Amy Lavin

I read an article last semester explaining how the NBA game has been transformed due to data analytics. The data showed that players had nearly the same probability of successfully converting a mid range shot as they do a 3-point shot. The 3-point shot has a 50% increased benefit over the mid-range shot worth only 2 points. Therefore, teams like the Houston Rockets and Golden State Warriors have taken the information from the raw data and take 3-point shots if they cannot score on a layup.

As a student, I always strive to achieve high grades. During finals, I have a clearer picture of what my grade will actually be. By taking the data of my grades over the semester from tests, quizzes, and assignments, I was able to calculate my current grade. I then analyzed my grades and devoted more studying time for tests in classes where I needed the best final test score to get the grade I wanted.

When working as a Retail Manager for Armark during the World Meeting of Families last semester, I sold “pope attire” to the mass crowd the event attracted. We kept a strong track of inventory numbers and used that data to determine what inventory needed to be replenished during the next re-stocking period. By effectively tracking and analyzing this data, we were able to maximize profits by keeping items in high-demand stocked throughout the event.

An important part of eCommerce is advertising. Companies such as Facebook and Google (YouTube) use data collected from customers to create personalized ads that are tailed to each customer. For example, if someone has recently been shopping for shoes, the software will in return display more shoe ads for a person. These tailored ads create a better experience for not only the customer, but for the seller because they are more likely to receive more business because their customer bases come from advertisements that are specialized to someone’s personal interests.

Over the summer, I decided that I was interested in figuring out just how much money I spend while I am at Temple’s campus. So, when the fall semester began, I started to record all of my purchases that occurred at Temple in an app on my phone. Besides the price, I would also record what kind of purchase it was, like food or clothes. When the semester concluded, I calculated the total I spent for the semester, as well as how much I spent in each category. Through the collection of this data, I was able to turn it into useful information and understand where I spend the most money and what on. Since then, I have concluded that packing my lunch would be in my best interest.

Working in retail for Dick’s Sporting Goods this past summer has taught me a lot about data and efficient ways to store it. For example, we have devices that not only keep track of our own inventory, but also the inventory of other Dick’s Sporting Goods stores in the district. This way if our store doesn’t have the product or size in stock that a customer is looking for, we can scan the item and up will come all the stores in the district that have the product. This has proved to be effective for our sales as well as the customer because they are getting the product that they desire.

I shadowed a co-worker who is currently working in the demand-planning department at my current job one day. He was working on a project that had lines and lines of information on an excel sheet. On that excel sheet was information that could estimate how long a distributor or a vendor will take to ship out a product to us. One column would have the vendor’s name, the next one would state the location, the next one have the approximation of when we received the package or if we have not. With the use of the filtering function on excel, he was able to create a graph that condensed the information and gave him an approximation of when a package or product will be received. For most vendors we could see that it can take about 3-5 days, then there are the more extremes which maybe coming from Italy, which can take from 6-7 weeks. (Fun Fact!: Most products we receive from Italy is shipped via boats). It was really neat to be able to see all of the data that was raw, to be turned into information that can help the company long-term.

As a student in high school I spent my last year working in retail, more specifically, Sears. While working there, the management had been going under some changes in how we go about interacting with our customers. Ultimately, the goal was to “convert customers into members”. Our POS systems now required us to ask our customers for their basic info (name, email, phone #) and input this data into our database. Just from giving this data to us, they became members and could now expect several coupons in their email on what they are most likely to buy next. This was all possible due to the information that was mined from the data collected.

During my internship with an Insurance broker, I gathered raw data from my clients like their premium, their total insurable assets, previous year’s losses and so on. I analyzed that data to determine whether or not their premium would rise or fall. For example, if they had a year with no losses, and they paid their premiums in full, and on time, then I would nudge the carrier to lower my clients’ premiums for the next policy period come renewal time. The opposite could happen as well, my cleints’ can have a terrible year filled with losses, and then I would not be able to control the carrier from raising the premiums, and would proceed to look for another carrier to place that client with. That continues to be my experience with raw data turning into information.

An example of raw data being turned into information comes from 2101 with professor Lavin from last semester. Amazon likes to track what items users often view and or purchase. With that data, Amazon is able to come up with suggestions and reminders for users. For example, if someone buys their laundry detergent from Amazon; within a month or so the user may see the detergent popping up as a suggestion for them to buy more detergent as they may be close to running out.

I worked as an underwriting intern for an employee benefits consulting firm where I gathered raw data regarding high claimants, such as number of high claimants, diagnoses to determine future losses, dollar amount over the threshold, and RX claims. I would use this information to figure out how the high claimants were affecting the overall insurable population, which helped to determine if rates needed to increase or if the client could benefit from switching carriers. In this way I turned the raw census data into information that we could use to benefit the client.

