1) Summary

This document walks through how to build the Promise Zone National Housing Preservation Database dashboard in ArcGIS. It does two things: 1) explains how the dashboard was made, and 2) actually provides the code necessary for preparing the data and the graphics. These elements will still have to be uploaded to ArcGIS to actually create the dashboard, but this markdown document should make clear how to create and use them.

It lays out five steps:

Importing and cleaning National Housing Preservation Database data
Importing and cleaning American Community Survey data
Making graphics for the dashboard
Automating text with embedded statistics for the dashboard
Uploading and assembling all these components in ArcGIS

Note: this walkthrough is not perfect. There are almost definitely some steps that I forgot to include, and you will need to fiddle with some things to get them to look exactly how you want. There are simply some limits to what it’s possible to build in ArcGIS. My hope is that this walkthrough will make it much easier for you to manually update the dashboard, though, and will serve as a foundation on which you can improve.

#If you are at a City office or on your VPN, you will need to load the ‘curl’ #package and run these three lines of code to get around the proxy server and #be able to use the census API
#function to get around proxy server for tidycensus
#library(curl)

#companyproxy <- curl::ie_proxy_info("proxy.phila.gov:8080")$Proxy
#Sys.setenv(http_proxy=companyproxy)
#Sys.setenv(https_proxy=companyproxy)

#install.packages("devtools")
#devtools::install_github("CityOfPhiladelphia/rphl")

#install.packages("janitor")

library(tidyverse)
library(acs) #to import american community survey data
library(tidycensus) #to import american community survey data
library(readxl) #to read the excel spreadsheet of nhpd data
library(sf) #for spatial data
library(rphl)
library(forcats)
library(janitor) #to clean dirty excel files

require("knitr")
opts_knit$set(root.dir = "C:/Users/Nissim.Lebovits/OneDrive - City of Philadelphia/Desktop/Transition Documents/Data/R Scripts and Datasets/Housing")

2) NHPD Data: Import & Wrangle

NHPD data must be manually downloaded as an Excel file here. You will need to register for an account, but this is free for government agencies. Select the dataset titled “Inconclusive and Active Properties PA”, then:

Download NHPD data
Import Promise Zone boundaries
Spatial filter
Clean categorical data
Add extra columns
Export to Esri shapefile

#import nhpd data for pennsylvania
all_nhpd = read_excel("./Inconclusive and Active Properties PA.xlsx") |>
              clean_names()

#filter for in philadelphia; convert to sf
phl_nhpd_sf = all_nhpd |>
                filter(city == "Philadelphia") |>
                st_as_sf(coords = c("longitude", "latitude"), 
                         crs = "EPSG:4326")

#fix typo in TargetTenantType column
phl_nhpd_sf$target_tenant_type[phl_nhpd_sf$target_tenant_type == "Eldery or Disabled"] = "Elderly or disabled"

#create column for earliestendyear
phl_nhpd_sf$earliest_end_year = lubridate::year(phl_nhpd_sf$earliest_end_date)

#import the pz; use to filter phl data
pz = read_sf("C:/Users/Nissim.Lebovits/OneDrive - City of Philadelphia/Desktop/Transition Documents/Data/R Scripts and Datasets/General Boundaries/Shapefiles/PZ_Shapefile",
             "PZ_Boundaries", stringsAsFactors = FALSE) |>
            st_transform(crs = st_crs("EPSG:4326"))

#Create a new column for a categorical variable indicating whether or not each site is in the Promise Zone. 

pz_nhpd_sf = phl_nhpd_sf[pz, ]

pz_nhpd_sf$in_pz = "true"

non_pz_nhpd_sf = phl_nhpd_sf[pz, , op = st_disjoint]

non_pz_nhpd_sf$in_pz = "false"

phl_nhpd_sf_full = rbind(pz_nhpd_sf, non_pz_nhpd_sf)


#select the columns that you want
phl_nhpd_full_forwrite = phl_nhpd_sf_full |>
                            dplyr::select(property_name,
                                          property_address,
                                          total_units,
                                          earliest_end_year,
                                          owner,
                                          owner_type,
                                          manager_name,
                                          manager_type,
                                          target_tenant_type)

