Part 1 of 3 – Being Data-driven
Morgen Healy, Principal, and Kate Hickey, Vice President
This is the first in a series of three articles on the differences between “data-driven”, “data-informed”, and “data-inspired” approaches to working with data, and when to choose among these approaches.
Like many of you, we think a lot about data. It’s the core ingredient to geospatial analysis, allowing us to reap benefits — ranging from improved traffic flow management, to more effective disaster response, to better understanding of large scale land use changes. Data is also essential to helping our organizations as we strive to improve processes, make good decisions, model potential outcomes, and assess our performance. But, we should also recognize that the models for how we collect, maintain, and analyze data are being upended, and the pace at which we must analyze vast amounts of data is unprecedented. We are told we need to be “data-driven” to be effective, but what does that actually mean and how do we actually do it? We need a flexible, useful framework to ensure we pay attention to what is most meaningful, and get the answers that will help drive all of our organizations forward.
Is being Data-driven always best?
We know we want to be leveraging data to improve our work, but how do we approach this? We recently conducted an informal poll at the annual Texas GIS Forum asking attendees if they’d rather: (a) Ride in a self-driving vehicle or (b) Drive a high-performance vehicle. The votes (as shown in the glass jars) were remarkably even, and the question posed led to some interesting conversations. What we heard, and probably all know intuitively, is that it depends on the situation and what you’re trying to do. You might want to be in a self-driving car during your traffic-laden morning commute so you could read your email or scroll Instagram. On the other hand you might choose to drive the high-performance vehicle if you were cruising the Pacific Coast Highway on a perfect blue-sky day. One is not better than the other. It depends on the situation and what you’re trying to achieve. Being data driven (literally like the self-driving car) is not always the best approach. Instead, we should think of these tools as existing on a continuum.
This concept of having different data strategies for different situations is described well in a March 2019 blog by Shayna Stewart at Y Media Labs. In her blog, Shayna describes three approaches to leveraging data — each with different characteristics and goals: being Data-driven, being Data-informed, and being Data-inspired. Each is equally useful and important. The challenge is applying the correct approach to the problem or question you are trying to tackle.
When should you be Data-driven?
A Data-driven process is spurred on by very specific information, as opposed to mere intuition or personal experience. A data-driven decision is made with hard empirical evidence, not speculation or gut-feel. And perhaps most importantly, when you are being data-driven, you have the exact data you need to make a decision. The data gives you your answer and next steps, plain and simple.
Public safety organizations often need to be data-driven. For example, we recently completed a GIS strategic plan for the California Department of Forestry & Fire Protection, or CAL FIRE. They are a large, complex organization with a $2.3B annual budget, and the task of operating over 700 fire stations across the state. Between 2013-17, CAL FIRE responded to 5,700 wildland fires and over 460,000 incidents. They have a lot of moving parts, to say the least!
CAL FIRE understands how data can provide insight into complex, fast-moving situations. The firefighter in the field has only seconds to make a decision, and must be able to answer a yes/no question immediately. There is no time for interpretation or weighing of options. One of the greatest technology concerns expressed during the strategic planning process was the risk of over-saturating a firefighter with data and information. This is the time to be focused and data-driven. Is the fire moving in this direction? Yes. Is there a person in this structure? No. This is not a time for exploring maps or mulling over options. It is a time for knowing definitively when to stay and when it’s OK to move on.
Emergency response to a 911 call is another clear case of a process that needs to be data-driven. A call comes in, and the fire truck and ambulance need to get the correct location as quickly as possible. Minutes and seconds can be the difference between life and death. Next Generation 911 (NG9-1-1) is attempting to improve our nation’s emergency response systems, top to bottom. At its core, as opposed to the original 911 system and enhanced 911 (E-911), NG9-1-1 leverages GIS data and geospatial capabilities to manage incoming calls, locate those in need of help, and dispatch first responders. NG9-1-1 relies on certain GIS datasets, such as jurisdiction boundary files, street centerlines, and most importantly, the building-level address data. Dispatching first responders is a purely data-driven process (as it should be), as there is no room for human error or interpretation. As described in a previous AppGeo blog, NG9-1-1 will help ensure this process is leveraging modern technology and the best-available GIS data.
Being Data Driven means that you have the exact information you need to answer a question or make a decision. Your next steps are clear, and there is very little room for interpretation.
Next Up – What it means to be Data-informed and Data-inspired. Click here to view the next blog in this series, in which we shift along the continuum and explore what it means to be “Data-informed” and “Data-inspired” and how to use these different approaches within your organization.