," he said. "It was trying to do something intelligent, but it fell flat on its face."
Mobile fusion apps aren't there yet
Most of the enterprise mobile apps Gilbert comes into contact with at uTest don't attempt that level of complexity. "They do one thing, such as helping customers pinpoint the nearest location of their favorite pizza chain," he said. Not surprisingly, these apps tend to fare pretty well in testing, he said.
The challenge for simple store locator apps -- and the opportunity for companies that own them -- is to enable customers to place an order once they have pinpointed their store. "Driving home, I click on an app, order a cheese pizza and specify exactly what time I want it delivered," Gilbert said. But right now the best way to ensure pizza for dinner is to call the store and say, "I'll pick it up in 15 minutes," he said.
Technically, it is not difficult for mobile developers to add order capabilities to store locator apps, Gilbert said. One way to do that is to install an applet on the customer's mobile device that stores name, address and credit card number data in an encrypted form. "With security controls in place, the applet could be accessed by many different mobile apps," he said.
He believes one reason that multilayered applications fall short the first time out, is that their developers are holding onto the Web application model of e-commerce, where customers enter data each time they place an order or the application provides a stored version of that data to the outlet that will deliver the product.
It's a different mindset with mobile, Gilbert said. Customers don't want to pull over and type in their information just to get their pizza delivered. "They want to click on the app and say, 'Have it at my house at 6:15,'" he said.
Dig Deeper on Mobile Application Testing Techniques and Tools