The visualization of the distribution and growth trends of art galleries in New York in recent years

Link here


For this report, I have chosen to study the growth trends and distribution of art galleries in New York in recent years. New York has long been known as “the capital of art”, with the highest value of art sales and the highest market share of art worldwide. The strong financial base and highly developed cultural infrastructure and regulations in the U.S. have contributed to New York’s leading position in the world’s art field.

However, this position has been under attack from various sources recently. In terms of competitors, China and other emerging markets have made rapid gains in terms of art share, and their rise is attempting to carve up New York’s existing art market share. In terms of passive context, the emergence of Covid-19 has forced the temporary closure of most galleries, museums, art events, and the number of travelers has dropped dramatically. As art businesses, artists and institutions struggled to cope with rising operating costs, New York became one of the hardest-hit American cities during the epidemic.
When it comes to art-related institutions, we often overlook the “gallery” category. Unlike museums, which are financially supported by the government, galleries are small, self-run businesses that need to take a more flexible approach to deal with external shocks. But it’s hard to believe that before the epidemic, there were 38,000 galleries in the United States, and New York galleries accounted for 16 percent of them.

And while most existing studies and surveys have focused on art institutions such as museums, there are few studies on galleries.
Therefore, this report will present the growth trends and distribution of New York galleries in recent years through different visual charts and will use tools such as Openrefine and Tableau Public to show the distribution of galleries in New York City and related factors, in order to find more ways to help promote New York galleries in the future.


When looking for a database to analyze, my first choice was Google dataset search, where I typed in keywords like “Gallery+New York” and got more. To get more data, I also searched on Statista and found data analyzing the added value of the New York art industry to the U.S. GDP over the last 20 years, as well as the growth trends of the New York art industry in recent years according to the U.S. Treasury Department. And to demonstrate the need for the existence of New York galleries, I searched Statista to get data on the visitors of a New York gallery in recent years. Finally, in order to visualize the distribution of New York galleries, I also collected the names, zip codes, and websites of the galleries.

When looking for the inspiration for the visualization type, I referred to two examples.
The first is a PDF of the New York market report, which visualizes the share of New York City with other U.S. cities in a pie chart and gives each city a different pattern. what attracted me most was the author’s unique visualization, which fills in only the New York portion in red and zooms out from the pie chart to focus the viewer’s attention on New York’s proportions. I will learn this in the report as a reference for the first half of the visualization.(Figure1,2,3)

The second is also a map of the location distribution of galleries in New York, an open map of New York City uploaded in 2014. The visualization marks the location of galleries on Google map with red dots, and the viewer can scroll and zoom in to get the exact distribution of galleries, as well as access the data used to generate the map through the NYC Open Data Platform. Despite the use of interactive technology, the visual component of the visualization is less than ideal. There are too many colors for the user to distinguish the focus point when using it, and the red dots are marked in a way that does not relate to the intended content of the gallery. Therefore, I will take the form of a map in this report, but improve the visualization. (Figure4)


Process and Results

After searching for relevant data on different source sites, I imported these .csv files into Excel to organize them and then imported them into openrefine for further processing once they were organized. Since I was not visualizing a single database, but multiple databases, I needed to link them together. The common points I found were “Art” and “Newyork”, which I introduced in a cascading fashion to the distribution of New York galleries I would eventually visualize. I divided my visualization project into two sections, growth trends and density distribution.

When I first organized it, I removed unnecessary headers, leading rows, and other information I didn’t need in the dataset and added the missing units of added value and headcount from the original .csv file to get the following data. (Table1 is a screenshot of the initial data, Table2 and Table3 are screenshots of the data after basic processing)

Table1(Original version)

I used this exported .csv file to load the data into Tableau Public and began creating visualizations.
The first part of the visualization is a trend visualization, divided into three databases, which are the trend of the number of art industries in New York and other U.S. cities in recent years, the trend of the contribution of the New York art industry to the national GDP in recent years, and the statistical trend of the number of people visiting New York galleries each year.

The first question that came to my mind when working with the first database was how to highlight the share of New York and the trend of growth. So, I grouped all other cities, including California, Texas, and Washington, D.C., under “Other”, leaving only “New York” and “Other” as the two city variables. (Figure 5)


I then present the changes in city shares in a bar chart, in chronological order. And, to highlight the change in the share of “New York”, I show the value of New York in blue (because blue is the representative color of New York City), while the other values are mostly in grey and white.
In the chart, we can see that the total value of New York’s art industry as a share of the U.S. has been steadily increasing between 2010 and 2016. Despite the lack of data for the last five years, we can still speculate from the growth trend that this share will continue to grow or stabilize in recent years, and either way, it will still represent a large part of the total value of the United States.(Figure 6)

Figure 6

NY’s share of Arts Industry in U.S. from 2010-2016

The second and third databases represent the added value of the New York art industry to the U.S. GDP, as well as statistics on the number of visitors to New York galleries each year.

