Susan Smith has worked as an editor and writer in the technology industry for over 16 years. As an editor she has been responsible for the launch of a number of technology trade publications, both in print and online. Currently, Susan is the Editor of GISCafe and AECCafe, as well as those sites’ … More »
Census Business Builder Tool Combines Census Geography and Data
July 27th, 2018 by Susan Smith
Location intelligence technologies are a ubiquitous part of business, both retail and commercial. A powerful tool at their disposal is the Census Business Builder – Small Business Edition as well as their Regional Analyst Edition. These are innovative, simple-to-use cloud-hosted mapping applications that allow users to easily navigate and use U.S. Census Bureau key demographic and economic data to help guide their research into opening a new, or expanding an existing, business. The tool dynamically combines Census geography and data, such as:
Andrew Hait, Economic Outreach and Marketing Staff, ITMD, U. S. Census Bureau, spoke with GISCafe Voice about the Census Business Builder tool and its ability to empower small business owners with easy access to information on businesses and customers, as well as ways to overlay multiple reference layers over data on maps, such as zoning areas, trade and workforce data. The CBB is updated every 3-6 months with new data and functionality and made available on the cloud.
Are you ramping up for 2020 Census?
We use GIS technologies in two ways at the Census Bureau. We use it internally to conduct our programs like the 2020 U.S. Census and we use it externally when it comes to disseminating our data. I work on the Business Census. A lot of people know about the Census they do every ten years, they know about our annual demographic survey, the American Community Survey. but I actually work on the business programs. We conduct 42 monthly quarterly and annual business surveys. Every ten years we do an Economic Census. So internally we use GIS technology to drive our internal systems, making sure we have every business in the U.S. to conduct a census, making sure we have every housing unit in the U.S. to do a population Census. And using GIS as a way to drive our collection of that data, we use GIS to drive how we tabulate that data, etc.
There’s a whole side of Census use of GIS technology that we see inside our building that the public never really sees. I work on our economic side, but I will tell you that we are making some major changes in how we identify every housing unit to make sure we have a complete count, using GIS technology, including imagery, in ways we’ve never done before.
For example, if you make sure we have a complete address list, we have to compare the list of housing units we had in the previous Census to the list of housing units we have in the current one. Traditionally we do that validation of those addresses annually. We drive every street in the U.S. and identify every single housing unit. That’s a little crazy and expensive. What we’re doing this time is using GIS technology and imagery to take a pass at those addresses to identify those addresses that you can clearly see with imagery. We’re looking at our communities where there are houses where there weren’t houses five years ago.
The imagery shows an area that was open land and now there are 110 housing units in there. We can use the imagery to identify those housing units and to do focused work with those addresses, sending people out to confirm that that one structure really is just one house and not a duplex. We use a lot of technology inside the building.
On the dissemination side, with the public, that’s where I’m more familiar with, we’ve been working with Esri to expand the customer base we have at Census from the more experienced data users who are familiar with our data, they know about our programs and are very statistically savvy. They understand statistics and they have the time to learn about data and products. Esri is exposing our data to people who are different than those types, that are more entry- to mid-level users who previously had no awareness that the Census Bureau even exists. They had no idea of the fact that they could use our data in the running of their business. For example, one thing that we’ve learned is that small businesses, who are deciding where to locate their first business or are working on a successful location and want to expand their business to another location, are doing their research less systematically because they are not aware of resources available to them.
For the last few years, we’ve been working on a data tool called Census Business Builders. I’m actually the project lead on that project, that leverages Esri technology map-based displays of data and presents that data specifically to entrepreneurs and small business owners, so it helps them make that more informed decision. I heard someone a couple of weeks ago say we’re democratizing GIS. We’re exposing GIS technology data in a map-based interface to users who would probably have no other exposure to that type of data. They previously accessed our data in tabular format and tables that don’t tell the story of our data. You don’t see those relationships in a data table as you do on a map. So, consider, I’m considering opening my restaurant in one of five counties in the southern California area. And one of my criteria is I want to locate my business where I can afford to pay my workers. I need to know what the average worker makes.
What are lessons learned, and as a result what have you been doing differently to achieve your goals?
In working with the Census Business Builder project and lessons we’ve learned, first, the expectations of data users today in terms of how information is presented to them is much different than how we’ve traditionally published our data. We’ve learned a lot about presentation systems. The Census Bureau is a large statistical agency. We are an agency composed of a lot of data nerds, but our data users aren’t that type of person. They like something more colorful, more interesting, more visual like maps, so we’re using color palettes on our maps today that are very different than the color palette you more traditionally used.
We’ve learned a lot about our users. They might want information that might span multiple programs and surveys at the Census Bureau where the survey to the data user is not that important. They don’t particularly care whether the data they’re accessing is from the American Community Survey or whether it’s from the Economic Census. In the end they just need information to guide their business decisions, so merging data across those programs has been especially challenging for us.
Census Business Builders is actually one of the first Census tools that is actually doing that, where we are combining demographic and business data together because the merging of those two data sets is where the interesting stories come in. We’ve also learned some lessons in that spirit of combining data across programs. Many users don’t even think of the Census Bureau as a single statistical agency. They want the data across multiple agencies, they don’t want to learn that they have to go to this agency to get this data and another agency to get another data. In working with Esri, we are not only combining Census data together, but we are combining that data with that of other federal agencies such as the U.S. Department of Agriculture. That Department has a large statistical arm called the National Agricultural Statistics Service (NASS). We now have agricultural data in our Census Bureau data tools and we’ve even put some third party data. Esri has their own Consumer Spending data. We actually have that data in with our data because our users have wanted that data merged.
From a technology standpoint we’ve had some real lessons learned there as well. Census Business Builder is the first census cloud deployed application. It’s on the Amazon Web Services. It was a challenge getting the app pushed out to the cloud, but now it’s there we’ve opened the door to other census bureau projects that can be deployed using the scalability, and performance of our cloud deployed applications. Our users don’t want to wait 30 seconds for that map to build, they want that map to build in a tenth or half a second. Cloud was a huge lesson learned.
The way CBB works is all the data comes in dynamically using our own API of the Census Bureau and other application programming interfaces. We’re the first data tool that’s consuming our own data using our own API. This makes it easier for others to consume our data. We cross a lot of bridges and recognize that pulling this data down dynamically into the map using an API takes a lot of burden off the app to store that data. Every year when we update the information, all I have to do is make small metadata changes, like change a 5 to a 6, and it’s much easier (2015 to 2016 data). It’s exposing our data to those who previously wouldn’t have had access to it and exposing technology to us and to them that they may not have the capacity to purchase themselves. Small businesses are not going to be able to spend thousands of dollars for a dataset, or to buy a complicated application. They just need information to help guide their business decisions and then they go off and do their job.
It’s very separate economic data from what we consider every ten years for population Census gathering. We conduct over 130 different surveys at the Census Bureau. We also have monthly quarterly and yearly business surveys. A lot of people don’t realize that Census has all that business data as well as the demographic, socioeconomic and housing data that they already know. Every school child knows we conduct a population Census because Census has a program called Census in Schools where we send out materials to K-12 schools so that the kid comes home and asks his family if they have filled out their Census yet?
We don’t quite have that on the economic side so there are some challenges conducting a business survey from a GIS perspective. Businesses are physically very different than houses. A house can only be so big, so from GIS perspective, geocoding, map spotting that house, publishing and tabulating housing units is very different from collecting and publishing data on businesses that could be hundreds of acres in size. Like a refinery or auto assembly plant or mall, that has individual stores within the mall, but the mall itself is a business. In the end the Census is charged with publishing data on both population and business.