I think you substantially over-estimate the number of tickets which can actually be 'solved' like this, and dramatically under-estimate the cost it would take t

I think you substantially over-estimate the number of tickets which can actually be 'solved' like this, and dramatically under-estimate the cost it would take to solve the problem. I'd have a hard time believing that the large number of companies who all operate this model are getting the cost-benefit analysis wrong by such a large margin as you seem to suggest.