Data

Tools

Indexes

Applications

Experiments

Awesome

An open API service indexing awesome lists of open source software.

https://github.com/markphamm/fall-2024-carmax-analytics


https://github.com/markphamm/fall-2024-carmax-analytics

Last synced: 3 months ago
JSON representation

Awesome Lists containing this project

README 

        
# CarMax Analytics Showcase Fall 2024



## Welcome to the CarMax Analytics Showcase Repo!



Here you will find the data and files you need to participate in the competition.



To register and find more details on this year's competition, please visit 



### Data Set and Important Files

The data is composed of purchasers and non-purchasers and their shopping journey details, such as number of website visits, marketing touchpoints, geographic details, and vehicle purchase attributes. Vehicle purchasers bought a vehicle from CarMax; whereas non-purchasers did not purchase a vehicle with CarMax .



Disclaimer: This is pseudorandomly generated and manipulated CarMax data. No data corresponds directly to reality, but is inpired by and mimics true data. No data represents current CarMax inventory and business.



Relevant Files:

* `Fall 2024 dataset.csv` - *the data set containing the relevant KMX purchases and non-purchases to be used in analysis*

* `Fall 2024 Showcase Data Dictionary.xlsx` - *information about each feature in the data set*

* `CarMax Analytics Showcase Skeleton Template.pptx` - *slide template suggestion, for reference. Use is not required.*



### Challenge

**CarMax is considering opening new stores in the United States. How should marketing and inventory strategy be approached to drive maximum sales depending on the location of the new store?**



# Frequently Asked Questions



This FAQ contains questions and answers from current and past competitions.

This FAQ will continue to be updated over time as new questions come in and are answered.



## Registration and Submissions Logistics



* Q: Can I register and then add a teammate later?

    * A: You can! If it won't bring your team to more than 4 people, you can email [email protected] to add your new teammate.

* Q: Can you give a better overview of the video segment? What are you looking for in terms of results and the report?

    * A: Video can include you speaking to results, showing graphs, or whatever you think will be helpful to synthesize the analysis you've done.

    You can find the full requirements for the video in our official rules at https://analyticsshowcase.carmax.com/rules.

* Q: Should the report only present the findings? Can I include graphs and bullet points in my report?

    * A: The report is meant to supplement your submission video and should outline your work, methodology, and results.

    It can include graphs, bullet points, and/or other elements to support communicating your process and findings.

    You can find the full requirements at https://analyticsshowcase.carmax.com/rules.

* Q: Do I have to show myself in the video?

    * A: No, that is not required

* Q: Can I present in whatever format I like in the video?

    * A: Provided it meets our submission criteria, present your findings however you feel is appropriate to get your message across most effectively.

    We are primarily judging submissions on the accurate and insightful findings that are clearly communicated.

* Q: How big is the competition?

    * A: The competition is open to any qualified university student (see https://analyticsshowcase.carmax.com for who can participate), so we are hopeful to see people from all different backgrounds and schools across the nation.

* Q: Who is the audience of the report, are they technical managers or business operations managers?

    * A: Managers and leadership across analytics will be reviewing submissions, so you can expect them to have technical expertise and business analytics backgrounds



## Analytics

* Q: Should I explain my methodology for how I analyzed the data and what insights I found?

    * A: The problem statement is primarily asking for your recommendation to our question, so you need to balance presenting your methodology in the report and the video with the presentation of your findings.

    One category of judging is rigor -- are you analyses sound and accurate? Demonstrating methodology is important for that.

* Q: How should I analyze this data? What are you looking for?

    * A: We don't want to be too prescriptive about what to do with your analysis.

    There is not a single right answer to the problem statement -- we are looking for you to give us your answer and support that answer using insights from the analysis you conducted.

    Try not to get too hung up on trying to anticipate what you think we want -- do your best within the guidelines of the contest and find accurate insights within the data.

    What analytical tools/techniques/software excite you?

    Start your approach there.

* Q: Is there a specific software/programming language I should use for the showcase analysis?

    * A: No! You can use whatever app, software, or programming language you are familiar with to perform your analysis.

* Q: Regarding the curiosity judging category, what does 'unique insights' mean?

    * A: When you look at the data you might find more 'obvious' insights that are very clear or apparent trends in the data.

    You will get a higher score in the curiosity category if you find some things that are deeper -- make connections and find patterns in the data that go a bit beyond the surface.

* Q: What kind of recommendations to CarMax are you looking for?

    * A: We want you to look at the data and your insights to make any recommendations to answer the question. These can come in a lot of different formats. There is no one right answer.

* Q: What accuracy metrics should you be using to validate your solution?

    * A: We are not looking for any specific numbers or thresholds.

    Since you have flexibility to define your approach to the problem, you should determine the relevant accuracy metric(s) you need to report and optimize on.



## Data

* Q: Why is the data roughly evenly split between states?

    * A: This is done to make some analytical approaches easier for the data for making recommendations for various states to make the amount of data more equal, and adds some nuance to any cross-state analyses.

    Please refer to the guidance above in `Analytics` for how to approach your analysis.



## Final Round

* Q: What happens if my team is selected as a finalist?

    * A: Finalists will be notified around the 8th of November and present virtually on 22nd of November.

    On the day of the 22nd, finalists will be partnered with a coach to run through things and finalize the presentation before presenting to leaders.

    At that point, all the data and work will have been set. We will make sure finalists have the resources they need well ahead of the presentation day on the 22nd.