SQL Sequel and Firefox

My second project at General Assembly involved learning how to use queries in SQL to aggregate data from a Mozilla Firefox beta test called a Test Pilot. There were some challenges at first from going to Excel to SQL since SQL was language based and had no interface and a limited UI. But as I used SQL, I quickly fell in love with the freedom and the creativity that was allowed when writing your own queries to find exactly what you were looking for. One part I didn’t love though was the preciseness of Syntax as I had had queries broken by single misplaced punctuation but that was the tradeoff.

The second project at General Assembly had a very clear objective which was to argue if developers should focus on Bookmarks function or the Tab function. As you’re reading this blog, you probably have several tabs of your own opened and are already forming your own opinions on this or maybe imagining a time when we didnt even have tabs. The data we were provided with was surveys filled out by the test users and a record of all of their actions from opening windows to even clicking links. When I looked at the data however, it was much different from the straight forward excel sheets I had been studying up until now because this data required a key to decipher it. It only took a few minutes to adjust though but it was an interesting experience having to reference things until I realized I could just insert the key straight into the cells.

To approach the objective of figuring out if developers should focus more on bookmarks or tabs, we first had to figure out who are our users which brought me to my first market analysis of the actual customers. In my first project, I analyzed the market through the properties but this time I had actual information on all of the users from things like gender to age to even what they like to do on the internet. To do this, I ran a battery of queries through SQL to get all the numbers I wanted, something that would’ve either been extremely tedious or not feasible to do in Excel. Then I visualized the results in bar graphs.

Through this analysis of the demographic and unorganized bar graph visualizing their browsing habits, I can find direction in where to take my further searches. For example, my initial theory was that people who repeatedly visit the same websites like game sites might more inclined to bookmark things while people who visit websites that provide large amounts of fresh content like news websites might be more inclined to open several pages at the same time through tabs. I also had a theory that older users might be more inclined to use tabs not as much as younger users because of the disposition of older people to be against change.

Some of my theories were correct while some ending up being a bust but this trial and error of diving in data and hoping to find gold was what I found exciting. I found the complexity of trying to understand your audience through the data was also challenging. The amount of angles I could’ve approached this project was endless which is how I learned the importance of setting objectives and goals so you don’t go off track.

In the end there were pros and cons for both bookmark and tabs and honestly an answer to focus development on which one over the other one was hard to find. The limitation of the data brought its own set of problems, only around twenty percent of users answered surveys which means that more data might be required. I had to side slightly over more for tabs though as we’ve had bookmarks for a much longer period of time so tabs deserve an equal amount of development.

For my full presentation with the other slides, my cleaning process and my queries, you can look at them here