#esri shapefiles do not accept null values (NAs),
#so we'll split the data into an expiring layer and a non-expiring layer
#and then upload them to the same map
phl_nhpd_expiring = phl_nhpd_full_forwrite |>
                      filter(!is.na(earliest_end_year))

st_write(phl_nhpd_expiring, "./phl_nhpd_expiring.shp")

phl_nhpd_non_expiring = phl_nhpd_full_forwrite |>
                            filter(is.na(earliest_end_year))

st_write(phl_nhpd_non_expiring, "./phl_nhpd_non_expiring.shp")

3) ACS Data: Import & Wrangle

In this step, we’re importing American Community Survey data to compare the Promise Zone to Philadelphia as a whole. We use area-weighted spatial interpolation to estimate numbers for the Promise Zone, and then perform some final simple calculations.

#import data by tract for interpolation
phl_tract_acs_data = get_acs(geography = "tract", 
                   year = 2020,
                   variables = c(
                                  "B11001_001E", #Total number of households
                                  "B25070_007E", #Rent 30.0 to 34.9 percent
                                  "B25070_008E", #Rent 35.0 to 39.9 percent
                                  "B25070_009E", #Rent 40.0 to 49.9 percent
                                  "B25070_010E"), #Rent 50.0 percent or more
                   geometry = T, 
                   state = "PA", 
                   county = "Philadelphia", 
                   output = "wide") |>
                   rename(tot_hh = B11001_001E) |>
                   mutate(tot_rentburd_hh = B25070_007E +
                                            B25070_008E +
                                            B25070_009E +
                                            B25070_010E)|>
            st_transform(crs = st_crs("EPSG:4326"))

#import full city data
phl_county_acs_data = get_acs(geography = "county", 
                   year = 2020,
                   variables = c(
                                  "B11001_001E", #Total number of households
                                  "B25070_007E", #Rent 30.0 to 34.9 percent
                                  "B25070_008E", #Rent 35.0 to 39.9 percent
                                  "B25070_009E", #Rent 40.0 to 49.9 percent
                                  "B25070_010E"), #Rent 50.0 percent or more
                   geometry = T, 
                   state = "PA", 
                   county = "Philadelphia", 
                   output = "wide") |>
                   rename(tot_hh = B11001_001E) |>
                   mutate(tot_rentburd_hh = B25070_007E +
                                            B25070_008E +
                                            B25070_009E +
                                            B25070_010E)|>
            st_transform(crs = st_crs("EPSG:4326"))


#interpolate pz data
pz_acs_data = st_interpolate_aw(phl_tract_acs_data[3:14], pz, ext = T)

pz_acs_data = pz_acs_data |>
                dplyr::select(tot_hh, tot_rentburd_hh)

pz_acs_data$geography = "Promise Zone"

phl_county_acs_data = phl_county_acs_data |>
                dplyr::select(tot_hh, tot_rentburd_hh)

phl_county_acs_data$geography = "Philadelphia"

acs_data_for_charts = rbind(pz_acs_data, phl_county_acs_data)

acs_data_for_charts = acs_data_for_charts |>
                        mutate(pct_rentburd = tot_rentburd_hh/tot_hh*100)

4) Make Graphics

The code below will make the appropriate charts to embed in the ArcGIS storymap. All you have to do is right-click them, “save as”, and make sure to add “.png” to the file name that you give them. You can then upload the resulting .png files to ArcGIS and embed them in the NHPD storymap.

  phl_nhpd_sf_full |>
  group_by(in_pz) |>
  summarise(total_units = sum(total_units)) |>
  ggplot(aes(x = "", y = total_units, fill = in_pz)) +
  geom_col(color = NA, alpha = 0.5) +
  geom_text(aes(label = total_units),
             position = position_stack(vjust = 0.5),
             alpha = 1) + 
  coord_polar(theta = "y") +
  scale_fill_manual(values = c("#5b6983", "#7fb0ac"), 
                    labels = c("Outside Promise Zone", "In Promise Zone")) +
  labs(title = "Assisted Housing Units in Philadelphia") +
  theme_void() +
  theme(plot.title = element_text(hjust = 0.5),
        legend.title = element_blank())

  ggplot(acs_data_for_charts) +
  geom_col(aes(x = geography, y = pct_rentburd), 
           fill = "#7fb0ac",
           alpha = 0.7) +
  geom_text(aes(x = geography, y = pct_rentburd, label = round(pct_rentburd, digits = 1)), vjust = 1.5) +
  labs(title = "Rent Burden",
       y = "Rent Burden (%)") +
  theme_phl(base_size = 14) +
  theme(aspect.ratio = 1,
        axis.title.x = element_blank())

