The deal you should never practice on.

Your rep is going to try the new thing for the first time on the biggest call of their quarter. That is not courage. That is a coin flip you scheduled.

We do this constantly and we call it coaching. You sit with a rep, you find the real problem, you tell them what to do differently, and then you send them out to attempt it, cold, in front of a buyer, in the one deal they cannot afford to lose. Then you are surprised when they revert to the old habit halfway through the call.

They did not forget. They never practiced.

Why do reps revert to the old habit in the deals that matter most?

Because pressure does not pull new behavior out of you. It pulls the most rehearsed behavior out of you. Under stress, people do not rise to the level of their intentions. They fall to the level of what they have actually done before, many times, without thinking.

Anders Ericsson spent a career on this. His finding was not that practice makes perfect. It was narrower and more useful: mere experience does not make you better. What makes you better is structured repetition with feedback, aimed just past what you can currently do. He put the whole idea in one line: “Living in a cave does not make you a geologist.”

That should land hard on any sales floor. A rep with ten years of experience has not practiced ten years. They have performed ten years. Those are not the same thing, and only one of them compounds.

Here is the other half of the math, and this is where it stops being philosophy. Single-deal exposure says that when one deal carries more than 40 percent of the quarter, it fails to close about 4 out of 5 times. Now read that next to the way we coach. We keep asking reps to debut new behavior in exactly the deal that is already the most likely to fall apart. We are stacking a first attempt on top of the single worst odds in the pipeline.

So separate the two rooms. Here is how.

1. Name the one behavior. Not five. Refine is not a makeover. It is one change at a time, because one is what a person can actually hold under pressure. Come out of a loss review or a ride-along with a single line: this quarter, you confirm the money before you build the proposal. That is it. If you hand a rep five improvements, you have handed them zero, and you will get the old habit back on Tuesday.
2. Give the behavior its first reps somewhere that does not count. This is the whole move. The first ten attempts at anything new belong in a low-stakes room. A one-on-one. A pipeline meeting. A drive between calls. Not the renewal that carries the region. Make it concrete. In your next one-on-one, do not ask “how are you going to handle the pricing conversation?” Ask this instead: “Say it to me. I am the CFO. Go.” Then let it be bad. It will be bad. That is the point, and that is the cheapest place in the world for it to be bad.
3. Coach the swing, not the scoreboard. When they run it, do not grade the outcome. Grade the behavior. Did they ask the question, or did they slide past it? A rep can execute the new behavior perfectly and still lose the deal, and if you only react to the result, you just taught them the new behavior does not work. Judge what they did, because what they did is the only part they control.
4. Hold through the friction weeks. Here is the part nobody warns you about. There is a stretch after a new behavior is installed and before it produces anything, where the rep is slower, more awkward, and less effective than they were with the old habit. That is normal. It is also exactly where most people quit and go back to what was comfortable. The friction weeks are the price of the change, and your job in them is not to add pressure. It is to be the person who says: this is supposed to feel worse right now, keep going, I have seen the other side of it. That is accountability delivered with empathy, and it is the highest form of respect you can pay someone you are trying to make better.

The spike and the system.

Thirty-plus years carrying a number, and a lot of years now coaching managers rather than just reps across a national team, and the pattern is not subtle. Every rep can pull off the hard call once. Give anyone a big enough moment and they will find something they did not know they had. That is a spike. It feels like a breakthrough and it changes nothing, because they cannot tell you what they did and they cannot do it again on demand.

The reps who compound got there a duller way. They picked one behavior, ran it badly in a room that did not count, ran it again, and kept running it through the weeks where it was not working yet. Then the day the big call came, they were not attempting anything. They were doing the thing they had already done fifty times. Isolate, Refine, Compound. See what you do. Change one thing. Let the years multiply it.

Rehearsal is what turns a good quarter into a good career. Performance just tells you what somebody already knew how to do.

The short version.

1. Under pressure, people do not do what they intend. They do what they have rehearsed. Nothing else is available to them.
2. Experience is not practice. Ericsson’s finding is that mere years of doing the work do not reliably improve performance; structured repetition with feedback does.
3. Never debut a new behavior in the deal that carries the quarter. Single-deal exposure already says that when one deal is more than 40 percent of the number, it fails about 4 out of 5 times. Do not stack a first attempt on top of that.
4. Refine one behavior at a time. Five improvements is the same as none.
5. Rehearse it in a room that does not count, then coach the swing, not the scoreboard.
6. Expect the friction weeks. New behavior gets worse before it gets better, and that is where most people quit.

Here is your homework: pick the one rep on your team who keeps reverting on the calls that matter, and find the behavior they slide past. Then give it its first ten reps somewhere that does not count, starting in your next one-on-one. Say it to me. I am the CFO. Go. Let it be bad there, so it does not have to be bad in front of the buyer.

This essay ran as edition 13 of The Frontline Sales Forecast, the weekly newsletter. One issue a week, published here on the Friday, then emailed and posted to LinkedIn the same morning. Same piece everywhere. Get the next one.