Currently, I am working a part time internship at Philadelphia Gas Works in the Gas and Acquisition Department. PGW controls all the gas that reaches the city gates, including that provided by external suppliers. In the department, they complete a lot of work with Excel and a system called Retail Operations to keep track of different suppliers who want to provide citizens in Philadelphia with gas. The department I work for needs to tell the suppliers how much gas they are permitted to bring into the city to provide for their clients. Using Retail Operations, the temperature is used to forecast how much gas will be needed for the week. If the temperature is especially high that week less gas is needed, if the temperature is drastically low then more gas will be needed. Forecasting uses past data, how much gas was used at the particularly temperature last year, and applies that knowledge to how much gas will be needed for that same temperature this year.

I used to work for a limousine company where I was assigned to insert date into their systems. By data I am referring to all pickups, drop-offs, times of arriving, revenue from each job, reviews about drivers, ..etc. So after inserting and organizing all the date; we find out who are our returning customers, loyal customers, what seasons or months that we are most busy in. Also an important point is knowing more about our drivers if they are satisfying the customers. All of which prepares the company to know how many and which drivers to have available in specific seasons. We also get a feel of what jobs to prefer over others because some might be taking more time and isn’t really worth doing anyway.

For my final project in Data Science I used data from the Federal Election Commission (FEC) to come to conclusions about the effect of Dark Money in the 2012 Presidential and Congressional elections. I downloaded data that outlined all individual contributions as well as contributions coming from PAC’s and aggregated it into a working model on Tableau. From there I found that in 2012, the congressional campaigns that spent the most outside money didn’t necessarily win. Also, I found that in both the Congressional and Presidential election that the most effective use of outside money was on ad buys.

Past experience in the management field of a franchise business proves how I can relate the usage of raw data that was turned into information. While; working in a franchise store, at the end of the day there is always a report that takes place. Which; explains when we had the most sales, how many customers we had in each hour, and a lot of more great data! That data was download on the computer and turned into information that could make the franchise more profitable. Sharing an example; after the process of turning the data into information I was able to determine how many employees I needed to assign for each shift. I knew what products are most wanted during specific hours resulting in what products I needed to order for my next shipment. The best thing about raw data, numbers, and information that it is quite impossible for it to be wrong. By reporting data at the end of the day and turning it into information makes a great business strategy.

Previously I have worked for a store called Five Below. As each product is being purchased there is a system in the back office where we can track which products are being sold and the quantities of products being sold. From viewing these we are able to recognize which categories or products are sold the most.

As a Risk Management and Insurance major student, I understand that collecting raw data from potential insured such as, their names, addresses, genders and occupations is extremely common and important for insurance companies to determine how much risk each potential individual insured has. The insurance companies compare their potential clients’ raw data to their huge database in order to turn the raw data into useful information-risk levels of the potential clients. The risk levels of the potential client are important information for insurance companies because insurance companies rely on the information to determine the premiums for their potential clients.

Now in 2016 many scientist are finding raw data and putting it to use. In my experience, I have seen raw data turned into information used in politics. In 2012, Obamas staff used metadata to focus on issues that interested voters. They also would use Facebook so when people liked or followed Obamas campaign it would give staffers much needed information about his voters and there network.

As a senior graduating in May, I conduct a lot of research on the Human Resources field to stay abreast of current trends and for interview preparation. Deloittte’s Human Capital Management sector recently posted an article on how Human Resource Departments will soon leverage data analytics tools to conduct performance reviews and manage talent. Managing talent is the new approach that companies are taking to ensure there employees are updating their skills respective to their line of work. Organizations will have their employees take online courses. Using this data, the organizations will then be able to gauge where their employees are in terms of the skillsets necessary for that industry. By leveraging this data, companies can continue to identify skill gaps and offer courses to help employees fill those gaps.

As a baseball fan there has been a lot of examples over the last couple years of how data has been turned into information that is used make a team more successful in preventing and scoring runs. One of the most obvious examples has been the use of shifts in the MLB. Teams have had a lot of data on where hitters have hit balls on the field and what types of pitches and pitch locations caused this. So teams have used this data to predict where a hitter will hit the ball when they are pitched to a certain way and in order to prevent that batter from getting on base they shift their fielders to the certain spots they think they are most likely to hit it (rather then keep the fielders in the classic/typical fielding spots). This strategy was criticized at first but now has been adopted by the majority of the league very quickly.

I ran a camp kitchen last summer in St. Croix, and kept a detailed inventory of food products and prices. Since the cost of food in St. Croix is much higher than state-side, I had to track every dollar spent. I collected data on each item, its price, quantity, where I bought it, and how much I actually used for one week. With these data pieces, I was able to generate information on which store had the best prices and how much of each item I really needed. With this information, I was able to stay hundreds of dollars below budget and create an efficient operations standard for the kitchen.

I read an article a while ago on the 76ers, and their strategy behind consistently losing games on purpose. In the NBA, the teams with the lowest records get the best draft pick opportunities for the following season. The 76ers have gotten both praise and much criticism from both the NBA and its general basketball community for intentionally tanking, however they’ve taken the statistics of wins and loses in games and used that data to discover that consistently underscoring by a specific amount of points each game can assure themselves of having a strong team next season. This is an investment decision no different then any organization might make at some point, to take a hit one year to assure themselves a better long-term gain in the future years to come.

