Today’s webinar is a really exciting one because I am joined by none other than Caspar from Silk.co. Silk is a data publishing platform which helps you visualize data in a fun and unique way. Together we’re going to be analysing the San Francisco housing market – a place I’d love to move (hint, hint import).
Get data with import
As always, the first thing we need is the data. We’re looking at the San Francisco housing market specifically, so I’m going to extract the housing information from Zillow.
Now, the first thing you should do anytime you’re looking to extract data from a list – like the housing options on zillow – is to pop the URL into our Magic tool and see what happens. Luckily for us, Magic works great on this site (it even paginates!) so all I need to do is delete any unnecessary columns and download it as a CSV.
Note: When you download your CSV, make sure you change the number of pages to the maximum of 20.
I did this process for both the rentals and the properties for sale – which took me less than 5 minutes! Now, it’s over to Casper at Silk…
Analyse data with silk
After a little bit of data clean up in Google Sheets, Caspar uploaded the data into Silk by pasting in the link to his Google Sheet (the sheet will need to be public) and selecting the sheet he wanted to viz.
Once your data has been imported you’ll be able to see a preview of it. From here you can do a little extra data cleansing such as separating values by columns – so you can have “5 bedrooms, 4 baths” as two entries. Once he’s happy with his data, Caspar imported it and began vizzing.
On the next screen – in Explore mode – Caspar showed us how to play with Silk’s different sorting functions. It’s handy to know how these features work so that when you start trying to analyse it you know what is possible.
The first way to analyse this data, is by putting it on a map. Using Silk’s filters, we can then look at individual zip codes and housing types to see how they are spread out across the city. Using the same mapping feature, you can also sort your data by estimate and see which area of the city has the most expensive properties.
The next analysis was to look at a scatter plot of the asking price vs the square footage. As you’d expect the larger properties are more expensive, but it’s a great way to find some outliers.
Next, Caspar made a simple bar chart of the types of listings so we could see how many condos there were vs free standing houses.
The really great thing about silk is that you can play with the data yourself to make new graphs and visualizations!
Join us next time…
Next week I’ll be doing a classic “Getting Started” webinar, where I take you through each of the different tools you can use to get data from the web.