target_tenant_hist  = phl_nhpd_sf_full |>
                        filter(in_pz == "true") |>
                        group_by(target_tenant_type) |>
                        summarise(total_site_units = sum(total_units))


ggplot(target_tenant_hist) +
  geom_col(aes(x = reorder(target_tenant_type, -total_site_units), y = total_site_units), 
           fill = "#7fb0ac",
           alpha = 0.7) +
  geom_text(aes(x = target_tenant_type, y = total_site_units, label = total_site_units), vjust = -.25) +
  labs(title = "Promise Zone Assisted Housing by Tenant Type",
       x = "Target Tenant Type",
       y = "Total Units") +
  theme_phl(base_size = 14)

ggplot(phl_nhpd_sf_full[phl_nhpd_sf_full$in_pz == "true", ], aes(x=earliest_end_year, y = total_units)) + 
  geom_col(aes(x=earliest_end_year, y = total_units), 
           fill = "#7fb0ac",
           alpha = 0.7) +
  labs(title = "Promise Zone Assisted Housing by Expiration Year",
       x = "Expiration Year",
       y = "Total Units") +
  theme_phl(base_size = 14) +
  theme(aspect.ratio = 0.75)

5) Text for Dashboard with Embedded Stats

today = lubridate::today()

phl_assisted_housing = as.integer(sum(phl_nhpd_sf_full$total_units))
  
pz_assisted_housing = as.integer(sum(phl_nhpd_sf_full$total_units[phl_nhpd_sf_full$in_pz == "true"]))

pct_pz_assisted_housing = round(sum(phl_nhpd_sf_full$total_units[phl_nhpd_sf_full$in_pz == "true"]) / sum(phl_nhpd_sf_full$total_units)*100, digits = 1)

pct_pz_total_hh = round(sum(acs_data_for_charts$tot_hh[acs_data_for_charts$geography == "Promise Zone"]) / sum(acs_data_for_charts$tot_hh[acs_data_for_charts$geography == "Philadelphia"])*100, digits = 1)

pz_rent_burden = round(sum(acs_data_for_charts$pct_rentburd[acs_data_for_charts$geography == "Promise Zone"]), digits = 1)

phl_rent_burden = round(sum(acs_data_for_charts$pct_rentburd[acs_data_for_charts$geography == "Philadelphia"]), digits = 1)

pz_assisted_units = sum(phl_nhpd_sf_full$total_units[phl_nhpd_sf_full$in_pz == "true"])

pz_family_units = round((sum(na.omit(pz_nhpd_sf$total_units[pz_nhpd_sf$target_tenant_type == "Family"]))) /(sum(phl_nhpd_sf_full$total_units[phl_nhpd_sf_full$in_pz == "true"]))*100, digits = 1)

pz_elderly_units = round((sum(na.omit(pz_nhpd_sf$total_units[pz_nhpd_sf$target_tenant_type == "Elderly"]))) /(sum(phl_nhpd_sf_full$total_units[phl_nhpd_sf_full$in_pz == "true"]))*100, digits = 1)

disabled_or_combo = round(((sum(phl_nhpd_sf_full$total_units[phl_nhpd_sf_full$in_pz == "true"])) -
                  (sum(pz_nhpd_sf$total_units[is.na(pz_nhpd_sf$target_tenant_type)]) +
                   sum(na.omit(pz_nhpd_sf$total_units[pz_nhpd_sf$target_tenant_type == "Family"])) +
                   sum(na.omit(pz_nhpd_sf$total_units[pz_nhpd_sf$target_tenant_type == "Elderly"])))) /
                  (sum(phl_nhpd_sf_full$total_units[phl_nhpd_sf_full$in_pz == "true"])) * 100, digits = 1)

no_target = round(sum(pz_nhpd_sf$total_units[is.na(pz_nhpd_sf$target_tenant_type)])  /
                  sum(phl_nhpd_sf_full$total_units[phl_nhpd_sf_full$in_pz == "true"]) * 100, digits = 1)

