Good tutorial. Thanks. For the 3rd query though, I think we could have added the 2nd argument (first_transaction_DT) to the first expression (Rank) itself. instead of creating 2 separate fields. -- rank() over (partition by user_id, payment_method order by payment_counts desc, first_transaction) payment_rnk
[Update] The founder was kind enough to reach out directly and resolve the issues, really appreciate it. I know they are a small team to get the product start and running, so do expect issues as it won't be perfect in the beginning, but also have patience and reach out to Dan if you have questions/suggestions. [original text] There are several mistakes in the solution across questions. I've submitted feedback but never heard back nor are the mistake corrected. This is a bit concerning. I would wait for the product to become more stable to subscribe or it would actually take you more time to find out why your solution does not work.
The question asks for product names with the highest price per category. If you just apply max, you will only get the price, not the corresponding product.
@@DataInterview that's not exactly true, if you include product name and id in the group by, it will function just as normal as their will be no ties to break, you can actually use group by in that sense
In the hard question part, some maild ids are left out... consider the example: maild_id trans_method trans_count minimum_date purchase_rank purchase_order email_id 1 credit_card 4 jan-10-2022 1 2 email_id 1 gifct_c 1 jan-11-2021 2 1 please provide a solution for one such uc....
