Page 3 of 4
Q&A: BI, Mobile, and Collaboration
Donald Farmer's 20-plus years of experience in data management and analytics include nine-and-a-half years leading Microsoft's BI product teams. In January 2011, he chose a move to QlikTech, creating a role as product advocate for QlikView.
In this interview, he discusses how mobile devices are fundamentally altering the face and adoption of business intelligence -- sometimes in ways that aren't immediately apparent. The ever-popular and insightful Farmer posts sporadic thoughts on business intelligence and loosely related topics at donalddotfarmer.com, and can be followed on Twitter (@donalddotfarmer).
BI This Week: How do you see mobile computing fundamentally changing the face of BI?
Donald Farmer: I see three ways. There's an organizational way mobile will change BI, there's a sort of experience way, and then on the pragmatic side there's a financial way it will change BI.
The organizational way is really important because we've seen this trend of top-down adoption of mobile. I was just chatting with somebody from another BI company about this, ... When we started looking at mobile a few years ago, when people started to think about mobile, we saw the adoption of mobile as being something [that] you'd give to your sales force, you'd give it to your field engineers, and it's going to be driven by Blackberry-based dashboards, KPI alerts, and reminders.
Actually, that's happened, but that has not been the interesting thing. The really fundamental change has been when CEOs, CXOs, and leaders in the organization have adopted these beautifully designed tablet devices and have said, "I want to do everything on my iPad." That has completely changed the landscape.
What started out as being this kind of bottom-up mobile BI has changed, and there's now a very top-down approach, where we're having to tackle the needs of strategic decision makers. We're not only having to tackle their needs but tackle them with a particular kind of user experience that they expect to have on the mobile device.
Organizationally, what that means is that the IT department isn't in control of this process any more. They're not leading the process, let's put it that way. They may be in control of it in the sense that they are managing and administering it, but it's not being led by IT requirements, it's not being led by IT disciplines, and it's not being led by IT formulating strategy. In almost all cases, they're running to catch up. That's very different from the way we've driven business intelligence in the past. I think that's a big organizational change.
Things have been turned on their heads, so to speak.
Yes, exactly. Of course, that in a sense leads to the financial change, which is kind of interesting. Gartner has said this -- that people are increasingly adopting these self-service solutions. They pick out company solutions, whether or not the IT department has approved. ... Very often, these are departmental solutions ... which means that ... the money goes increasingly to enabling and empowering clients rather than building expensive, heavy-duty infrastructures at the back end.
For example, you'll find that people used to put a huge amount of money into the data warehouse and client tools were, in many cases, an afterthought. That is no longer the case. There's a lot more [money invested] very quickly into client tools. It's interesting to me that there has been this substantial change. Again, that's partly because IT no longer owned this process, but it's also a reflection of the importance of the empowered client these days.
Looking at your three categories of how mobile has changed BI, they all seem to tie back to a big shift in both IT and user experiences.
There's absolutely this important change in the experience generally, and that's really the change in the social or organizational side of this. I think there's also this very specific change happening with the user experience. It comes from something Jakob Nielsen described some years ago -- the difference between what he called the "lean-back" and the "lean-forward" application.
What he meant was that there's literally a change in physical attitude between working on an iPad and working on a desktop. On a desktop, you lean forward. You're focused. You're solving very specific problems. With an iPad or mobile device, you tend to be literally holding it at arm's length and you're leaning back. You're observing and taking in information in a somewhat more passive way. Even if you're engaged in it, and you're clicking, there's a slightly different mental approach to it.
It's subtle, but when you capture that user experience, the difference is that on the one hand, an exploratory user experience on a mobile device is more a question of being aware of what's going on in your business through data and exploring things which pop out at you -- things that you notice, things that you recognize, things that you have become aware of through visual clues.
When you're exploring on a desktop device, it's much more about "I have a specific hypotheses that I want to explore." That's the kind of lean-forward one. It's sitting down, focused on a particular idea.
