The Sequence, Not the Sentence: Why Enterprise Deals Break Where You Are Not Looking

The final meeting went well. So what killed it?

I want to start with the most common conversation in enterprise sales, and the most useless one.

A deal that everybody expected to close does not close. There is a post-mortem. And in the post-mortem, the last call gets examined in detail, because the last call is the thing everyone attended and everyone remembers.

Someone says the pricing conversation could have been tighter. Someone else thinks the CFO was never really convinced. The rep says the champion went quiet at the end, which is true, and which explains nothing.

And then, usually, someone says the sentence that ends the discussion without resolving it: the final meeting went fine, honestly, I do not know what happened.

They are right. The final meeting did go fine. That is exactly the problem, and it is why the post-mortem is looking in the wrong place.

The thing nobody wants to say out loud

By the time you reach the meeting that appears to decide the deal, the deal has usually already been decided.

Not in that room. Not in that conversation. Weeks earlier, in a series of small moments that nobody flagged at the time, because none of them looked like the moment anything was at stake.

This is the part I find genuinely hard to get people to sit with, because it runs against how deals feel from the inside. Deals feel like they build toward a climax. There is a final presentation, a negotiation, a decision. It has the shape of a story, and stories have turning points.

But the turning point is almost never where the drama is. The dramatic meeting is where a deal that was already lost finds out.

Where it actually breaks

Let me be specific, because "it broke earlier" is easy to say and useless unless you can name where.

It breaks at qualification. A first call goes well. The prospect is friendly, engaged, asks good questions. The rep reads warmth as intent, marks it qualified, and moves on. Nobody asks whether there is funded, identified, urgent pain, or whether this is a curious person having an interesting conversation on a Tuesday. Three months later a deal that was never real disappears, and everyone treats it as a loss rather than what it was, which is a qualification that never happened.

It breaks at stakeholder mapping. The rep builds a genuine relationship with the champion. The champion is enthusiastic, responsive, helpful. What nobody tests is whether the champion has any real influence with the person who signs. Enthusiasm and authority are different things and they are very easy to confuse, particularly when someone is being nice to you. The deal dies the moment it reaches a person the rep has never met.

It breaks in the committee. The rep prepares thoroughly for the people in the room. The decision, meanwhile, gets made in a room the rep is not in, by six to ten people, several of whom the rep has never spoken to, each holding a different definition of value, one of whom quietly prefers the incumbent and will say so when the rep is not there to answer.

It breaks at the multi-threading that never happened. One relationship, carefully maintained. Then in week eleven that person goes quiet, changes role, or simply stops replying. And there is nothing else in the account. No second thread, no parallel relationship, no way back in. The deal does not get rejected. It just stops existing.

Look at those four. Not one of them happens in a meeting anyone would have flagged as high-stakes. Not one would show up in a recording as a bad call. They are all quiet. They are all early. And they are all decisions the rep made, or failed to make, while everything still looked completely fine.

The sequence, not the sentence

Once you see this, something uncomfortable follows about where sales training and sales tooling actually point.

Almost everything is aimed at the visible moment. Pitch practice. Objection handling. Demo certification. Roleplay the hard call. Conversation intelligence records the meeting and analyses what was said in it.

I want to be fair here, because I think conversation intelligence is genuinely valuable and I am not interested in pretending otherwise. Knowing what actually happened in a call, rather than what a rep remembers happening, is real information and most organisations were flying blind without it.

But notice what all of it has in common. It is pointed at the conversation. And if the conversation is where deals get revealed rather than where they get decided, then the entire category is instrumented at the end of the story.

You are recording the crash with excellent equipment. You are learning nothing about the decision three stages back that made the crash inevitable. That is the sequence, not the sentence.

What this means for how you train

Here is where it stops being an observation and starts being a practical problem.

If deals break in the sequence rather than the sentence, then training a rep to handle a single conversation well, however realistically, is training the wrong unit.

A rep can be excellent in the room and still lose consistently, because being good in the room is a late-stage skill and the errors were early-stage judgment. The question that decided the deal was not how do I handle this objection. It was should this account have been in my pipeline at all, asked eleven weeks earlier, and never asked at all.

That judgment is not a communication skill. It is pattern recognition across a sequence. It is knowing, from having been in this shape before, that a champion who agrees with everything is usually a champion with no power. That a friendly first call with no budget conversation is a deal that will die politely in Q3. That when one relationship is your only relationship, you do not have an account, you have a coin flip.

Nobody learns that from a slide. And nobody learns it from rehearsing a single conversation, no matter how good the simulation of that conversation is, because the thing that needs practising is not the exchange. It is the sequence of decisions around it.

