ITBusiness.ca

Big data and self-service – A Q&A with Information Builders’ Jake Freivald

Prescriptive Analytics

With business intelligence driving more and more decisions within companies both large and small, New York-based Information Builders has been working to give its customers more “self-service” options.

That mantra was repeated during the company’s User Summit in Orlando, Fl., this week, with some of its products geared towards giving non-data scientists more insight into their figures and numbers.

Jake Freivald, vice-president of corporate marketing, has been with Information Builders for 15 years. On Tuesday, he talked to us about the company’s direction, its strategy, and how it plans to help its customers understand their data.

This interview has been edited and condensed for length and clarity.

 

ITBusiness.ca: During this conference, we’ve been hearing a lot about “self-service” for business intelligence. Can you speak to that?

Freivald: One of the things that’s really interesting about this company is that we’ve always focused on business users – people who would not normally be expected to understand technology and use tough technology.

So in the business intelligence arena, and business analytics arena, you get a lot of people focusing on analysts. We think about analysts as the ones who are the most critical to us, they’ve got to have the right tools, they’re going to be the five people in the back room who make some kind of magic inferences from the data that will save the company a million dollars or whatever the case may be.

And yeah, those people are extremely important … But there are a whole bunch of other people out there who are making decisions everyday. And one of the taglines for our company is that everyone makes decisions. And everyone from the executives to the analysts of course, to the people who are running a supply chain, to a nurse or a doctor or people working in a hospital – whoever they may be, all those people are making decisions.

And so what’s self-service to those people? That’s part of the question. They don’t want tools. They don’t need predictive analytics, per se. What they need is some kind of application – what we call an info app … You should be able to have that information accessible from your phone, from your browser, or whatever you’re on, and you should be able to manipulate that information.

We demonstrated some of that today. We talked about self-service analytics using InfoDiscovery, which was the big announcement today. It’s very visual, it’s very graphical, very interactive, and so on … What you’re left with is a single set of graphs and dashboards that somebody can use. They can understand, OK, if I lasso a piece of a map, the data will change, or I will see the piece of data that is relevant on the map.

 

ITBusiness.ca: Today [Information Builders president and CEO] Gerry Cohen mentioned how important it is for people to use Excel … But even that requires a certain level of familiarity with data. So when you say [InfoDiscovery] is good for any business user or power user, how savvy do they have to be?

Freivald: It’s a good question. Let’s talk about the difference between authoring something and using something.

The user of InfoDiscovery, something that’s been published, takes not as much skill. They have to understand what a visualization is, they have to understand that different colours mean different things, different sizes of dots mean different things. They have to know when they lasso something, something’s going to change. There’s a little bit of familiarity that you have to get there, but you’re not going to hurt yourself by playing with those, because you’re not fundamentally changing the data, you’re not bringing in new data sources. All that stuff has been done for you already.

The author is going to be a power user or an analyst – someone who understands data generally, the system they’re getting it from, the quality of the data, they understand that if they’re going to join data from a spreadsheet to data from their [customer relationship management] application, those two things together – they have to correlate in the right ways. That’s going to take someone who has at least a solid business understanding of the data. It doesn’t mean they have to be a database administrator or have a deep knowledge of SQL Server for Microsoft or any of that stuff, but they have to understand the way that the data goes together.

ITBusiness.ca: Everyone’s talking about how to make use of big data and unstructured data, and how to get some use out of it. [Keynote speaker] Thornton May mentioned if you know how to monetize big data, you are the hero. So what is the strategy for Information Builders going forward to make that big data understanding or monetization a reality?

Freivald: Interesting question. Part of what’s interesting about it is that it’s not just a big data question. Big data got a lot of people excited, but monetization of data that you already have is hugely important regardless.

Ford, for years, decades I think, had warranty claims and collected reports about them in a backroom – reports that no one was looking at. They took that information and pushed it back to dealerships in North America, and put it on a dashboard that anybody could understand … These guys would look at the dashboard, and they’d be able to say to a customer, you want to work with us and you want to get your warranty repair work done here, because we have the lowest return rate of any in the area.

The dealer gets the advantage, but Ford also saves money because now, the work is done better, and people in the other dealerships were looking at this and saying, I don’t look so good. And because I don’t look so good, I’m also losing warranty claims business with Ford, so I actually want to increase my quality … That saves Ford money, to the tune of $25 million a year.

You can change culture with data, like Ford did, creating a culture of quality … If you’re talking about monetizing data to create something new, that’s a huge resource we haven’t yet tapped.

Exit mobile version