less_than_five_exp = pz_nhpd_sf |>
                      filter(as.integer(earliest_end_year) <= (as.integer(lubridate::year(lubridate::today()))+ 5)) |>
                      summarise(sum = sum(total_units)) |>
                      as.data.frame() |>
                      dplyr::select(-geometry)

five_to_ten_exp = pz_nhpd_sf |>
                      filter(as.integer(earliest_end_year) >= (as.integer(lubridate::year(lubridate::today()))+ 6) &
                              as.integer(earliest_end_year) <= (as.integer(lubridate::year(lubridate::today())))+ 10) |>
                      summarise(sum = sum(total_units)) |>
                      as.data.frame() |>
                      dplyr::select(-geometry)

ten_to_twenty_exp = pz_nhpd_sf |>
                      filter(as.integer(earliest_end_year) >= (as.integer(lubridate::year(lubridate::today()))+ 11) &
                              as.integer(earliest_end_year) <= (as.integer(lubridate::year(lubridate::today())))+ 20) |>
                      summarise(sum = sum(total_units)) |>
                      as.data.frame() |>
                      dplyr::select(-geometry)

twenty_one_plus = pz_nhpd_sf |>
                      filter(as.integer(earliest_end_year) >= (as.integer(lubridate::year(lubridate::today()))+ 21)) |>
                      summarise(sum = sum(total_units)) |>
                      as.data.frame() |>
                      dplyr::select(-geometry)
                                               
no_exp = sum(pz_nhpd_sf$total_units[is.na(pz_nhpd_sf$earliest_end_year)])

Below is the text for the dashboard. I’ve coded it so that it automatically incorporates the relevant statistical calculations. The Markdown document will spit out just the text, but you can unhide code to see the calculations themselves.

This dashboard displays Philadelphia data on federally-assisted multi-family housing taken from the National Housing Preservation Database (NHPD).

The map tool can be used to identify individual housing developments: • points are sized according to the total units at a development • points are colored according to how soon they will expire

As of 2022-07-14, the NHPD counted 33281 total units of federally-assisted multi-family housing in Philadelphia. Of these, 3395 (10.2%) are in the Promise Zone.

The Promise Zone faces disproportionately high rates of rent burden: 40.4% in the Promise Zone versus 22.7% citywide. assisted housing in the Promise Zone targets different types of tenants. Out of 3395 total assisted units: • 26.8% are for families • 24% are for older adults • 9.1% are for disabled people • 40.1% of units have no specific target

Expiration timeline: • Next 5 years: 653 units • 6 to 10 years: 140 units • 11 to 20 years: 1232 units • 21 years or later: 152 units • No expiration date: 1218 units

6) Putting It Together

Once you have all these pieces ready to go, you can upload the relevant pieces to ArcGIS and then assemble everything in the StoryMap.

This will require you to create five separate components:

the NHPD map
the rent burden map
the residential construction permits map
the various graphics
the dashboard text

Once you have created these, you can upload them as appropriate to the ArcGIS storymap and format them as you see fit. This may change a bit each time you make the map, and you may have to do some fiddling with the visualizations; what I’ve created here can certainly be improved.

i) Creating the NHPD Map

The NHPD map is the most complicated part of the dsahboard. You’ll have to take the shapefile created earlier in this script and upload it to ArcGIS Pro. You can then push the resulting layer to an ArcGIS Online webmap. That webmap can be turned into a dashboard which, when properly formatted, can then be embedded in the final storymap.