Those two user experiences are very, very important to capture -- both of them. Neither of these is more important than the other. You have to have both these user experiences, which is why we don't just have a mobile edition [at QlikTech]. We still have a very powerful desktop edition.
You touched on this idea when you talked about organizational changes -- the whole bring-your-own-device phenomenon, and how that's changing corporate structures, how people use devices, and so forth. How does BYOD tie into both mobile and BI?
It's critical ... but there's some other things which I think are also really important and aren't just about the mobile. ... How do I put it? It's not just about bring-your-own-device. It's also becoming about bringing your own software.
It's one thing to say, "Everybody buys an iPad…" But what about when you bring your own software? That is a distinctive trend. People are bringing their own software to these discussions and it's great. It's actually a very interesting experience. What people are doing, of course, is acquiring software -- very often through free downloads.
It's a great way of getting business intelligence into an organization at the level where it's actually going to be used and consumed. Previously, the entry point for business intelligence again used to be primarily through the IT department which meant that there wasn't that immediate feedback. We like that immediate gratification of immediately getting results. These desktop tools and these free desktop editions -- in our case, free personal edition -- they give the business user instant gratification that they have to have through rich and enterprise-ready backend to support that business user when they get serious.
There is this level of instant gratification. They try something. They can very quickly get results. We actually find bring-your-own-device is one thing, but bring-your-own-software is a very powerful trend -- one that's favoring us [at QlikTech] very much.
You've made interesting comments on your blog about collaboration and how important that is to BI. You earlier mentioned the social side of computing. Can you talk about your views of collaboration?
One of the things I've tried to get across in talking about collaboration is that collaboration just happens. It's just out there. You can't expect people not to collaborate. We often talk in business intelligence about adding collaboration to business intelligence, but it happens anyway. If people are chatting around the water cooler, they are collaborating.
What you have to think about with collaboration tools is what additional value we can bring. It's not about enabling collaboration. Often in business intelligence, we've assumed that it's somehow our job to enable collaboration, but what we really have to do is provide additional value in collaboration.
So often what we've done in business intelligence in the past has really just been to provide a library where people can post [things] and comment on it -- we call that "collaboration." That's not collaboration. That's terrible collaboration. Human beings are much richer and more engaged socially than that. They have to be able to chat. They have to be able to exchange ideas in real time. They need to have the library for storing things. The way in which they interact is so complex, having one model of collaboration isn't enough.
A lot of the work that we've done in our collaboration architecture has been to provide different models of collaboration. We provide a kind of offline mode of collaboration where you post things to a library, but we also provide live, interactive collaboration. We provide quite rich ways of storytelling the data as well, where you can create a series of bookmarks where those bookmarks and comments are associated with a particular state of the application. So if somebody clicks through your bookmarks, they will actually see the application, its choices and selections changing as you click through them. You can actually start to tell a real narrative story with the data.
What's keeping us from reaching a point where BI information is truly, easily accessible to users? Is it devices, is it software, or are we there and we just don't realize it?
It's just time. We're in the growth phase of mobile BI and it's exceptional. Every single one of our customers asks us about it. It's gone from being something which only a few specialized people asked us about a few years ago, to now being the first question that every executive asks: "Can I get this on my iPad?"
We'll talk about infrastructure, we'll talk about metadata, then we'll talk about all the other things. The first question, though, is, "Can I get it on my iPad?" I think in a sense, we're already there. ... There are no great technical barriers. Of course, there's still more work to be done, but I think the growth in the market is telling its own story.
If you could pick a few words to describe where mobile and BI is headed, what might you say?
It's everywhere. It's as simple as that.
The other thing I would add to that is, we will stop thinking about mobile as being a special mode of BI. Mobile will become the default of business intelligence. Business users will simply consume on mobile devices or on increasingly lightweight desktop devices. The desktop as a BI environment will be increasingly for reserved, heavier, number-crunching, deeper investigation, and more of that focused, sit-forward analysis. When it comes to what most of us do, which is a more lean-back analysis of our current situation, and reflecting on our current status, our business, and trends, most of that will be on a mobile device.