Again, when working with the database of the added value of the New York art industry to the U.S. GDP, I continue to use blue as a proxy for the “value of New York” and use an area chart to better represent the growth in GDP. As we can see from the graph(Figure7 below), the contribution of New York’s arts industry to U.S. GDP has been steadily increasing from 2001 to 2019, with the value becoming 2.4 times its initial value over the twenty-year period.

Figure 7

NY annual added value to U.S. GDP in last 20 years

For the annual count of visitors to the New York gallery, I also chose to use blue as the main color and labeled the chart with some information that I thought was important. But this time I chose to use a line graph because it better reflects the diverse trends in the number of visitors. I chose the average value of 20 million as a reference value, with the data above in orange and the data below in blue, as a distinction between whether the number of visitors can sustain the basic operating costs.

Figure 8

As we can see from the chart above(Figure 8), the number of visitors to New York galleries rose rapidly from 1970 to 2000, fluctuated from 2000 to 2008 (with a break in 2004 possibly due to the economic crisis in the US), and remained stable at around 30millions from 2009-2019. The year 2020 was the second breakdown, this time due to the impact of the New Crown epidemic, and 2021 saw a return to growth. The above information indicates that New York galleries have been on a growth trend for a long time in the past, and are likely to continue to do so in the future until they return to their pre-epidemic levels.

Annual visitor numbers to NY museum and galleries in last 50 years

I am satisfied with the default Tableau font as it is a nice sans-serif font that is easily legible, but I did increase the size of the headings slightly and reduced the font size of data that I considered less important in the visualization.
All of the information in the database above reveals a message that the art industry in New York is growing and has a promising future. I also chose to use a variety of chart formats in the visualization, such as area charts, line graphs, etc., and to strongly highlight the main body of information in terms of color and font size. One, because I wanted to experiment more with different possibilities for expressing trends with the new tool, and two, I wanted to reflect diversity in my final visualization report.

After completing the visualization of the trends, I then decided to visualize the current distribution of New York galleries by geographic location on a map. In order to get more comprehensive information, I merged the two databases together, making the data variables include gallery name, contact information, official website URL, longitude and dimension, region, and specific geographic location. (Table 4) I wanted to create an interactive map-like format that would allow users to find galleries in their area by zip code and to jump to the gallery’s official website when clicked. I also wanted to express the number and density of galleries in the area by the shade of color.

Table 4

Using Figure 9 as an example, I imported Zipcode into Tableau Public and selected the region as the United States, and the system generated a map of the gallery represented by dots. I originally planned to express the density in terms of color shades, but in the form of dots, I thought the shape size would be a better visual representation of the density. I also tried to change the Background layer to compare the distribution of New York galleries with the distribution of the local population and found that the density distribution of galleries did not correlate with the number of population. Therefore, I changed the layer back to the preset layer.

Figure 9

However, we can still see from the above graph in New York City(Figure 10) for example: the closer the area to the city center (Manhattan city), the larger the shape of the blue dots and the higher the density of galleries. Conversely, the further away from the city, the smaller the blue dots and the lower the density of galleries. This is what I would expect, as larger cities tend to have higher standards of living and more access to art education, so it stands to reason that the number of art institutions would be higher.

The graph below shows the distribution of galleries in Long Island City( Figure 11) and Brooklyn (Figure 12) respectively, and we can see that the number and density are much smaller than in New York City. Therefore, it shows that the promotion of art education and institutions in these areas still needs more effort.

In the above gallery distribution map, users can click on a dot to get the name and official website of the gallery for further reference. And users can also filter the zip code they want to check in the zip code area on the right side, and then the area will be displayed on the map.

Density distribution map of New York State galleries


Google dataset search, Statista – finding the dataset
Excel – pre-processing the database
Openrefine – editing the database
Tableau Public – data visualization

Reflection and Future thoughts

As I am still familiarizing myself with the new tool, this lab report did not reach the initial interactive goal, which will be improved and worked on in future projects.
In the meantime, I will continue to explore the relationship between the density distribution of the gallery and other factors, such as the traffic coverage of the area and the financial expenditure index. I hope that more research on the relationship between different factors will help find ways to be able to promote New York galleries more in the future.