TomHunter1
commented
6 years ago 1) It fluctuates highly and I think the idea of a follow up touch point is a great add. 2) Yes, Meyers Briggs was going to be cited as one of my self-criticism of my own idea. Apparently, this happens alot. 3) Good point 4) Fair point. I'll go into detail about data sources and applications for use cases. 5) I agree with this point. This was part of my planned build out, so you read my mind. 6) Agree- fair point will change
I think this is an interesting idea for driving accountability, which might be the most fundamental aspect of goal-setting. Making it social is a good idea because it helps achieve that purpose, plus making everything social is marketable nowadays. This is trying to tackle a difficult problem, which isn't bad, but the results will probably have a decent level of uncertainty and variability. I like decision trees as much as the next guy, but I'm skeptical that that the probabilities will have much precision. I'd love for you to prove me wrong.
I have a few small suggestions, but I think overall if you flesh it out with a few details, it is on the right track:
1) Rather than pick a rigid personality measure and stick with it, you may just want to check in with the user every 6 to 12 months and re-test. I am not an expert, but I think the traditional view on personality is that it is fairly steady, but some newer schools of thought regard the self as more fluid.
2) Meyers-Briggs specifically has been criticized quite a bit. Although popular, I might not choose that one. Some others I've tried are Strong Interest Inventory and StrengthsQuest (just as food for thought).
3) I actually think the ethics of this are pretty straight-forward and don't warrant too much discussion. You're helping users achieve user-defined goals. The only possibility I see for gray area is some sort of "suggested goals" section. Subtle persuasion is a really invasive aspect of modern life. Users specifying unethical goals is not worth mentioning in my opinion - seems solvable with a "Flag as Inappropriate" button.
4) I can't tell exactly where the data is coming from for your NLP. Is it in-app or from public message boards or from books written by Duckworth or Bandura? Even if you have a great parsing algorithm, what makes the sample of data from which you derive your algorithms any better than the strangers' suggestions you dissuade from using in paragraph one? Or maybe I am misunderstanding and the public information is generally a good basket of ideas, but just need sorting as to what applies to whom.
5) I'd actually like to see a little more literature included on whether or to what extent personality differences affect optimal goal-setting strategies. Are they really that individualized, or are there universal strategies such as nudging, accountability, and other tricks that work on everyone? Is this supported by expert opinions or is it your own hypothesis? I'm just genuinely curious.
6) Minor thing: I thought this was supposed to be more of an idea pitch, but the past-tense at the end makes me think it's already completed or in work.