IRE Hands-On: Real Estate Data with Tableau Public

IRE Hands-On:
Real Estate Data with Tableau Public

In this demo we’ll create a workbook describing the real estate trends of a city or neighborhood. While we use a neighborhood in Washington, D.C. (Shaw), the techniques and dashboard could be applied anywhere with real estate data.

Get the data

1. There are several places to get real estate sales data. To get the best result, you want one record for every home sale (address) and latitude and longitude for each address.

Redfin allows download of search data by 500 records at a time. You must be logged in to download.
Your local MLS may allow you to download sale records.
Many local governments allow you to search home sale records. Some allow download.

2. Make sure your data is in the right format:

Workbook has one sheet with data
Header row is row 1
No blank rows or columns

3. Connect to the data.

Open Tableau Public and Click “Open Data.”
Choose Excel & find your file
Select “Single table” and find the first sheet
You know you’ve done it right when all your fields show up in the Data Window on the left hand side of Tableau.

Map all the home sales.

· In the Data Window, ctrl-click on Latitude, Longitude and Address and click “Show Me.” This will result in every address being shown as a separate map. To get them all together, drag Address from Rows shelf to the Level of Detail shelf.

· You will see a world map, because some points are on “Null Island,” or lat,long of 0,0. Right-click on that point and choose “Exclude.” Your maps should now look like this:

· Now drag and drop fields to get color and size on the view. First, drag “LAST SALE PRICE” from the data window to size. This will size each home sale by the sale price of that home.

· Next, drop “BEDS” onto color. This will make your view look like this:

· Now let’s do some formatting. Click the arrow in the top right of the color legend field and select “Edit Colors.” We’ll choose an orange-blue diverging color scale to add contrast. Don’t use green-red unless you are attaching good-bad judgments to the values, which in this case we aren’t.

· Now go to the drop-down on the color shelf (see image below.) There are some overlapping marks, so let’s add a border and some transparency to the marks.

· Finally, drag BATHS, Sale Date, and HOME TYPE to the Level of Detail shelf. We’ll use those later.

· Right-click on the sheet name (“Sheet 1”) and rename it “Sales by Location.”

Create price trend view

· Create a new sheet.

· Drag LAST SALE DATE to Columns and LAST SALE PRICE to Rows.

· Click the “+” on the Date pill to expand to quarters, then click again to expand to months. Then drag Quarters off the view and into a blank (grey) place. Your view should look like this:

· Unfortunately, this shows the sum of all sale prices, which is not meaningful. We can click the drop down on the LAST SALES PRICE pill to change to average sales price:

· But we’re still missing something. A better measure of the market is price per square foot, because the simple average can be distorted by the sale of very large or very small homes. Price per square foot isn’t in our data, so let’s create a calculation.

· In the data window, right-click on LAST SALE PRICE and choose “Create Calculated Field.” Name it Price per Sq. Foot and make the calculation: sum([LAST SALE PRICE])/sum([SQFT]). The sum() function makes sure you don’t miscalculate averages on grouping. Your calculation should look like this (look for the green check to know it is ok.):

· When you click OK, this new field will show up in your data window. Drag it out to the view and drop it on top of AVG (LAST SALE PRICE) to replace it in the view.

· Finally, let’s make sure we don’t have any null values distorting the data. Drag SQ FT form the data window to the Filter shelf. Choose to filter on All Values, then Special à Non-null values.

· Your view will look like this:

· Finally, rename your sheet “Price per Sq. Foot.”

Create the dashboard

· Create a new dashboard by right-clicking next to your last sheet and selecting “New Dashboard.”

· Size it to your page size: click :”Edit” in the size section in the lower-left of the dashboard. This will help you avoid scrollbars and make sure the viz looks good in your page.

· Add the price and location views by dragging them onto the dashboard, or just double-clicking them

· Arrange and lay out the views by dragging them. Your dashboard should look like this:

· Change and edit titles by right-clicking on the title and choosing “Edit.”

· Add any explanatory text by selecting “Text” from objects and dragging onto the screen. Here, we’ll add our source information as text.

· Add a Quick Filter on BEDS

o Click on the top right corner of the location view à Quick Filters à City

o This will apply only to the lower view, so align it there on the dashboard

o Bonus: In the data window on either sheet, right-click on BEDS and select Field Properties à Number format. Then select Number, 0 decimals, so you get only whole numbers.

· We also want to add a filter on HOME TYPE.

o To do that, you need to go to the map sheet and drag HOME TYPE from the data window to the Level of Detail.

o Then go back to the dashboard, and you can add the Quick Filter from the Sales by Location view.

o Make this filter global by clicking the drop-down in the upper right and selecting “Make Global.”

· On the lower left of your screen, Double-click on Title in Objects and title the dashboard.

· Edit your tooltips for each view.

o Select the sheet

o On the main menu, select “Edit à Tooltips.”

o Format as desired

· Right click on legend and filter titles to change and format them. Right-click on headers to remove them.

Test the interaction

1. Click around and see how it works

2. Fix the size legend so that home sales keep their sizes as you change from one to another

Your dashboard should look like this: