In recent years, organizations have realized significant benefits from investing in geospatial technology and data science capabilities. GIS and data science naturally complement one another, each supporting the work of the other. Often, however, an organization’s data science team resides within one department, and the GIS team lives elsewhere in the org chart.
At first glance, this division could appear advantageous. It may seem easier to manage separate teams responsible for their own business unit-focused analytics. However, if these two disciplines are disconnected in an organization without shared data or unified governance, it can result in higher costs; inaccuracies; reduced operational efficiency; and missed opportunities for deeper, more sophisticated insights.
The Price of Fragmentation
A GIS or data science team can often be created in response to a need within the organization. Say the CFO of a company wants a financial dashboard to track compliance with new regulations. The CFO’s division forms a data science team that develops a data warehouse and dashboards to answer specific financial questions.
At the same time, the manager of the company’s marketing department wants a map showing customer demographics in different sales regions. The manager hires a geospatial technology team that purchases GIS software, geocodes customer data into a geodatabase, and generates maps. The manager can now visualize market areas, gain new insights into customer purchase patterns, and improve targeted advertising.
In this scenario, the CFO doesn’t need to care much about customer demographics and the marketing manager has little concern for the company’s financial reports. Neither has a problem with siloed analytics capabilities across the company since their own parochial needs are being met. In fact, they may not even be aware of what the other’s team is doing.
Now, say other executives at this organization recognize the value provided by the maps and dashboards and start building their own stovepiped analytics or GIS teams within their business unit silos. As analytics capabilities proliferate in an uncoordinated way across the company, efforts are duplicated; enterprise architecture becomes increasingly complex; and costs for data storage, infrastructure, and software licensing multiply.
Even worse, the lack of integration limits the company’s ability to gain enterprise insights, making it difficult for the CEO to answer simple questions such as “How many customers do we have?”
Attempting to answer that question, the organization’s financial data team might decide to define customers as clients who have submitted payments within the last 12 months. The team queries its financial dashboard and tells the CEO there are 1,251 customers.
Simultaneously, the marketing executive’s GIS team defines customers differently. They tally all accounts with valid mailing addresses and tell the CEO there are 1,486 customers.
It’s easy to see the problem here. Now consider a scenario in which what the CEO actually wants to know is how many customer interactions happened within the last two years, regardless of purchase history. The business rule that defines customer is not documented in a centralized metadata repository, and not communicated with either team. People become frustrated; time is wasted; and after so much investment in GIS and dashboard technology, nobody can answer a question that is, on its face, relatively simple.
If the organization had consolidated its GIS and data science teams into a one-stop data shop with an integrated data model and standardized data definitions, multiple problems would likely be solved. For one thing, those standardized data definitions and clear, structured lines of communication reduce the risk of miscommunication between the CEO and the rest of the organization. Additionally, the CEO would get their answer more quickly and with minimal cost and annoyance.
Maybe you’ve heard this one before: Imagine you ask four blindfolded people to each touch a different part of an elephant. One person, feeling the tusk, might think they have a pointy spear. Another person, touching the tail, would call it a rope. A third might grab a leg and say it’s a tree. A fourth, feeling the trunk, might conclude they found a snake. Nobody gets the full picture and realizes it is an elephant. This is what happens when the GIS and data science teams are separate and uncoordinated.
The Benefits of Integration
When strategically and holistically evaluating an organization, it often makes sense to integrate GIS and data analytics teams. There are many benefits to using this approach. With it, you can:
- Gain efficiency through streamlined processes.
- Reduce costs by consolidating databases, infrastructure, and software licenses.
- Unlock more sophisticated and aligned visualizations of complex spatial information that drive deeper insights and empower more effective decision-making.
- Achieve more effective problem-solving through collaboration.
- Improve quality and consistency of data products.
- Enable AI, predictive analytics, and other advanced data capabilities.
- Create a one-stop shop by combining maps, dashboards, and data onto a single ArcGIS Hub landing page.
- Scale enterprise investments to reveal holistic insights and optimize business operations.
Ideally, companies should consolidate—or at least coordinate—data science and GIS teams so that they work together and stay in sync. This can involve defining and documenting the calculation rules and metadata for key metrics; establishing a centralized workflow for triaging and managing ad hoc data requests; and working to source all GIS map applications, dashboards, and ad hoc analyses from one authoritative data source. Doing so will create efficiency; save money; and—most importantly—ensure that everyone gets the same answer no matter what the question is, who asks it, or which tool is used.
There are many ways to achieve this kind of integration. Multiple ArcGIS tools, for example, already support this convergence. With ArcGIS Notebooks and integration with Python, organizations can unify spatial and nonspatial data science workflows in a single environment. ArcGIS Hub is a cloud platform that can become your company’s landing page and centralized data catalog for easy sharing of data, geospatial apps, dashboards, and other data products. Advanced data-sharing capabilities that support open data standards such as REST are built into ArcGIS tools. You can also easily connect to internal company data, ArcGIS Living Atlas of the World, or other data sources and drag data directly onto your map.
As organizations increasingly adopt data science and AI to drive decision-making, the ability to spatially enable these models becomes a competitive advantage. Move beyond answering stovepiped data requests, toward a future that includes real-time operational intelligence across your enterprise GIS.
Taking advantage of these tools—and, crucially, the mindset behind them—positions you to be a leader in integrating geospatial and data science systems as analytics and operational intelligence become increasingly vital in a rapidly changing world.
About the author
