Name Lisa Dawes
Choose the SQL syntax option you learned about in the course. This is the syntax you'll be required to use in the take home exam. Postgres
Our database has 3 tables: transactions, products, and customers.
Using the sql file provided answer the following questions. You can assume that the sql file works properly. If you chose SQLite use https://bit.ly/2GEMY53 If you chose Postgres use https://bit.ly/2LsEJwm Create a query that returns the name of the most popular item in every state and the state. SELECT state, MAX(Product_Count) AS Popular_Product FROM ( SELECT state, COUNT(products.product_id) AS Product_Count FROM customers JOIN transactions ON customers.customer_id = transactions.customer_id JOIN products ON transactions.product_id = products.product_id GROUP BY state, products.product_id, products.product_name ) Q GROUP BY state
Create a query that returns the name and purchase amount of the five customers in each state who have spent the most money. none
Create a query that returns the five most popular items for users with a ‘gmail’ email in the past 30 days, based on number of sales. none
Experimentation: Answer each question with no more than 250 words. We’re launching a redesigned home page for our site, and want to study how it affects visitors. What are some of the key things you’d want to study? What metrics would you measure? Provide an explanation for every metric. A few key things that I would like to study include:
-
Number of total site visits - To assess web traffic that associated with the redesign
-
Number of site visits per user - To assess if customers are visiting the new web page more often
-
Number of returns to the site per user - To assess if customers are returning to the site
-
New users vs. returning users (plus corresponding page activity) - To evaluate differences between new and returning users and how the redesign has affected that traffic
-
Amount of time spent on the site - To assess if . customers are spending more time on the site and what features/pages on the site may cause . them to linger longer
-
Sign-up rates - To assess if more customers are signing up/buying product because of re-design features
-
Incoming questions from users - To evaluate if customers have a particular question because of the redesign (i.e., is something not clear?)
-
Popular Pages - To assess if any web pages increase in popularity due to redesign features
-
How users get to the site - To assess if the redesign if affecting key word searches
Other Considerations: Cost of the redesign, Customer Feedback
Using those metrics, how would you statistically evaluate if the new site was a success? If overall site traffic and sign-up rates/product sales increases I would consider the redesign a success. As a part of the evaluation I will work to see if there are any other project/factors that might have affected these metrics as well (e.g. new ad campaign). In addition, if I receive positive feedback and/or less questions from customers I would consider the redesign a success.
How would your approach vary if we could offer our redesign to a subset of customers in perpetuity? I would look at the same metrics however, following this approach I would be able to evaluate different user experiences based on the original site and redesigned site. I would also be able to possibly customize the redesign for that particular subset and/or offer/test more expensive features before rolling it out to all users.
Using the data in the link below, attempt to model a customer’s propensity to join our loyalty program: https://bit.ly/2Etq2Ux Please attach your code here via PDF exported from a jupyter notebook. L_DawesThinkful_Take_Home_Exam.pdf http://nikib.ro/wn/screenshots/24207f74570e-L_DawesThinkful_Take_Home_Exam.pdf
Attach a one page, double spaced write up detailing your approach. Write_Up__Take_Home_Exam.pages http://nikib.ro/wn/screenshots/77c4f82438a7-Write_Up__Take_Home_Exam.pages
vesvxpr8rtxlnedhvesvxlgctgdm736n Typeform sent you this email on behalf of a typeform creator. We aren't responsible for its content. If you suspect abuse, like suspicious links, please report it here.