Which is also why nobody measured it

There is a measurement problem sitting underneath all of this, and I think it is downstream of the diagnosis rather than the headline.

Enablement reports completion rates. Ninety-four percent finished the methodology course. Four point six out of five satisfaction. Green across the board. And in the QBR, when someone asks why the deal was lost, the two facts sit next to each other with no relationship at all.

It is easy to treat that as a failure of rigour. It is not. Completion is what the tooling could count. Measuring whether a rep can actually navigate a sequence of judgment calls under pressure required a human on the other side of the table, running the scenario, watching the decisions. A manager, or a senior rep playing the difficult CFO. That is the highest-quality sales training that exists and it is also the least scalable thing in the entire commercial organisation, because it consumes the most expensive and most contested hours you have.

So it happened in bursts. Kickoff. Onboarding week. The top decile who earned coaching. And for everyone else, across the other forty-eight weeks, there was content, because content was the only thing that scaled.

Nobody chose completion because they believed it measured readiness. They chose it because it was the only thing available that could be counted at scale. That constraint was real, and it is the constraint that recently lifted.

What you would see instead

Let me make this concrete, because "measure judgment" is easy to say and the whole difficulty is in what the artifact actually is.

Imagine, instead of a completion report, a manager opens a view of their team the week before a critical quarter. Not satisfaction scores. Not who finished what. A readiness picture, resolved per rep, per stage of a deal.

They can see that one rep qualifies rigorously but stops after building a single relationship, every time, in every account. They can see that another maps stakeholders well and then concedes margin on the buyer's first push, before real pressure has even arrived. They can see that a new hire, four weeks in, is already reading committee dynamics better than someone who has carried quota for two years, which is a promotion signal that would otherwise have taken a year to surface.

None of that comes from a survey. It comes from having watched each rep move through realistic deal sequences, under pressure, with their actual decisions scored against the framework the company already runs. Not a new methodology layered on top. The organisation's existing MEDDPICC or SPICED, finally made observable at the level of the individual decision.

And the coaching that follows changes in kind. Not "the team should work on qualification," which is what an average produces and what nobody acts on. Instead: these three reps, this specific pattern, this exact point in the deal where they lose control, and here is the movement six weeks later. Coaching stops being a broadcast and becomes a scalpel, aimed by evidence instead of by the manager's memory of the last deal review.

The conversation this changes

There is a practical payoff here beyond diagnosis, and it is the reason I think this matters more than a philosophical argument about metrics.

Enablement leaders lose budget arguments not because their programmes are weak, but because their evidence is denominated in a currency the CFO does not use. "Ninety-four percent completion" is not a weak answer. It is an answer to a question nobody asked, and everyone in the room hears the mismatch.

Evidence of readiness changes the currency entirely. Not "the team completed the enablement" but "eleven of our forty reps consistently advance deals without testing whether the champion has real authority. We know which eleven. We coached specifically for it. Here is the movement six weeks later, and here is what that is worth in pipeline that did not evaporate in Q3."

That is a sentence a CFO can act on. It is also the sentence that turns enablement from a cost centre into a revenue function, which is an argument the profession has been trying to win for twenty years with the wrong ammunition. You do not win it with better slides about completion. You win it by changing what you are able to see.

Where I sit, disclosed

I should be transparent, because I have a commercial interest in the argument I just made.

I build a company in this space. We make simulations where reps rehearse an entire deal cycle, territory planning through negotiation, against an AI buyer that pushes back, and every decision they make is scored against the framework the company already runs. The bet is exactly the argument above: that the sequence is the unit that matters, not the conversation.

You should weigh my view accordingly. But I would separate two claims that I hold with very different confidence.

The first: that enterprise deals break early and quietly rather than late and dramatically, that the industry instruments the loud end, that this makes the failure invisible until it is expensive, and that the whole gap stayed unaddressed because real sequence practice was never economically scalable. I would defend every part of that in a room full of people trying to take it apart. None of it depends on my product and most of it is older than my company.

The second: that any particular tool, including mine, is the right response to it. That is a bet, and bets deserve scepticism, including from the person making them.

The question I would actually like answered

I have now had a version of this conversation with enablement leaders and sales managers across quite a few companies, and one pattern keeps repeating. When I ask where their deals break, almost nobody says the final meeting. They say qualification, or the committee, or a champion who turned out not to matter.

And then, almost universally, their training is pointed at the final meeting anyway.

So the question I find more useful than "how do we improve our sales training" is a narrower one, and I would genuinely like to hear the answers:

Think about the last significant deal you lost. Not where it ended. Where did it actually turn?

I suspect most people reading this already know. I suspect very few of them have any way to prove it, coach for it, or catch it happening again next quarter.

Frequently asked questions