When one starts a PhD, should it be based upon interest or tractability of the project? If a project fails is switching advisors a reasonable option within the same department?

I have a couple options for PhD, one is an area I am more interested in but it's a more open ended problem with potentially less straightforward solutions. It also gives a background in more areas I am interested in learning and more industry applicable(focused on making novel system which works).

The other I don't think I want to work on for 4 years but probably will be more straightforward, it's also less useful to industry(I think) and more theoretical in nature. It is a continuation(mostly) of what I already know.

I'm not sure what to do, I have no interest in staying in academics so a PhD which is more focused on an applied topic which could be used to spin up a company is a lot more interesting to me but I obviously don't want it to fail.

  • 1
    What is your motivation for doing a PhD? Aug 15, 2020 at 19:30
  • 3
    It sounds like you should just do a start-up if you're not that interested in doing research? If your main goal is to do a start-up then why wait 5 years to do it? You should also be aware that earning PhD does not imply you will make a high wage, in fact, in the long run, economically you might be better off just not doing a PhD. Aug 15, 2020 at 19:49
  • I don't have all the skills for a startup and the PhD will cover additional classes and give me more hands on experience with a few systems I wouldn't normally have access to. I'm interested in research but I want it to be transferable to a physical product. Aug 15, 2020 at 19:51
  • So why is the more straightforward/theoretical topic even an option? It seems entirely disconnected from your motivation. Aug 16, 2020 at 21:04
  • 1
    The more theoretical project is more straightforward? That's a new one.
    – JeffE
    Aug 17, 2020 at 1:47

2 Answers 2


You give a false dichotomy, because you should be doing a project that you are interested in and one that can be completed. There's no point in choosing a topic you hate working on for 5 years but is solvable just as there's no point in choosing a topic you're passionate about but not even those who have studied it for decades could solve (possible, but highly unlikely). Therefor, the answer to your question is to choose a solvable problem that you are interested in.

However, if you're interested in living in a binary world, then choose a project that is tractable and you hate. If you are concerned about getting a PhD, then it's better to complete a project and receive the degree than to work hard on a problem and receive no degree.

  • Hmm I see your point, I think the real question is - suppose I work for 2 years on a problem and don't make much progress but can exit with a second MS. Should this be viewed as a failure or not? If I am getting something from the PhD process itself, independent of the degree it seems that should be factored in. Aug 15, 2020 at 19:57
  • Also what about changing advisors if it goes bad? Doable? Aug 15, 2020 at 19:59
  • 1
    What does it matter to us if you gaining a second masters is considered to be a failure or not? It’s really up to you to define what you consider to be success and what you don’t. Second, changing advisors is a serious thing you shouldn’t take lightly. What would be the reason you would be switching advisors? Furthermore, how do you know you’ll be able to switch to the advisor you want when you come to the decision to switch? And what would you define “goes bad” to be? Aug 15, 2020 at 20:38
  • 1
    It’s not uncommon, but that doesn’t take away from the seriousness of the implications in doing so. For example, you may end up on a new project you don’t like, you may end up with a prof who doesn’t tailor to your style, you may end up burning more time than you’d like, you may be in a position where what you think will happen doesn’t pan out the way you’d like, you may end up developing skills you don’t want to be developing. I say all of this because it sounds like you haven’t considered the implications too carefully of switching advisors. Aug 15, 2020 at 20:55
  • 1
    @GrayLiterature Oh, no, I've found lots of interesting open problems to work on. I made progress on enough of them to be successful, but most of them are still open and interesting.
    – JeffE
    Aug 18, 2020 at 13:14

I think that a good PhD adviser will guide you to a project that is interesting and doable. Look for potential thesis advisers and see what their past students have done.

Not the answer you're looking for? Browse other questions tagged .