Uploading the shapefile

Open ArcGIS Pro
Start a new, blank project
In the “Contents” section, right click the basemap layers and remove them.
Under “Map” in the toolbar, select “Add New Data”. Navigate to the filepath for phl_nhpd_expiring.shp and import the file.
Right click on the phl_nhpd_full layer and select “Attribute Table”. Right click on any column name in the table and then click on “Fields”. In “Fields”, go to the “Alias” column and rename each column according to what you want the popup name to be. I went with “Name”, “Address”, “Total Units”, “Earliest Expiration Year”, “Owner”, “Owner Type”, “Manager”, “Manager Type”, and “Target Tenants”. Save the map.
Right click the phl_nhpd_full layer and select “Symbology”. Set the primary symbology to “proportional symbols”, and the field to “ttl_nts” (originally “total_units”–remember that Esri shapefiles have limits to the length of column names, so it will automatically trim them). Set the minimum size to 5 and the max size to 50. Save.
Click on “vary symbology by attribute”, then open the “color” subtab and change “field” to “erlst_ (originally”earliest_end_year“). Set the color scheme to yellow-green-blue. Then, next to”template“, right click the little dot and, under”color properties“, set”transparency" to 50%. Finally, make sure to check the box for “show legend”. Don’t forget to save.
Right click on “Map” under contents, and then select “Properties”. Under “general”, make sure to check off “Allow assignment of unique numeric IDs for sharing web layers”.
Repeat steps 4 through 8 for phl_nhpd_non_expiring, except do NOT add the “vary symbolgy by attribute” with the color setting. Just make the color for all of these points gray. (Do this by click the dot next to “Template”).
Navigate to the “Share” tab and click “Web Layer” (NOT “Web Map”). Set the name to something like NHPD Housing for Dashboard and the summary to something like “Data from the National Housing Preservation Database cleaned and formatted for a Promise Zone dashboard.” Tags might include “Promise Zone”, “Housing”, etc. Set the layer type to “feature”. Set the location to your folder, and set sharing to Everyone and City of Philadelphia, and then click “publish”.

Push the layer to ArcGIS Online

In your browser, navigate to phl.maps.arcgis.com and go to “content”. Open the layer you just created (“NHPD Housing for Dashboard”) in Map Viewer. Title the map “NHPD Housing for Dashboard” and save.
A city basemap should automatically be present. You’ll also want to add the Promise Zone boundaries. Do this by going to “Add”, then “search for layers”. Select “my organization” and search for “Promise Zone”. Select the “Philadelphia_Promise_Zone” layer uploaded by maps.phl.data. Navigate to the “Details” tab and drag the housing layer to be on top of the Promise Zone layer. Save.

Turning the webmap into a dashboard

Click on the “home” tab in the top left corner. Next, click the little square of nine dots just next to your profile picture in the top right corner of the screen. Select “Dashboards”, and then “create dashboard.” Name it something like “NHPD Housing Dashboard (June 2022)”, and add the tags “Promise Zone” and “Housing”.
Click the plus sign in middle of your dashboard template, and then select the “NHPD Housing for Dashboard” web map that you created in steps 10 and 11. In settings, turn on popups, the legend, and the search and zoom features. Click done and then save the dashboard.
Click “add element” in the top lefthand corner. Add a header with a title, and add a sidebar with a map legend.
In the header, add two number selectors: -In the first one, set label to “Expiration date range:”. Set display type to “slider” and input type to “range”. Limits should be from “statistic”, and the field for limits should be “earliest expiration year”. Under the “actions” tab, set the selector to filter the “expiring affordable housing” map layer. Save. -In the second one, set the label to “total units”, display type to “spinner”, and input type to “range”. Limits can be “none”, with lwoer default set to 0 and upper default set to 1000. Under “actions”, set the filter to both the expiring and non-expiring housing layers. Save.
Save the dashboard. Make sure its sharing is set to public. Publish. Copy the link to it. -Embed the dashboard link in the housing storymap. Move it to where you want it.

ii) Creating the Rent Burden Map

Creating the construction permit map can be done entirely in ArcGIS Online:

Open a new map.
Under “add data”, search in “living atlas” for “rent burden”. Select the “ACS Housing Costs Variables - Boundaries” layer published by esri_demographics. Add it to the map.
Filter the layer so that County contains “Philadelphia County”.
Set the symbology to “Counts and Amounts”. Set the attribute to show to “Percent of Renter Households for whom Gross Rent (Contract Rent Plus Tenant-Paid Utilities) is 30.0 Percent or More of Household Income”. Set the theme to “high to low” and the fill spectrum to yellow to blue. Set the outline color to white and the thickness to 1 pt.
Add the Promise Zone layer: select “my organization” and search for “Promise Zone”. Select the “Philadelphia_Promise_Zone” layer uploaded by maps.phl.data.
Turn off popups for the PZ layer by right-clicking the elipses next to it under “content” and selecting “no popup”.
Save the map.
Navigate to your ArcGIS profile. Click on the little box with nine dots in the top righthand corner of the screen and select “dashboards”. Click “create dashboard”. Title the dashboard something like “Promise Zone rent burden”. Click “create dashboard”.
Click the add button in the middle of the screen. Add the construction permits web map.
Click “add element” on the left side of the screen. First, you’ll add a header with a title. Then, you’ll add a legend.
Save the dashboard.
Make sure the sharing is set to public. Share the dashboard, then copy the link to it.
Embed the dashboard link in the housing storymap. Move it to where you want it.
To set the appropriate map zoom level, simply save the web map to the desired map extent. This will automatically update the StoryMap. There’s no way to sync the map extent across separate web maps, though.