I worked as an intern for a financial magazine production company back in Singapore. I was once given the responsibility to observe and report trends from what kinds of information magazine users prefer. I was given a huge chunk of data which showed how many people preferred reading huge chunks of informations compared to images and infographics. There was obviously a greater number of individuals who preferred images and infographics to chunks of paragraphs. From this I came up with the recommendation that the magazine should incorporate more images and infographics, which led to the increase in sales.

During my time working for Postmate’s, I was enlisted to the “street team”. If youre unfamiliar with Postmates, it is essentially a delivery service app that delivers food or whatever you want/need within an hour to you. With the street team we had to do some flyering, or “canvasing” to get the word out about Postmates in Philadelphia while, then it was still early and not as big as it currently is. We took the transnational data as well as all of the customer location data to see where the most popular spots for Postmates in the city was. Using that we could see what areas needed more attention to fully cover all of Philly and try to inform all areas of the city.

In high school, I ran cross country. When it came to qualifying for states, it was hard to compare runners across different districts, because we all ran on different courses in different weather conditions. They had all the information about the courses we ran and the time we ran, but that raw data didn’t do much to explain who deserved to go to the state championship. So, the committee had to match up our “numbers” that we wore during the race up to the time we crossed the finish line, and compare those statistics to runners from other districts. From that, they were able to decide which runners had run the fastest on the hardest courses, and who deserved to go to states.

Over the summer I worked for a country club called Squires. The day of the week would determine how many members would show up. After a while of gathering the data of which members and how many members would come on certain days, that data would determine what would needed to be properly down for that day. We used that data to plan accordingly as to how much food, drinks, and workers we would need for that day as well as set up a schedule to the tee times that they desired.

At the country club I worked at over this past summer, we started having live music on Friday nights. For the first month of the summer, we had a different genre play each week, and in the following days asked the members what they thought of the performer. The ‘Jimmy Buffet” style of music got the highest appreciation by far, so we stuck with that genre for the rest of the summer, and it was a big success.

I am on the rowing team here at Temple and last year I experimented with my diet as how it impacted my physical output I can produce at practice. Everyday I would chart my food and also record the time it took me to complete the workouts at practice. After a while I looked back to see the results and could determine the best combination of carb/protein/calorie intake that my body best responds to so I can improve my fitness and speed.

At my current employer we have setup our virtual server environment to monitor and report on several key factors, two of them being Memory and CPU utilization. We can see peak usage hours, low or no usage as well as when both get over utilized. The data metrics are reported in real time as well as daily summary reports. We can adjust resources on the fly (most of it is automated now) in order to try and reduce contention for system resources. If certain application servers are constantly pegging resources we can make adjustments when needed.

An experience where I turned raw data into information was when I participated in the Data Analytics challenge last semester. My group members and I looked at raw data from the Pennsylvania Ballet website, transferred it to an excel sheet, and created a pivot table to utilize the information. We were able to discover what state and city sold the most ballet tickets, what show was the most popular, and the method in which customers purchased their tickets. We then used all of that information to come up with our marketing campaign.

In high school i worked a country club where perfecting service was always our goal. There were 4 types of memberships you could get for all different prices and upon choosing one we asked a few different questions such as drinking preferences, kids name and age, food allergies, and a few others i cant exactly remember. From this information,we were able build on as they came in for dinner and eventually could predict the inventory needed to meet the needs of customers

I love to watch movies and TV shows and have noticed that big data plays a huge role is businesses such as Netflix or Comcast. When I watch a movie or show on say Netflix, they will make suggestions on which shows or movies you may want to watch that are similar to the ones you’ve previously watched. For example I was watching a documentary on Netflix. After I had watched it, there was a category on my page that said “Because you watched…” It gave me a list of other documentaries that were similar to the one I had just watched. This is important for customers of Netflix and other similar companies because it will help them expand their range of customers and improve customer satisfaction and well as increase profits.

The place where I work, they were used to be analog in their some place. Now they updated almost everything in digital format. For example, they use to keep employee attendances just punching a card. Now they gave a card with chip which employees have to swipe or touch with a device then device can collect all the data.

A few years back I worked for a car wash. It was the busiest wash in town and we would often have days where the line of cars was backed up onto the street a few blocks down. By taking the raw data from our average sales we were able to determine whether or not we would have enough soap and materials to withstand the massive demand on those busy days. Once we gathered that data, we put it into a broader perspective and use the raw data itself to calculate how much the wash truly costed per car. By looking at the data and producing information, it was determined that the wash profited tremendously due to charging $8+ per wash when the wash truly only spent close to $0.10 on supplies per car. The profit was massive and gathering the raw data made it simple and easy to view our goals.