Spatial analysis can be a means of generating deep insights.
Business intelligence is about generating insights from data that lead to decision and action, presumably to the betterment of an organization's fortunes.
Spatial analysis is the use of data that references a specific geographical area or location.
Examples include objects or events in a specific area, roads and property addresses, GPS data to track movement patterns, and terrain such as forest or farmland. And it frequently combines characteristics including geographical coordinates with attributes of the location, such as whether it's a house, commercial building or open space.
All that information can add an extra -- often critical -- layer of context to insights typically gleaned with BI, making them more astute, according to David Stodder, senior research director for business intelligence at data and analytics training and research firm TDWI.
David StodderSenior research director for business intelligence, TDWI
"Spatial analysis is a key to new data insights," Stodder said during a webinar on Jan. 24 hosted by TDWI. "Location is powerful for analytics. Location intelligence -- answering spatial analysis questions -- can lead to new insights and provide new perspectives on things."
In particular, spatial analysis can reveal why something might have happened, correlations between events that might otherwise seem unrelated and trends, he continued.
And it goes beyond simply adding a map to existing BI applications. It can be used in data science models to move beyond merely inclusion in the examination of data to being a means of enabling deep analytics regarding what happened at a location and its potential effects on everything at that location.
"It's the kind of thing that can shed light on those areas by giving organizations a better geographical sense," Stodder said. "Often, the location is what pulls [data] together and makes it make sense."
Applications of spatial analysis
Some primary applications for spatial analysis are discovering opportunities for expansion and reducing risk.
Enterprises generally want to grow, and that often means moving into new geographical locations, whether through the addition of new physical locations or targeting customers in a new area.
While organizations can learn key information about their competitors through all sorts of data, only spatial analysis can inform them where those competitors and their customers are located. And it's with that information that an organization can better determine whether an area might be ripe for expansion or is too much risk given the foothold a competitor might have.
It's also by understanding the geographic location of its own customers -- combined with other data such as the economic makeup of its clientele -- that an organization can extrapolate information about whom it appeals to and then devise targeted marketing efforts.
"It's helpful in a lot of industries to know geospatially what the competition is doing, in terms of locations and marketing and other kinds of efforts," Stodder said.
Supply chain management is another common application for spatial analysis.
Resilience and efficiency are critical to successful supply chains. When supply chains break down at any one stage, it likely affects all subsequent stages of the supply chain and results in delays. And when supply chains are haphazardly put together with vendors spread far and wide, delivery times take longer than necessary, and costs related to fuel and personnel rise.
Location is therefore critical when seeking suppliers and piecing together a supply chain.
"Location is very valuable information," Stodder said.
Other common uses for spatial analysis include local law enforcement and international intelligence in which GPS data is important, and tracking health care trends.
Spatial data is also at the root of climate change and other environmental analyses, enabling organizations to analyze current issues and uncover warnings about future problems to devise plans for how to deal with climate-related events.
"You can look at just about any space or industry and find a use case for geospatial data," said Matthew Forrest, vice president of spatial data science at location intelligence vendor Carto, who also spoke during the webinar. "Climate change and resilience is a huge area, and it all revolves around geospatial data."
As a result of the value spatial analysis can add to traditional BI, Stodder noted, TDWI's research shows that 34% of organizations it surveyed already incorporate geospatial data in their analysis. Another 8% have not yet done so, but have a plan in place to incorporate spatial analysis.
Challenges of spatial analysis
While a means to better insights than BI alone, spatial analysis has challenges.
In particular, speed and scale pose problems.
The amount of data organizations collect worldwide is growing at an exponential rate. According to Statista, the total amount of data created, captured, copied and consumed in 2010 amounted to 2 zettabytes. By 2020, that had grown to 64.2 zettabytes. And by 2025, it's expected to reach more than 180 zettabytes.
Just as the overall amount of data being collected is growing, so too is the amount of geospatial data.
Mobile devices and a rising number of IoT devices are leading to an explosion of new data sources, and organizations somehow need to harness all that data to derive meaningful insights.
"Collecting and managing new and larger data is definitely a challenge," Stodder said.
Traditional on-premises analytics deployments simply don't have enough storage space or compute power to keep up, he continued. As a result, using a cloud data warehouse or data lake is becoming an imperative.
Not only do cloud-based repositories offer unlimited space and more compute power than on-premises systems, but they also make integrating data from disparate sources easier, which lessens some of the significant data preparation burden placed on data engineers.
"Sources can be quite varied, and volume is definitely rising," Stodder said. "So, it challenges organizations that are working with traditional types of data management and data storage systems, and often is a reason they're moving to the cloud where they can take advantage of cloud storage."
Beyond scale and speed, data integration and getting trusted and accurate data are challenges for spatial analysis.
And again, a unified data structure in the cloud can lead to better outcomes, according to Stodder.
Cloud storage repositories make it easier to automate data integration. They also enable organizations to connect data from different departments to break down the isolated silos that can result in data duplication and lack of data consistency.
"It streamlines the ability to analyze the new data and new data sets," Stodder said. "The moving of the data, extraction of the data [and] finding of data is less difficult. Having a unified data architecture and beginning to reduce data silos is important to making spatial data discovery and analysis more complete."
Eric Avidon is a senior news writer for TechTarget Editorial. He covers analytics and data management.