Learn how to SOLVE a data analytics case study problem

Today I'm tackling a data analytics problem asked in data engineering and data science interviews. For most of these questions, there are multiple solutions involving both analytics and writing a SQL query.
Follow along and solve the case study on Interview Query: www.interviewquery.com/questi...
Relevant Links:
DoorDash analytics case study guide: www.interviewquery.com/p/door...
Analytics case study guide: www.interviewquery.com/p/data...
👉 Subscribe to my data science channel: bit.ly/2xYkyUM
Use the code "datasciencejay" and get 10% off data science interview prep 🔥 : www.interviewquery.com/pricin...
❓ Check out our data science course: www.interviewquery.com/course...
🔑 Get professional coaching here: www.interviewquery.com/coachi...
🐦 Follow us on Twitter: / interview_query
Ready to find data analysts and scientists who can tackle complex problems like the one in this video? Outsearch.ai leverages cutting-edge AI to identify top talent who possess the analytical skills and SQL expertise showcased here: www.outsearch.ai/?...
0:50 - Difference between product metrics and data analytics questions
2:32 - Data analytics problem description
3:00 - Framework for tackling data analytics
4:50 - Visualize the output
7:29 - Writing the SQL Query
9:03 - Where's the error?
Follow Jay on:
Twitter: / datasciencejay
LinkedIn: / jay-feng-ab66b049

[@iqjayfeng]

If you enjoyed discussing this data analytics question with us, explore more on our website! Button A/B Test: www.interviewquery.com/questions/button-ab-test? Causal Email Journey: www.interviewquery.com/questions/causal-email-journey? Rolling Bank Transactions: www.interviewquery.com/questions/rolling-bank-transactions?

[@iqjayfeng]

Thanks for watching! If you're interested in learning more about Data Analytics, then check out our newly launched Data Analytics course! The course is designed to help you learn everything you need to know about working from basic concepts to more advanced techniques. www.interviewquery.com/learning-paths/data-analytics

[@aditi3601]

These case studies videos are helping me alot. Please come up with more case studies.

[@filippoperissinotto8795]

Hey Jay, very useful video, thanks for sharing! If you had a take-from-home assignment like this, do you think it would be useful to run any more advanced statistical analysis? Maybe analyzing the population means with a t-test? Cheers

[@thefirstcindy]

Interesting! Good information to have

[@isabellaxu5572]

Thanks Jay, great example for a SQL + case study practice! I'm a beginner and still learning, can you please explain a little bit more about the normalization and casual inference you mentioned at 6:59? Or just pointing me to some material will be appreciated.

[@iqjayfeng]

Nothing on deck but we’re coming out with a data analytics course soon going over it!

[@dakshmukhra]

Great video ! Really in-depth

[@iqjayfeng]

Much appreciated!

[@LinhNguyen-cb5on]

Hi Jay! Why does the output still show 0 under the at_least_once_call_connected column? I thought we filtered that out when we call MAX in the sub query. Please advise. Thanks

[@BonniePerez]

Amazing resource! :) Well don!! TY! Keep going! :)

[@kabalayouri419]

Hey ! I’m new to this, but I wanted to ask why double joining on the same tables ?

[@CaribouDataScience]

Good stuff. Would this type of case study be something that entry-level analysts would be expected to solve?

[@iqjayfeng]

Probably not but it would make you the top of the pack if you could

[@IsonTin31]

I'm trying to get into Data Analytics. It seems really interesting. I have no background or experience with this line of job. Gotta learn the basics then maybe come back to this video. XD But overall, great video. :)

[@subhadrakumariram]

can you also solve data analytic problem wrt ecommerce example number of product vs revenue trend over number of years

[@melikakhamisi6816]

Hi Jay, why did you use Max aggregate function for "at least once call connected"?

[@iqjayfeng]

When you run MAX on only values that are 1 or 0, you’re basically setting a binary outcome for if they at least connected once. For example if I have nine 0s and one 1, then the MAX function will return 1 showing that there was at least one call!

[@nonoisephotography5809]

and I will find my way back there too!

[@shobhitmathur6262]

Hi Jay, This was very helpful! I was using this to prep for an interview and wanted to understand your approach here. Why did you left join marketplace into chats? Wouldn't that exclude any transactions that did not bother with the chat function ? If we do marketlace left join chats, then avg(count (m.id)/count(c.id)) as score, case 1 if call connected is 1 else 0 as used_chat group by used_chat would also work right?

[@iqjayfeng]

LEFT JOIN allows us to all transactions but not all the chats

[@user-gy7uu9gt8n]

I think the chat function is for customer who have product problems. So we only focus on those who attempt to make calls and whether successfully connected calls lead to real purchase. Yes, the query Jay provided would exclude transactions that did not use the chat function, but it does not matter. They just don't need it.

[@ASO-xh5vu]

Hell yeah I think it’s time we do it in the kitchen. I totally and sincerely agree with you

[@Mr.Arshad-oc1fd]

Great work 👏 But in this case study you didn't show numerical data as in questions

[@BrianBin]

I rarely see case studies good vidoe

[@iqjayfeng]

Thanks!

[@ChineseSe7en]

This too advance for me, what should I do?

[@niiniiy9940]

00:30

[@BruceNJeffAreMyFlies]

Lmao I don't know what's wrong with my thinking, but I thought 'success' was just whether or not people used it... 'Well just compare the number of buyers to the number of people who made calls'... I feel too dumb for this field now...

[@iqjayfeng]

Lol that’s not dumb - it’s just a common fallacy that you will now completely see through now when it comes up again!

[@sridharpadhy120]

teach you a lot of the fundantals of soft. You don't even need to sit through the entire tNice tutorialng to get a good grasp on the basics.

[@azabeish2942]

Too much water then solving. Too much talk.