Answer these questions in best of your abilities within 24 hours. If you are stuck on a technical aspect, write down in words how you would go about answering this questions and what other information would you need.
- Attached is a sample dataset with 100,000 rows of random placements, and media exposure by impressions. What queries would you use to analyze the data? (Hint: you need to think about cleaning the data first. Common data problems include duplicates, missing, errors in the data)
- Now that you identified issues with the data, do you notice anything particular about this dataset? What queries would you use to investigate? (Hint: think about what you know about digital marketing and do you think these would be good exposures?)
- Segment the media exposures into 5 groups. What queries would you use to help you with this? Create a histogram in your answer.
- How many green T-shirts do you think are sold in a year? (This is an open ended question, let us know how you would go about this figure)