3 ms·
I wouldn't do the process, just the feedback. Build a value add around the feedback, such as interview critique/coaching services, training, etc... Make it abou
by jeffmould 11y ago
I wouldn't do the process, just the feedback. Build a value add around the feedback, such as interview critique/coaching services, training, etc... Make it about optimizing the interview process to make smarter hires.
- applygo 11y agoThere are plenty of services that offer interview prep advice and numerous independent consulting services in feedback. These don't scale and not what I want to do. I'm doing more research and think there is something to "collective intelligence" or wisdom of the crowd. I'm not going to argue if a crowd is wise or not, but data show that the crowd gets it right within a fraction of the right answer. What I'm proposing is a type of feedback, but one that can scale globally. The crowd should be able to predict closely what a person is like to work with, or so I think. This addresses the extreme cases we've discussed on the thread here. I'm still researching this - can crowds predict with the same accuracy as it is at arriving at an already known answer like how many jelly beans are in a jar etc. Here is some examples of what I'm talking about: http://youtu.be/iOucwX7Z1HU http://youtu.be/iOucwX7Z1HU http://arxiv.org/ftp/arxiv/papers/1406/1406.7563.pdf http://arxiv.org/ftp/arxiv/papers/1406/1406.7563.pdf http://m.sciencemag.org/content/330/6004/686.abstract?view=abstract&uritype=cgi http://m.sciencemag.org/content/330/6004/686.abstract?view=a...