5 Questions Every Data Scientist Should Hardcode into Their Brain

In this video, I break down 5 questions every data scientist should hardcode into their brain to make problem discovery second nature.
Read more: medium.com/towards-data-scien...
The Data Entrepreneurs
🎥 TH-cam: / @thedataentrepreneurs
📰 Medium: / the-data
📅 Events: lu.ma/tde
👉 Discord: / discord
--
Book a call: calendly.com/shawhintalebi
Homepage: shawhintalebi.com/
Socials
/ shawhin
/ shawhintalebi
/ shawhint
/ shawhintalebi
Support ❤️
www.buymeacoffee.com/shawhint
Introduction - 0:00
Background - 0:46
5 Problem Discovery Questions - 1:55
1) What problem are you trying to solve? - 2:30
2) Why..? - 3:23
3) What's your dream outcome? - 4:30
4) What have you tried so far? - 5:05
5) Why me? - 5:45
3 Key Takeaways - 6:20

[@ShawhinTalebi]

📰Read more: medium.com/towards-data-science/5-questions-every-data-scientist-should-hardcode-into-their-brain-3948e215750f?sk=37b76f4d1a9d1784932c36f84d2b0008

[@kenchang3456]

Thanks for sharing Shaw. If I could add, as you're talking to your client, elicit what their view is of a solution or solutions, after all they're the SME. The caveat is that what they see is often bounded by the tasks they are involved with and may not see any effects changes they are asking for, have on the business process(es) they are a part of, either upstream or downstream which is your value add.

[@ShawhinTalebi]

Thanks for sharing that insight! I like that. Pieces of that might start to come through when asked what their dream outcome is.
How would you go about sussing this out?

[@kenchang3456]

@@ShawhinTalebi there are many times when I'm talking to users, you can dig through and find out what data is theirs, meaning what data is generated by their group or task and what did flow in what data if any flows out of their task. If they are being "done to" by inbound data then you'll certainly hear about it and if they have not had any success working with upstream tasks then they're probably doing incurring overhead trying to compensate for dirty or incomplete data coming from upstream. Likewise, when crafting their solution for their task, they may not be aware that what data is being passed out of their task may cause changes downstream. And if those changes are not planned for then you end up with a process that just grinds. The worst case scenario is having the user walk the business's process to get what they need done because the business doesn't have a well designed business process and that includes spotty or bad communication between departments. Simple example is I call my doctor for an appointment with a specialist and the make me call the specialist directly without offering to transfer me or at minimum give me a phone number. Common sense stuff that should be part of a process but isn't. Looking forward to more of your videos. You're really good and you have a great mind.

[@csmac3144a]

I've been in IT consulting for 35 years and I haven't heard a better summary of how to initiate a successful client relationship.

[@ShawhinTalebi]

Wow thanks for kind feedback. Any questions or notes you'd add to the list based on your experience?

[@GetPaidToLivePodcast]

You're at 10k! Congratulations Shaw, you're killing it!🤩

[@ShawhinTalebi]

Thank you 🙏

[@ifycadeau]

Soo good thanks for sharing!❤

[@ShawhinTalebi]

My pleasure, hope the helps!

[@anneest]

Thanks for sharing, indeed with experience these questions appeared more natural to ask eventually. And I agree with 'not reinventing' the wheel, and ask the customer what they tried already and why it failed or partly succeeded only. This is really good 👍

[@ShawhinTalebi]

Thanks, glad you liked it :)