Four Recommendations If You Are Exploring Spatial BI or Location Intelligence

This article was written by Mark Flaherty, CMO at InetSoft.

Today’s article is about spatial Business Intelligence or spatial BI. Some people refer to it today as location intelligence. In the world of business intelligence we use information to make smarter decisions. Typically we try to understand business activity dimensionally: who, what, where, when and how.

Now GIS Systems or Geographic Information Systems add one more important dimension, and that’s where. It’s not just a descriptive where, but a where defined by x and y coordinates. Many people say that as many as 80 percent of all transactions are location based. For instance, a phone call happens between people in two locations. Transactions or sales happen in a store. Deposits happen in a bank branch.

So adding location to your business intelligence can significantly aid your analysis of the business activity. Yet, these two environments, business intelligence and GIS Systems, have historically been very separate and used by different people using different systems.

Let’s go though each system and compare and contrast before we talk about how to bring them together. First from a data perspective, business intelligence uses facts and dimensions stored in a relational database. Typically we may want to know sales which is a fact broken down by sales people, by product, by geography, period and channel for the purpose of asking or answering the question why. For instance, why did sales drop last quarter?

What Is Spatial Data?

Spatial data or GIS Systems store spatial data. And what is spatial data? It’s points, lines, and polygons. So, a point might be a store on a map or a ZIP code. It could be a house. A line is perhaps a road, a pipeline, or a utility transmission line. And the polygon is simply a shape. We have some standard shapes of state, country, ZIP codes or we can also create them using GIS Systems. These are custom shapes such as a custom sales territory, a floodplain, the path of a tornado and even shadows of a skyscraper.

Now, GIS Systems store spatial data, as I said, as x, y coordinates, and you can add this as a column in a relational table. Typically these are x and y coordinates that the spatial engine then can apply some complex calculations to in order to support various types of applications. For instance a children’s clothes store may want to know the number of high income families within 50-minute drive time of their store so they can target market to those families.

Also, an insurance company may want to know all the customers in a specific floodplain with the path of an oncoming hurricane so they can assess their risk exposure to these natural disasters. Spatial engines or GIS Systems also overlay all kinds of business data or other kinds of data on top of this spatial data, such as transactional data, interaction data that comes from your core of business operation systems. Also demographic data and psychographic data, such as age, gender, sex, income or beliefs, hobbies and lifestyle are very popular type of enrichment data. Another is environmental data. It shows things like floodplains and natural topographical features.

As I said before these two systems, BI and GIS, have been very separate, and we see below that on the BI side we use BI and ETL tools running against BI and ETL servers running against data warehouse databases to support the type of dimensional analysis I described earlier. These are used by business analysts and business managers and in a well designed BI system, this addresses most users in an organization.

View a 2-minute demonstration of InetSoft's easy, agile, and robust BI software.

Monitoring by Spatial Analysts

GIS Systems on the other hand are geared to a more specific, more narrow audience. They use map-based applications running against a spatial engine and a spatial database that is tuned to support these complex geometric calculations. And there are two types of users. One is those who create these custom maps for use by an organization such as a railroad that wants the layout on a map of all its rail lines. Then it monitors the performance of those lines against specific metrics. Spatial analysts do the monitoring and the analysis to find out what happened right, what happened wrong and to make changes.

While BI and GIS have been two separate systems, now there is an opportunity to bring them together. And there are many benefits to doing so. One, if we can bring these systems together we can finally combine all of our data in one place. Two, we can also spatially enable all of our business data. Today I would suggest a lot of enterprise data is not spatially enabled, probably the vast majority of it.

Third, we can add spatial dimension to our reports if we haven’t already, so that we can have a map sitting side by side with a chart and a table, and if user wants to circle certain features on that map, it will automatically update the chart and the table. And finally, we can deliver spatial data, spatial analysis and location, intelligence to all of our employees, not just a subset of mapping analysts.

But there are challenges along the way. One, not all databases support spatial data and spatial functions. Two, not all companies have GIS or have geocoded even their customer data. And also spatial calculations require lot of horsepower, and some databases may not be up to the task. And fourth, a lot of BI tools offer varying support for GIS location intelligence capabilities.

Read what InetSoft customers and partners have said about their selection of Style Scope for their solution for dashboard reporting.

Identify Potential Uses for Spatial Data

So for BI professionals, I have four recommendations if you are exploring spatial BI or location intelligence. One, identify potential uses for spatial data in your organization. Some of the examples I gave may trigger some ideas for your organization. Two, explore the spatial capabilities of your BI and database tools that you are using in your BI environment to see if they are up to the task for supporting spatial data.

Three, work towards combining your spatial and your business data in one repository. Ideally I think that make sense to do it in your data warehouse. And four, make use of syndicated data to expand and enhance your enrichment data that can be used to overlay on your maps to provide more value for your users.