"What do you want to do with your life?"
I don't want to claim to know the best, objective answer to the question, but I have an approach. So when we're saddled with a difficult question, one of the things we (people) do is wrongly try to tackle the whole problem at the same time. As complexity increases, it gets harder and harder to do this. The workaround is decompose the main issue into smaller, more workable problems. So, first you have to know what to break the big issue into, and in part this is why people go to therapy. As for the question what do I want to do with my life, I would decompose life first into the foreseeable future, and then decompose the foreseeable future into a reasonable timeline.
Okay so for me, good milestones are what do I want to accomplish in 1yr, 3yrs and 5yrs. You can also think about 1/5/10 years, whatever you can conceptualize. Then think about what you'd like to see done by those milestones. It doesn't have to be what actually happens, just what you'd like to see happen. And so then when the next person asks you what you want to do with your life, you can firmly tell them how you think your next several years should progress, since no one can see 30 years into the future. Even nailing down a year is more helpful than having nothing planned at all.
How do I convey in interviews that despite any specific career goals I am a diligent worker?
The main hurdle, and as a science graduate you'd probably understand, but if I hire someone to do biological research and I look over their experience and realize they've never used a pipet or cultured cells, I'm most likely to hire someone who has the minimum bench experience over the one who has none. There is a chasm between diligence and experience that you can overcome with hard work, just not in a timeframe any reasonable employer might require. You have positions where the employer expects you will need to be developed, and others where you'll need to hit the ground running. Marketing manager feels like the latter case.
If you think your skillset is worth the lack of domain experience, you need to sell that in your interviews. No generalities, be specific and sell why your skillset will allow you to succeed. You might even find a gap in your understanding that will hold you back in interviews if you try to convince yourself first before you try to convince the interviewer. e.g. "I'm really good with Microsoft Access" versus "I lead the development of an Access database that got our receiving department on track with tracking such-and-such. And this is transferable in that it'll help me ... [something to do with role you're interviewing for]."