Friday, July 11, 2014

Subscribing to Dog Food

I'm sure most dog owners out there can understand that I want to get a break on the cost of dog food. Our youngest dog has allergies and needs limited ingredient, unique protein dog food. It's not that cheap. A few sites offer a discount if you subscribe and get a scheduled shipment. It seems Petco has a good deal - they give a discount along with free shipping and we also get reward points.

The question that remains is how often do we need to receive our shipments. We buy food about once a month, but I'm not exactly sure. Of course I'm going to turn to Tableau to help me figure this out. First step - using my receipts to input the data into Excel (you could also export your bank transactions, but I don't always buy dog food when I go to pet stores).

Save the Excel file, open Tableau, connect to data... and we're off!

I start by looking at the Dates of Purchase, which I drag out onto the row shelf by right-clicking and dragging. This gives me a few options of how I want to view my date field (I select the Discrete option at the top). I use this feature every day and it saves me so much time!

Next I want to calculate the day difference between the previous date and the date of purchase. I start by creating a field called "Previous Date".

I bring that out into my view as well (using the same right-click and dragging). Because the field is a table calculation I need to set how I want it to compute it's values. Clicking the little arrow on the right side of the pill will give me the option to Compute using > Table (Down).

Now I can calculate the day difference. Since the Previous Date field is an aggregation, the Date of Purchase needs to be as well. There are many different aggregation options I could use, but I chose the attribute option since I know I'm looking at each possible date and the most granular level.

Double clicking on my new calculation adds it to my view. While this is already very helpful, the question I am trying to answer is how many weeks between deliveries do I need. I could simply modify my calculation to take the day difference and divide by 7, but I created another calculation so that I could see the values side by side.

I added a few additional attributes to my table and hid the Previous Date column because it's just necessary for the other calculations.

You can see that I could get by with receiving an order every 3 weeks. I'm hoping our dogs continue to make the most out of a bag of food and that we aren't stranded without food. I do not want to pay for a bag of food every 2 weeks.

Thursday, January 16, 2014

Wedding Map

If you don't know, I'm currently planning my wedding to a beautiful woman. She also works in the business intelligence arena so portions of our job responsibilities are similar.

One of my favorite things we do together takes place on some Saturday mornings when we each grab a cup of coffee, sit at the dining room table, and work on pet projects. Mine typically involve Tableau Public, Fiverr, or our wedding hashtag #BrandiLovesTara.

We both have a desire to enter a contest put on by Visualizing, but that hasn't happened yet. There is so much we want to learn that we are constantly getting sidetracked (much like I am right now).

I wanted to talk about our wedding invitations. Maybe you have seen invitation with a little map plotting the different sites related to the wedding events (here is an example). With lots of out of town guests I think a map (in addition to directions) would be helpful. Google Maps provides a way to save locations to a saved map. You can even change the look of the pins and the background map. Unfortunately, with a grey map street names are not available and the original Google Map looks like this.

My next thought was of Tableau and the maps available. I know they look like what I want and that I would have more control over the washout, labels, shapes, sizes, etc. So I took my locations and put them into Excel. I used to put in the addresses and find latitude and longitude.

I then used the latitude and longitude from my file to plot the different sites and made slight modifications to the map (such as showing streets). A search on Google provided appropriate images, althought I changed their color using Microsoft Office Picture Manager. Back in Tableau, I set each location to be their respective image. Here's the map:

What do you think? Please comment below!

Saturday, January 4, 2014

Burpee Equivalent

My Aunt Stacy posted an interesting article on Facebook regarding the amount of burpees a person must do to counter junk food they eat. I find this type of information extremely important because it helps to educate individuals on the effect of their eating habits. Many people indulge in the immediate satisfaction of junk food but do not realize the long term affect this has on their physical appearance.

When I read the blog post I immediately thought about how I could visualize it. I took the data, threw it into Excel, and then used Tableau Public to make it interactive. A user can select their general weight and see the number of burpees they must do in order to burn off the calories of the specific item of food.

Check out the whole thing here:

And here is a static image:

Tuesday, November 12, 2013

Toronto Condo Bubble

As a full time Senior Analyst I look at large data sets and build sets of dashboards to answer business questions for our clients. That said, I am also an entrepreneur and I have a "gig" for purchase on a site called Fiverr. People can purchase my "gig", send me their data, and I will build them a dashboard. This gives me an opportunity to work on some pet projects with different datasets.

One of those projects was for the Toronto Condo Bubble. I was asked to create an index of housing prices in Toronto by type of home from January 1996 to August 2013. I added the ability to switch between viewing the price or the index. Additionally, you can use the slider to view specific months. By clicking on the value in the color legend, it will highlight the respective line in the trend above.

You can view the dashboard live on the Toronto Condo Bubble website, here, or you can check it out below. Yet again, feedback is appreciated!

Pinnable image:


Wednesday, November 6, 2013

Updated: Rights of Same Sex Couples

Due to the impending legalization of gay marriage in the state of Illinois, I have updated my dashboard that visualizes the rights of same sex couples. To make it even more fun for you, I am embedding it below, so you should be able to play around with it right here in this blog post! Want a pinnable image? Here you go:

You can select various issues and see the color of the states change to reflect the laws in effect. Additionally, the circles represent the number of same sex couples in each CBSA. Hover over a state or CBSA for some extra detail.

Do you like it? Is there something you would like to see me add? Comment below!

Thursday, October 24, 2013

Visualizing Commuting Data

I listen to NPR pretty much every morning. I love getting my news while I drive, saving me from having to read long articles later. This morning there was a bit about commuters and I realized how much I missed taking the train into work every morning. I was able to read my newspaper, listen to music, check my email, catch up on social media sites, enjoy a podcast, etc. So decided to download some census data about commute time by various types of transportation (driving alone, carpooling, public transportation, walking, other, or work from home). Here is what I have created:

The first dashboard initially gives a high level overview of the different types of transportation and which state has the highest percentage of their population using that type. The bar charts below show each state and the percent of their population using the various forms of transportation. In the live version of the dashboard you can sort the columns or you can select a single state and see the distribution.

The second dashboard shows the same high level information at the top, but the map below shows different CBSAs colored by the average time it takes for an individual to get to work. Obviously, the bigger cities have a longer commute time (most likely due to traffic congestion). The size of the circles represent the number of people being considered in each city.

The third dashboard focuses on a single CBSA, selected using a filter on the right. This dashboard shows various ZCTAs (zip code like areas defined by the census bureau) and the number of people in that ZCTA that choose a certain type of mode of transportation, which is also selected using a filter on the right. In this case, you can see which areas tend to have a good number of people using public transportation.

Friday, October 11, 2013

Wedding Hashtag

My partner and I both love our technology. We appreciate social media and plan on incorporating it into our wedding through the use of a hashtag. Our guests will be able to snap a photo, apply filters in Instagram, use our hashtag and it will appear in our wedding feed. Guests can also send that Instagrammed photo to Twitter or directly tweet about our wedding using our hashtag and that will also appear in our feed.

Our plan is have these two feeds (Instagram and Twitter) up during dinner or dancing or something. We hope our guests take part and have fun with it - make it into an interactive event. I wanted our wedding to be something different. Simple, elegant, with some traditions, but also I wanted pieces of our wedding to break the norms. I want our guests to feel comfortable using technology - as long as it is about us. :)

Seriously, though, I hope people have fun with it and see it as a form of entertainment and not rudeness. I'm sure some people will think it is tacky, but it's us and that's what that day is going to be about. #BrandiLovesTara