Good is not almost as good as great
I arrive at Westchester Toyota and pass two or three salespeople loitering outside. Inside, there were two or three more, sitting in a line of chairs, waiting for the signal from the headmistress at the counter.
My guess is that even for a thriving brand like Toyota, most of these guys weren't paid so much. They were 'good' salespeople, lifers who showed up, did what they were told and closed a sale here and there.
It soon became clear that the salesperson who was assigned to me wasn't 'great'. The dealership had messed up: He had no record of my appointment, no file, no history of why I came. But he just punted. He made no effort to engage with me or look me in the eye or empathize with my frustration at the complete waste of time my call yesterday had been. He gave up after about ten seconds, bummed out that he had lost his place in line. So I left.
Driving home, I started to think about the discontinuity in the graph of salespeople. Discontinuities are interesting, because that's where you can see how a system works. In this case, it's obvious that a great salesperson is going to sell far, far more than a good one. Nine women working together can't have a baby in one month, and ten good salespeople still aren't going to close the account that a great one could. That's because it's not a linear scale. The great ones reach out. They work the phones when they're not first in line. They understand what a customer wants. They're not just better than good. They're playing a totally different game.
My best advice: Fire half your salesforce. Then, give the remainder, the top people, a big raise, and use the money left over to steal the best salespeole you can find from other industries or even from your competition. You'll end up with fewer salespeople. But all of them will be great.
And the good guys? Have them go work for the competition.