iii) Creating the Construction Permit Map

Creating the construction permit map can be done entirely in ArcGIS Online:

Open a new map.
Under “add data”, search in “my organization” for “construction permits”. Select the “PERMITS” layer published by the Department of Licenses & Inspections. Add it to the map.
Filter the layer so that PERMITDESCRIPTION contains “residential building” and TYPEOFWORK contains “new construction”.
Set the symbology to “heat map”.
Add the Promise Zone layer: select “my organization” and search for “Promise Zone”. Select the “Philadelphia_Promise_Zone” layer uploaded by maps.phl.data.
Turn off popups for the PZ layer by right-clicking the elipses next to it under “content” and selecting “no popup”.
Save the map.
Navigate to your ArcGIS profile. Click on the little box with nine dots in the top righthand corner of the screen and select “dashboards”. Click “create dashboard”. Title the dashboard something like “Promise Zone residential construction permits”. Click “create dashboard”.
click the add button in the middle of the screen. Add the construction permits web map.
click “add element” on the left side of the screen. First, you’ll add a header with a title. Then, you’ll add a legend. Finally, you’ll add a rich text element where you can include text that explains the dashboard. Currently, this element reads: “This heat map indicates areas of high residential construction activity. This includes only new construction, not modifications to existing structures, and covers both single and multi-family buildings.” However, this can definitely be improved.
Save the dashboard.
Make sure the sharing is set to public. Share the dashboard, then copy the link to it.
Embed the dashboard link in the housing storymap. Move it to where you want it.
To set the appropriate map zoom level, simply save the web map to the desired map extent. This will automatically update the StoryMap. There’s no way to sync the map extent across separate web maps, though.

iv) Uploading Graphics

To take graphics from this markdown document to the storymap, right click on them in the markdown and select “save as”. Name them, making sure that the filename ends in “.png”. You can then upload the resulting .png files directly to the storymap and place them where you want to. If you want to edit the graphics, you can change the underlying R script.

v) Adding Text

Finally, you can take the text from above in this walkthrough (section 5) and paste it into the storymap as appropriate. If you decide to edit the text for future versions of the webmap, make sure that you also edit this markdown document accordingly so that your edits are reproducible the next time you update the storymap.

vi) Publishing the Dashboard

Finally, you’ll want to publish the storymap, which you can do by clicking “publish” in the top righthand corner of the storymap editor. When you do this, make sure that the dashboard itself, as well as all the underlying parts, are set to public sharing. If the dashboard is public but the makes that make it up are not set to public sharing, then viewers will not be able to see all elements of the map.

7) Outstanding Issues

The issues below remain to be fixed. Some of them can be solved with some persistence; others are simply not worth the effort. You’ll have to gauge the combined abilities of the dashboards team and your own comfort with tools like Adobe Illustrator and Photoshop in order to decide what you want to work on.

Issues that can be fixed now:

The NHPD point map legend currently does not show a color spectrum for expiration dates for the expiring housing layer. I think this might need to be fixed in ArcGIS Pro before pushing it to a web layer, but I’m not sure how.
The year data for the NHPD point map show up in ArcGIS as numerics, not dates, meaning them appear with commas (e.g., 2,022 instead of 2022). This is a problem with Esri shapefiles, I believe, which automatically convert certain datatypes. You can probably fix this in ArcGIS Pro before pushing the layer to a webmap.
The titles of the embedded dashboards could be made more precise.
All the maps need clearer instructions for how to use them. Consider adding these in the dashboards themselves, rather than above or below as plain text.

Issues that can’t yet be fixed:

The Promise Zone boundary is still wrong. City Planning is supposedly working on fixing them. Follow up with them to see if anything has moved forward.
There’s currently no straightforward way to improve the map legends. You could try doing this using something like Adobe Illustrator or Photoshop, which I believe is the standard approach in cases like this, but honestly it’s probably not worth your time for a dashboard that so few people will see.

NHPD Dashboard Steps

Nissim Lebovits

7/14/2022