Interpreting Data to Make Business Decisions

Please refer to the link of the excel spreadsheet while reading this post or you may have a difficult time following it.

Imagine you are in charge of cutting operational costs for an air transportation supply company due to a declining economy. You are given data on employee performance, salaries, positions, and overtime earnings for each employee. It is your duty to make decisions by drawing conclusions from this information. However, the prevalent question is how to make such important decisions based on the data you’ve received?

To start, we must first interpret the data. This involves understanding, analyzing, and questioning the data given to us. Understanding the data is a crucial first step as you must know what you’re reading. To do this, simply look through the data. What are each of the columns representing? In our spreadsheet the columns are detailing (from left to right) the employee ID number; the date each employee was hired; the department they are a part of; the job performance score given to them by management using a scale of 1-4; management’s comments for each employee; their annual salaries; the annual bonuses earned by each employee; and their amount of overtime earnings for the last year. Now that we’ve established what each column is representing, we must determine which columns possess data that will help us draw our conclusions. For example, the employee ID is probably not a column we need to account for when doing our analysis. I feel the most important are the job performance evaluations, annual salaries, and overtime pay columns. I have chose to avoid the management comments section as this could hold biased data. Also, all non management positions earned a $2,000 dollar bonus so that information is consistent for all employees. Next we actually read the data. Is this data qualitative (referring to data retained by first hand observation) or is it quantitative (referring to data retained by numerical data)? Most of the information that we have chose to look at other than the department column contain quantitative data. This makes our job a bit easier as we will be interpreting numbers, which provides us with a simple means to compare the data.

With a basic understanding of how to read the data, its time to move on to analyzing this information. What patterns do you notice? For the first column (after refining our criteria), we can notice that the majority of the employees were hired in the year of 1987, with a few having been hired in 1992 and 1997. The next column shows us that we are only dealing with 11 different positions held within the company, those of which you can find in the provided spreadsheet. In the third column we find that a 3 must be the average employee score, meaning these employees are performing satisfactorily. Our following column shows the disparities between salaries within each position. The management salaries are clearly outliers. Lastly, in the final column we notice very few employees put in any overtime hours.

Now it is time to create questions based on the data that will steer our final conclusion making process. I came up with a few questions in reference to this data. Does employee performance have any correlation to salary? Based on the data, the answer to this first question is no. We can see that employees with low job performance scores are paid equally if not more than other employees with the same position. Were the new hires after 1987 due to company expansions? I believe the answer to this question is yes, as the jobs added included a research department, an IT department, more salesmen, and more manufacturing positions. All indications that the company has expanded. Does working overtime lead to having a higher job performance score? Our data also confirms this as all employees who worked overtime scored a 3 or higher. Finally, what is not provided in this data set? This table is missing the shifts worked by each employee. I feel this is another area that we could compare with employee performance. For example, maybe working the graveyard shift is causing a decrease in performance.

Once we’ve interpreted the data, it is time to make our decision/s. Our company’s goal is to cut operating costs by 10% from $5 million. After seeing that almost 15% of our company is underperforming while still being paid equally, I think it is time to do some firing, unfortunately. Provided we only have the data we were given available, adjusting salaries will not be a enough of a contribution towards reaching that $4.5 million mark. Because we have the job performance evaluations, this process should be pretty painless. However, to weed out any management bias during evaluations, it may be a good idea to use a secondary source and have the employees also give their feedback on their peers in the workplace. This will identify any employees who may have received higher scores due to manager bias. However, for the sake of this blog, we are only going to base our decision on the data we have available. I’ve decided to let go all non-essential employees or employees in management and also cut overtime. These actions allowed us to reach our goal within a few thousand dollars. To find the last $10,000, the company could look into primary resources like material costs.

Through this blog we have come to a decision that will reduce the operation costs of our company by 10%. We did this by interpreting and analyzing a set of data as well as formulating conclusions that will drive our decisions using said set of data.

Link to excel document used:

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: