i'm not fond of commenting on hn-as-marketing-channel posts, even if it's within the bounds of the guidelines, but here goes...
> "I don't try to assess character because I don't think you can effectively."
> "My typical strategy is to just hire and fire quickly. I don't do interviews, and I just hire someone for a small job (5-10 hours) and see how they do."
to restate, you can't assess character in a few meetings/interviews, as there's just not enough data (it's well within the honeymoon period of any human relationship). humans are quite good at assessing character over the long term however. your "typical strategy" is employed, or at least should be, to mitigate the inability to assess character in the short run.
but, you didn't employ that strategy in your situation. fire fast would have been after they didn't deliver the first set of assets--you'd give them one more chance (with fair and direct warning), and after that, they should have been gone. instead, you kept at it for many more months. you failed to manage your own project, and that's really the bottom line learning here, not all the other stuff you wrote about. by the time you did fire them, you had enough data to assess their character and fired them based on that, rather than employing your fire-fast strategy.
that's not to try to condemn you in any way, as management is ambiguous and surprisingly complex (NP hard), but you left a gaping management hole that the agency filled with their own priorities and goals. i've been on both sides of this coin, and one of the unobvious inefficiencies of outsourcing is the need for twice the management (on each side). your solution to just hire a freelancer would work, not because it's a small project and you'd be "rightsizing", but because it'd make it obvious and necessary that you'd be actively managing the project.
> "I don't try to assess character because I don't think you can effectively."
> "My typical strategy is to just hire and fire quickly. I don't do interviews, and I just hire someone for a small job (5-10 hours) and see how they do."
to restate, you can't assess character in a few meetings/interviews, as there's just not enough data (it's well within the honeymoon period of any human relationship). humans are quite good at assessing character over the long term however. your "typical strategy" is employed, or at least should be, to mitigate the inability to assess character in the short run.
but, you didn't employ that strategy in your situation. fire fast would have been after they didn't deliver the first set of assets--you'd give them one more chance (with fair and direct warning), and after that, they should have been gone. instead, you kept at it for many more months. you failed to manage your own project, and that's really the bottom line learning here, not all the other stuff you wrote about. by the time you did fire them, you had enough data to assess their character and fired them based on that, rather than employing your fire-fast strategy.
that's not to try to condemn you in any way, as management is ambiguous and surprisingly complex (NP hard), but you left a gaping management hole that the agency filled with their own priorities and goals. i've been on both sides of this coin, and one of the unobvious inefficiencies of outsourcing is the need for twice the management (on each side). your solution to just hire a freelancer would work, not because it's a small project and you'd be "rightsizing", but because it'd make it obvious and necessary that you'd be actively managing the project.