An Introduction to the Storefront Index

Zachary Kitt
4 min readNov 8, 2018

Joe Cortright and Dillon Mahmoudi introduced the ‘Storefront Index’ in a 2016 report available here. The metric is a simple one, counting the total number of businesses within a city that meet several conditions (publicly accessible, densely located, and close to the city center). Despite its simplicity, this metric identifies the vibrancy of major metropolitan centers and indirectly measures other features such as walkability, safety, and economic health — all of which contribute to the quality of life of local citizens.

Although Cortright and Mahmoudi only calculated scores for the nation’s fifty largest metropolitan areas, they provided enough detail for replicating their method for smaller towns and cities. To demonstrate the utility of this metric, I will apply it to Oxnard, CA: a coastal city of over 200,000 people (and this author’s hometown).


Cortright and Mahmoudi acquired their data through a third party, but Oxnard’s data portal was all I needed, although some additional preprocessing steps were required. I had to convert the Business Data dataset’s 11,000 business licenses into a list of storefronts. Doing this required de-duplication and manual curation of some categories, but the end result was a list of Oxnard’s 1,200 storefronts.

The preprocessed data was loaded into QGIS as a delimited text layer. The latitude and longitude columns were assigned their respective point coordinates — in this case, WGS 84 (GPS). These coordinates were then re-projected into a local coordinate system (EPSG:26745) for more accurate distance measurements. (Note that points can be plotted against XYZ tiles as a sanity check.)

Once the points were mapped, QGIS’ distance matrix tool was used to calculate the distances between each location’s nearest neighbor, so that storefronts more than 100 meters away from their nearest neighbor could be removed. Next, the distance to nearest hub tool was used to filter out locations more than three miles away from Oxnard’s city hall (manually added as a separate layer).

Figure 1: filtering out Oxnard’s neighborless and distant storefronts


After performing the steps described in the methodology section, the original count of 1,200 storefronts was reduced down to 963. This final number represents Oxnard’s own storefront index score. Surprisingly, the score of 963 is competitive with the quantities attributed to larger cities (circa 2014) in Cortright and Mahmoudi’s research, including the cities of Austin and Pittsburgh. This number may be deceiving, however, due to the use of different datasets.

Plotting the qualifying storefronts (i.e. those that are densely populated and close to city hall) provides greater insight into city-wide trends, and certainly more insight than would be visible by simply plotting all storefronts (see Figure 1 for a before-and-after comparison). The resulting visual reveals a predominantly north-south, linear orientation. Significantly, only two regions show clustering, the larger of the two being anchored by city hall (the yellow dot) and the lesser being distributed around the Ventura Freeway (Highway 101). These results will make sense to locals, as the first cluster represents Oxnard’s historic downtown and business improvement district, and the second cluster mainly represents a new outdoor shopping center known as The Collection.

Neighborhoods with high walkability and convenience scores will most likely be adjacent to or interlaced within these clusters. Getting to these areas, however, is another question. The city hall cluster sits on the connecting road between both Highway 101 and the Pacific Coast Highway, the arterial roadways of the Central Coast, and The Collection is a freeway exit off the 101 itself. The rest of the city’s storefronts are essentially strip malls, none of which are enough to meet the variety of entertainment and consumption needs of nearby households (see Figure 2). This indicates critical gaps in Oxnard’s neighborhood development fueled in part by the historical presence of major highways.

Figure 2: Strip malls influence the index.


While the Storefront Index is supposed to measure economic strength, it is also supposed to shed light on vibrant communities. In this sense, fast food restaurants and big box retailers don’t contribute as much to a neighborhood’s character as do independent bookstores, pinball arcades, and bars. The Storefront Index would need to acknowledge this difference to better reflect neighborhood quality.

I also propose an additional step that removes the likelihood of sparse or linearly oriented storefronts increasing a city’s score. Starting from the hub point (in Oxnard’s case, City Hall), only a single chain of storefronts linked by distances of less than 100 meters should be included. This would exclude clusters of stores that are marooned on the boundaries and strengthen the index’s relationship to walkability scores.



Zachary Kitt

Writer of code. Interested in data-driven policy. Graduate of @JacksonYale and @UCSBGlobal.