throwaways66722
throwaways66722 on Hacker News
I see little difference. Sometimes it's an EM + 1 or 2 data scientists. Sometimes its an EM + 1 data scientist + 1 or 2 BA/Associates. We solve similar problems, its just the data science people tend to solve it with data and the traditional consultants solve it with interviews and whatever. That's not a swipe at them either. There's a significant quantity of problems that can be easily uncovered by talking to someone who knows the answer but can't get the business to acknowledge it OR by spending a week churning through data to discover it yourself. You might feel better with the hard fact base in hand, but doing those interviews and using the traditional consultant skill set tends to make you far more targeted and efficient in scoping and planning.
"QB" I think operates a little differently, whereby they get the data and tend to do more stuff on their own with dedicated modelers, data engineers, and designers. More "I am the expert here, leave me be". (Also not intended to sound like a swipe. That's a reasonable point of view sometimes).
1) You don't need to work your way up the approval chain. Your idea is already going to people with the authority to act on it.
2) You have significant ability to just call people and get them to share their data and domain knowledge
3) You have little risk of long term problems from saying something that someone doesn't like, so long as its not the person who hired you.
4) The pace is probably very very fast relative to the speed of most large orgs. Probably not sustainably fast tbh. It's fine because it's several weeks of hard focus and then a bit of a break.
If you feel like you're doing fine and are answering important questions then you probably are doing just fine and your company's data culture is healthy. IMO people who build analyses that few people want or need are acutely aware of it.
I would take a different perspective. The issue is not 23 year olds being better than others. It's the fact that the 23 year old from McKinsey comes in with incredible political capital equivalent to whoever hired him at the top, and that enables him to be multiple times more productive than someone else. Most organizations are... slow.
Most data scientists... don't know what to do. And most clients... don't know what to do with them. So they do little data science projects and build dashboards that few people use. Coming out of McKinsey you can quite confidently say you've learned how to identify top problems, organize a project around it, and solve it in a fast timeline with realistic solutions. That's pretty much gold. It's really useful to be able to say you've worked with c-suite leaders.
Finally, it's probably a function of just fast growth internally. You gain responsibilities much faster at McKinsey than at most most companies. The company invests in your development more. And when you leave, you get weeks of paid time off to search for a new opportunity.
*current mckinsey data scientist who got an offer from both Gamma and McKinsey, and turned down Gamma despite being offered a higher salary.