Data science projects face up to an 80-95% failure rate primarily due to poor data quality, lack of clear business goals, and organizational silos rather than the models themselves.
7 replies (most recent on top)
Very true, garbage in garbage out. Dan needs to reengineer that team.
OMG! We need QUALITY Data? What a shocker!! Good luck to Dan trying to build any AI platforms with literally garbage data. Due to a total lack of accountability or any ownership whatsoever....the data is a total disaster + train wreck!
Anil better get in 2nd gear soon
Mainly because there isn't any such thing as data "science"
I was just in eweb and the DS team needs to be re-engineered.
Previous regime had to many spokes in the hub. Now, a dysfunctional data science organization that is run by Google.
Tiger teams can solve this! We will need a staff of 50 incremental headcount to talk to your teams and repeat back to them what they tell us with pretty slides - at scale! Together we can build the future with organizational clarity and grace to delight our customers and deliver world class 5g to keep humans glued to their devices and make the world a better place. For you and for me and the entire human race.