Meta Data Science interview questions

A recruiter reached out to me, we did an initial phone screen, I passed, and was told I would do a interview online with a member of the Data Scientist team. I was told by the recruiter that the interview would have three parts: a product/business sense section where you discuss FB products, how you would improve them, and what type of metrics you would use to gauge success; SQL coding to show basic competency in coding; and a section on statistics. I was told that the first section would be the most important, and that candidates who struggle on the SQL coding or statistics sections can continue through the process if they excel at the product section. As a result, I spent all my prep time reading about FB products and thinking about how to improve them and how to quantify success, and brushing up on statistics questions, like many of the ones you can find here on glassdoor. The actual interview was entirely SQL coding. I was not really mentally prepared to do live coding for someone else for 45 minutes, and feel like I was gaslighted by the recruiter in being told how I should spend my time preparing for the interview. I also had a lot of problems scheduling the interview; my first attempt to interview fell through because the recruiting team failed to notify my interviewer internally that he was supposed to interview me. (Which I find a bit funny for a social network company.) so we had to reschedule to a few weeks later. Overall: this was a negative experience, and I hope other candidates read this so they are not caught by surprise during an interview.