The first data role after school is always a shock: nobody seems to care about the theoretical properties of the shiny models that you can build. Everybody around you talks about things you have never heard before: data pipelines, KPIs and business impact. This is the big gap between academia and industry. As in many situations there is a problem of incentives: academia has its own research goals and many academic programs don't have a mechanism to keep up with industry needs. But they still enroll plenty of students who will get their "Analytics" degree. At the same time, many companies prefer to look for experienced candidates because it takes too much effort to partner with universities and train fresh graduates. Of course there are exceptions* but as universities continue to evolve, I am hoping to see more interactions (especially at the graduate level) between the two worlds. *I am affiliated with Columbia University 's IEOR Department, which offers 5 Master programs with a mix of theoretical and applied courses, and opportunities to work on industry projects.
Can confirm I did an ostensibly industry focused masters in data science in 2012 and didn't learn a thing about data pipelines. In their defense though, it was 2012-- Airflow and dbt were both released two years later!
Data Scientist/Analytics Engineer/Analyst with deep experience in mortgage credit | SQL, dbt, Snowflake, Sigma, SAS, Dataiku, Python
2wOklahoma State's Masters of Business Analytics is an exception to this. I've been hiring their interns for years and find they arrive with business understanding and practical skills.