Courtney Carlson

After looking over the data provided to Widgets Inc containing 4.5 years of sales and promotional data for 5 regions, there were many insights created. The data proved to be relevant and consistent overall after fixing a few minor functioning errors. The data was split into three classifications; frequencies (categorical data), descriptives (scale/continuous data), and correlations. Each of these sections gave rise to a series of insights due to the numbers they derived. In terms of frequency data, I was able to come up with one major insight; there will be more data for the earlier quarters of the years than the later due to the fact that there is an odd number of data for months acquired since the time range is January to June. The descriptive data gave way to insights that most of the promotional budget was allocated towards media spending and the least amount was allocated towards SMS. The correlations data proves that as an increase in spending occurred in SMS, the advertising sp...

This week we solidified our knowledge on how to complete a data analytics process. We went through exactly what our steps should be and what different aspects go into those steps. First off when you are presented with data, you need to be able to communicate with the analysts in a way that they understand. In order to do so you  have to first know exactly what the problem at hand is and come up with a research question that clearly states what your end goal. The next step would be to clarify what data you need and where you are going to go to get that data. In the class situation we already were provided the data so now we had to organize the data and see exactly what kind would be useful. After you then have to analyse the data by looking at the variables and coming up with facts and findings. After this you know have to turn those findings into insights that ultimately answer your research question. This process involved a lot of work as well as studying. In class we also looked ...

Interesting: 1.)  I think that it is interesting that there are five assumptions that are required for a regression model to work, to be unbiased, to be interpretable, to be efficient and or consistent. I thought that a regression model just proved any correlation between variables but there is more that goes into it then just an explanation. If these assumptions are failed then something in the model has to be chased so that consequences can be accounted for. (Cha 4) 2.) I did not know that there are two types of explicit equations (probabilistic and deterministic). I thought the concept of an equation was that one side of the equation equals the other side of the equation but in probabilistic equations that is not the case. Since there is a ransom error term it changes the outcome of the other terms; the dependent variable, the constant, the independent variable, and the coefficient. (Cha 4) 3.) Logistic regression is similar to ordinary regression in that there is a dependent...

Interesting: 1.) I find it interesting exactly what a buyer persona is. It is a fictional person who represents a particular company's ideal customer. I knew that company's would look into their consumer and target market to see exactly what they are looking for in a company, but having an actual created personal profile of a customer makes the process more personal and in depth. 2.) I also think that it is interesting that the persona gets so in depth with the "day in a life" feature and the finances of the perceived consumer. I never thought that these were features that were important because it varies so deeply from consumer to consumer. Finances is important because the products may be marketed to a certain income class depending on its quality but I was not aware that the daily activities were important since they change so much. 3.) It is also interesting that you can have more than one buyer persona for a product because usually a product is put in a niche ...

This week I only attended Tuesdays class but in that class we learned valuable information. Going into the week I was not aware that there was a difference between facts/findings and insights. I had an idea that facts could be a type of insight or vis versa but that is not the case. We learned that a fact is information that will lead to an insight. Insights are turning that information into an understanding that can be further developed into a hypothesis or experimental question. In our day to day lives we are presented with factual information about many aspects of our lives and how we interpret or develop those facts creates our own insights. After being able to decipher the differences I began to notice how our minds naturally create insights. In my ILP class on Wednesday we were looking at educational stats in underdeveloped countries around the world. After looking at this data, which were my facts/findings, our teacher asked us "what does this mean?" After this questio...

Interesting: 1.)  One thing I found interesting in the "Facts Insights" article is how CMI created a global marketing information system which organised data and presented it in a format that was viewed uniformly throughout the company. This means that at any time and by any person, the data will presented the same based on the same numbers and found through the same process. This created one version of the truth so that the whole company would be on the same page at all times. 2.) I find it interesting how they set apart an insight from a finding by stating that an insight transforms your actions while a finding merely guides them. Later on it then goes into detail about how an insight is the deeper understanding of facts that call for analysis. A fact is something that can just be a surface detail. Insights can be facts but facts are not always insights. 3.) I find the whole-brain mindset interesting in the aspect that marketers need to look at how they can reach the lef...