RevOp  /  Post-mortems  /  07
Post-mortem 07

The manager who coached deals that were never going to close.

Module 18 · Coaching & enablement Stage · repeatable and up Instrument · The Execution Score

This is a better post-mortem than "the manager was too busy to coach", because the coaching was happening. It was consistent, it was skilled, and it produced nothing.

What the week actually contained

Taking rep questions. Support. Shipping help. Deal coaching, fairly consistently. One or two call reviews a week, chosen at random. A very light layer of framework, and a lot of moving between whatever a rep needed to keep going.

None of it was slacking, and none of it moved the things that drive performance.

What was happeningWhy it produced nothing
Deal coaching, consistentlyOn deals that were never going to close. Effort correctly applied to the wrong object
One or two call reviews a weekUnscored. An informal conversation about the call
Rep questions, support, shippingReal, urgent, and endless
The mechanism

Deal coaching on a pipeline that is half fiction is not coaching. It is rehearsing fiction.

Every session was sincere and well run. The questions were good. The advice was sound. And the deal at the centre of it did not exist in any form that could be closed.

The part that makes this hard to catch is that it is invisible from outside. A week of coaching fiction and a week of coaching real deals look identical on a calendar, take the same hours, and generate the same amount of visible management activity.

You cannot audit this by looking at what a manager spends their time on. Only at what the time was spent on.

The two findings, and managers will argue with the second one

An unscored call review is a conversation. A scored one is a datum.

Without a score there is no trend, no comparison between sellers, and no way to tell whether the rep improved or you simply had a nicer chat. Two reviews a week for a year produces roughly a hundred conversations and zero evidence.

So the missing piece was never the reviews. It was the scoring. That distinction matters, because "do more call reviews" is the advice everyone gives and it would have changed nothing here.

And the reactive work is not trivial work

The standard framing is that the work which compounds is the first thing a reactive week eats. True, and it undersells the trap badly.

The thing doing the eating is legitimate, urgent, and coming from people you are responsible for. It is your own rep, stuck on a shipment, asking for help. Nobody ignores that, and nobody should.

Which is why the fix is structural rather than behavioural.

You cannot will your way past your own team needing you. You can only give the compounding work to someone whose sole job it is.

That is the part worth arguing with, because it removes the comfortable answer. There is no discipline solution. There is no better calendar. A manager who is available to their team is doing their job, and the compounding work has to live somewhere else.

What replaced it

An enablement manager, with three duties. The third one was not on anybody's job description when we started.

Call reviewsRun to a schedule rather than at random
ScoringThe thing that was actually missing. A number, not a chat
Removing phantom pipelineDeleting fiction from the forecast, which turned out to be the first job

Head coach runs the game. Assistant coach develops the players. And in this case the assistant coach's opening task was not development at all. It was clearing out the deals that were making everyone else's work meaningless.

Which is the uncomfortable sequencing lesson underneath all of it. Coaching cannot be fixed before the pipeline is real, because until then you are improving how well someone rehearses something that is not going to happen.

How many of your call reviews produced a number last month?

Not how many happened. How many produced something you could put in a row next to last month's. If the answer is none, you do not have a coaching problem yet, you have a measurement problem, and they need fixing in that order. The model is free and it starts where this does, with how much of the pipeline is real.

Run the model, free No account. Your numbers stay in the link.