I've written a lot about orchestration. Why it works. What has to exist underneath it. What it pays. What it can't do on its own.
I've never written down how to actually do it.
It's three moves. It takes about five minutes. And the middle one is the only reason any of it works, which is why almost nobody does it.
What You're Actually Orchestrating
Start here, because everything else follows from it and most people get it wrong.
You are not orchestrating the sale. You're not getting a commitment to buy, or a budget approval, or a verbal yes on a number.
You're orchestrating the conversation. That's all. And that's why it works.
Nobody commits to a purchase 90 days out on a call. Plenty of people will commit to talking about one, especially when the person asking has just told them not to buy yet.
One: Introduce
Name the thing.
The customer has told you what they're trying to do. You say the best way to do that is the thing you sell. Be specific about which thing, and connect it to what they just said they want, not to your product catalogue.
This part is obvious and it's where most people stop. Stopping here isn't orchestration, it's mentioning something. Mentioning is better than silence and it isn't the play.
Two: Deny
Now take it away.
They're not ready for it. Say so, and say exactly what makes you say so. Not a vague "let's revisit later." The specific thing that has to be true first, and isn't yet.
You just told a customer you have something to sell them, and then told them not to buy it.
Think about the last time anyone did that to you in a commercial setting. Most people have never had it happen once.
That's the whole lever. Everything the framework does to trust, it does here. Skip this move and you've made a soft pitch with a delay on it, which the customer will read exactly as what it is.
And it isn't a technique. It's true. If they can't use the thing yet, selling it to them now produces an item sitting in their account doing nothing but reminding them it isn't working. Denying them is the correct call on the merits. The trust is a side effect of being right.
Three: Agree
Then get the smallest possible commitment.
When you reach that point, we'll talk about adding it. Is that fair?
They say yes. In many years of listening to these conversations I've never once heard anything else, because it's close to the most fair thing anyone gets said to them in a commercial relationship.
Notice how small the commitment is. They haven't agreed to buy anything. They haven't agreed to want it. They've agreed that a conversation should happen at a moment you both just defined.
That's enough. People work hard to stay consistent with commitments they've made out loud, and this one costs them nothing to make and sits in their head for months.
The Template
The best way to do what you're describing is [the thing].
You're not ready for it right now, because [the specific condition that isn't met yet].
Once you've [reached that point], and on your current pace that looks like about [timeframe], we'll talk about adding it to your account. Is that fair?
That's it. That's the five minutes.
Six Things That Get Called Orchestration and Aren't
The word has spread faster than the practice, so it's worth marking the boundary.
Mentioning something without withholding it. "We also have this, let me know if you're interested." That's move one on its own. It's better than silence and it produces none of the trust, because nothing was taken away.
A promise with no condition attached. "Let's revisit in Q3" names no milestone, so nothing can ever trigger it. The customer hears a brush-off, correctly.
Anything keyed to the calendar. Months since signup, fiscal quarter, the renewal date. That's your clock. Orchestration runs on theirs.
Deferring because it felt awkward. The deny move is a stated judgment about readiness with a named condition attached. Avoidance dressed as patience has no condition, no timeframe and no agreement, and it's the version most likely to be mistaken for restraint.
Naming a setup milestone instead of a readiness one. "When your onboarding finishes" or "once the integration is live" tells the customer when they can start. A readiness milestone tells them what they'll have achieved that makes the next thing worth having. A starting line is not an achievement, and this is the most common false positive by a distance.
Raising it at renewal, even if you mentioned it earlier. Renewal is a contract date with no relationship to their progress, so whatever you said in month three, the conversation is now happening inside an inventory review.
The test for all six is the same. Was a specific thing named, withheld against a stated condition, and agreed to for a moment you both defined? Two out of three isn't a partial score. It's a different move.
Specificity Is the Whole Skill
The framework fails when it's vague, and it fails in a specific way: a vague version sounds like a brush-off.
"We'll circle back when the time is right" tells the customer nothing and commits you to nothing. Name the milestone. Name what has to be true. Then, separately, tell them roughly when they'll get there at the pace they're currently moving.
Those two things are doing different jobs. The milestone is the trigger, because that's what actually decides readiness. The timeframe is only there so they can plan, budget, and get whatever internal approval this will need. Ready and able are different conditions, and the second one needs lead time you're now giving them.
Being that specific also does something to how you're heard. A person who can name the next milestone, say what has to be true, and estimate the timing knows what they're talking about. Vagueness costs you that, every time.
When Not to Orchestrate
There's a point where it's too late to orchestrate and you should just do the work.
If the customer is close enough to the milestone that procurement, implementation, and setup would run past it, orchestrating wastes the window. Introduce it properly and start the process now, so the capability is live when they arrive.
Roughly 60 days out is where I'd draw it, and that number is yours to set rather than mine. Be careful with it, though, because it looks like a calendar rule and this whole approach argues against calendar rules.
It isn't the same clock. Milestones run on the customer's progress. The 60 days runs on your fulfilment lead time, which is a property of your business and has nothing to do with where they are. Long procurement and a heavy implementation push that number out. A one-click add-on pulls it in.
Two Things Have to Be True First
This framework is small and it sits on top of two things that aren't.
You have to be delivering what they already bought. Delivery failure precludes expansion, and running this on an account still waiting for what it purchased in March isn't orchestration, it's a fresh insult with a timeline attached.
You have to have something to orchestrate toward. The deny move requires a real thing that exists, has a price, and isn't included yet. If everything you sell went into the first deal, there's nothing to introduce and nothing to withhold. That's the dependency, and it's why this fails in companies that stuffed the initial sale.
It Isn't Only for Expansion
The same three moves work anywhere something should happen later.
Asking for a reference before you've earned one is the most common version of this mistake. Introduce it, deny it because they haven't got results yet, agree to come back when they have. Same for a feature they've already paid for and can't use until three other things are done.
Anything that should happen at a moment the customer hasn't reached yet is a candidate. Introduce it, take it away, agree on when.
The commercial version gets the attention because that's where the money is. The move is the same.
Lincoln Murphy formally named and popularized Customer Success starting in 2010 and has spent 15 years connecting it to expansion revenue and commercial outcomes. Read The Premise.