Tuesday, January 10, 2012

The Other BI: Birst and the Mobile Ad-Hoc Query

This blog post highlights a software company and technology that I view as potentially useful to organizations investing in business intelligence (BI) and analytics in the next few years. Note that, in my opinion, this company and solution are not typically “top of the mind” when we talk about BI today.

The Importance of Mobile Ad-Hoc Query Software Technology to BI

The mobile business market in general, and mobile BI in particular, have taken off over the last year, as end users “roaming around” the warehouse with a smart phone or using a tablet for local central-database-syncing sales calls lead to a “three-legged mobile stool”: smartphone for in-house, smartphone or tablet for local, and laptop for emerging-market instant-on “virtual offices.” And yet, what kind of mobile BI? The main focus of many vendors, including the majors, seems to be BI that analyzes mobile uses, not BI on mobile devices. Those that do aim to do BI on a tablet seem to be emphasizing reporting, not analytics. So where’s the BI that the rest of the industry seems to be focusing on? Where’s the exploratory, in-depth, ad-hoc querying on mobile devices, with a similar look-and-feel to the enterprise’s data-warehousing-type analytics?

Well, the capabilities are there. Something like Google Analytics provides the common infrastructure software for apps running on iPhone, Android, and traditional operating systems, allowing write-once-deploy-many. But Google Analytics is scant on ad-hoc query support, preferring instead to emphasize delivering flexible access to canned reports. Most other vendors seem to do likewise, at the level of their own BI interfaces.

That’s fine for now, as end users slowly get their heads around simple, routine querying. But very soon now, they will want the kind of slide-with-your-hand, pinch-and-expand way of looking at data in more depth that the new mobile interfaces presently provide for searching through a list of YouTube videos for the perfect Facebook post, or zeroing in on the best app for finding a restaurant in the area. For example, there’s birst mobile for the iPad, which can “flick through charts and tables, filter to information of interest and drill into more detail.” And once end users get that, they will indeed apply it to all three legs of the stool (as laptops move to support the new touch screens in their new generation).

The business benefits of that are very much analogous to the business benefits of desktop productivity software like Excel and PowerPoint in the 1990s. They are consumer-market-driven, ubiquitous improvements in employee productivity that, as in that generation of technology, eventually lift the entire global economy to a faster rate of increase. The business benefits of today’s mobile BI are top-down: they are about driving the business’ notion of BI to their shop floors and offices as the employee walks around or takes a trip or sets up a new office on the cheap. The business benefits of mobile ad-hoc querying are about employees piggybacking on the flexibility of those queries to generate new uses for mobile BI and analytics in general, because analytics is being used in places and situations where it never could be used before.

The Relevance of Birst to Mobile Ad-Hoc Query BI

I cannot say, as I have for other Other BI solution vendors, that Birst is a completely safe and well-established vendor. Birst is relatively new, and specializes in cloud BI, a BI market that is much more dynamic and unpredictable. What I can say is that, based on the evidence of its web page, Birst is further along than most major vendors towards delivering common-look-and-feel, fully functional mobile ad-hoc querying in a full BI suite, and that it makes sense that this would be so.

No, it’s not because Birst was able to design for mobile ad-hoc querying from the ground up; that might be true of Birst’s cloud BI, but not of its mobile BI, since Birst is not primarily aimed at mobile BI. Rather, it is because cloud BI, as is the case with cloud markets in general, have a greater emphasis on the SMB customer, and therefore on the naïve corporate end user playing around with the data. It’s relatively simple to extend that kind of user interface to ad-hoc querying on mobile platforms, and the obvious way to do so on the cheap is to provide a similar app with a similar look-and-feel on each mobile platform, based on standardized if not open-source infrastructure software.

I rather doubt that the major competition for cloud-BI-type mobile ad-hoc querying will be the major BI vendors like IBM, Oracle, SAP, and SAS, over the next 2-3 years, although I could be wrong. They seem content to follow their business customers, rather than getting ahead of consumer or SMB customers. No, it may very well be that the main competitor for cloud ad-hoc BI in 2015 is the generation of hundreds and thousands of iPhone, tablet, and Google Analytics-platform apps that together cover much of the immediate use cases of mobile ad-hoc querying, and which slowly integrate together into full-fledged business “pocket querying.” Right now, however, that’s an expensive way to go, because the business has to figure out how integrate and customize an as-yet-unknown set of apps. No, right now the best path forward is a flexible “core” mobile ad-hoc querying and BI suite with which new apps can integrate as needed. Birst, among others, would appear to have such a “core.”

Potential Uses of Mobile Ad-Hoc Querying for IT

As with desktop productivity software, IT should both support (as in, follow the lead of) end users employing mobile ad-hoc querying and also integrate their querying into an overall enterprise BI architecture. Right now, this means ensuring that a “core” platform like Birst is both fully deployed on mobile platforms (whether used or not) and integrated with existing analytics. Thus, the end user may, if he or she wishes, combine the latest Web-searching mobile app with access to corporate data using the same look-and-feel.

Over time, as the market sorts out, it is possible that a platform like Birst’s will be superseded by an agglomeration of apps, as suggested above. No harm, no foul; one can simply fade out the “core” as its integration functions are steadily replicated by the loosely-coupled apps. Or, the major vendors can catch up in their mobile ad-hoc query functionality. Sorry, a replacement is not necessarily justified in this case. Remember, this is being driven by end-user uses, so there is no point in risking the possibility that the vendor hasn’t got it right yet, and especially when the Birst-type platform being replaced depends on standardized, flexible infrastructure software that is well suited to following the ongoing changes in business-user use cases. As in the Bring Your Own Device (BYOD) case, simplicity for the end user trumps simplicity for IT.

And finally, in some cases IT may want to use this technology as a way of getting one’s feet wet in both mobile ad-hoc querying and cloud BI – a two-for-the-price-of-one deal. This does not mean that IT must tackle both topics at once; it simply means that IT can do them at least partially in parallel if it wants to. That would seem especially appropriate for medium-sized organizations that are in a driving hurry to find a kind of analytics technology where they can outflank the big guys, at minimal cost and with minimal effort.

The Bottom Line for IT Buyers

Somewhere in the next 2-3 years, some sort of mobile ad-hoc querying is going to happen, and IT is going to start seeing support calls about it. Sometime after that, it will really take off, and IT will be scrambling to keep up. At that point, IT will be really glad if it has at least a “core” Birst-type mobile ad-hoc-querying/BI platform that will allow it to sort out the confusion.

In the meanwhile, IT buyers should either start the process of creating a pre-short list, or kick the tires of and/or implement a tentative “core” platform. As I have said, Birst would seem to be one candidate for pre-short list or implementation.

It’s hard to tell at this point where some other candidates will come from. Perhaps Google will expand Google Analytics’ capabilities to mobile ad-hoc querying, and they will become a de-facto standard; perhaps SAP/Sybase will move its standout mobile infrastructure software in that direction. Or, someone may acquire Birst and add on its own superb core-platform features. Or Steve Jobs may return and make the swipe-and-pinch interface as dated as hula hoops … naaah.

So, IT buyers, you thought you were getting comfortable with buying for mobile BI? You ain’t seen nothing yet. Still, start familiarizing yourself with solutions like Birst’s, and you should be better set up for the next 2-3 years. If not …

I suppose you could pray for the return of hula hoops. One of these centuries, that should work.

No comments: