Only five to ten percent of your pipeline is ready to buy now. The other ninety percent isn’t lost. For most of business history, it was simply invisible.
We grew the agency I built by twenty to twenty-five percent a year, and every year we lost around ten percent of our clients. For a long time I filed that loss under churn and moved on. It took me longer than I’d like to admit to see that most of those clients were never lost at a moment I could point to. There was no row in any report where it happened. They drifted, somewhere I had no view of, and by the time it showed up in the numbers the decision was months old.
That’s the part that has stayed with me, and it’s the part this piece is about. The people who decide whether to keep buying from you, or to buy from you in the first place, do most of that deciding in a place you cannot see. The question that matters in 2026 isn’t whether you should look after the ninety percent who aren’t ready yet. Everyone agrees you should. The question is how — how you actually engage that many people, early, as individuals, without it costing more than it returns. For most of my career there was no good answer. Now there is.
Nurturing was always the answer. So why does it keep failing?
Lead nurturing fails, most of the time, because it is one-way. The business does the talking. It builds a sequence of emails, points it at everyone who didn’t buy on the first pass, and waits for a hand to go up.
The waiting is the flaw. A nurture sequence is built to sit still until the lead does something visible — clicks the link, books the call, replies. So it treats a person who is quietly weeks from a decision exactly the same as a person who will never buy, because from the outside, on the day, they look identical. Both are just names on a list that hasn’t converted yet. The same five emails bore the one and insult the other, and the business calls the whole list “cold” and asks marketing for more leads.
The buyer moved earlier. Your structure didn’t.
Buyers now do most of their deciding early, and they do it quietly. They read, they compare, they ask people they trust, they form a shortlist — often before they have spoken to a single salesperson, sometimes before they have given you their name. By the time someone fills in a form or asks for a quote, the real work of deciding is largely done.
Most businesses are still built for the old order, where a buyer announced themselves at the top and was walked down a funnel while sales waited at the bottom for a qualified lead to arrive. The handoff between marketing and sales — the seam where one is meant to pass the lead to the other — was set late in that journey, because that’s where the decision used to happen. The decision has moved forward. The seam mostly hasn’t. So businesses arrive to engage at the moment the buyer has already half-chosen, and wonder why the conversation feels like catching up.
By the time a lead raises its hand, the deciding is mostly done. You’re not entering the race. You’re arriving after it.
Most of what matters is invisible
Here is the harder truth underneath all of it. The signals that actually tell you someone is moving toward a decision mostly happen where you can’t see them. The prospect who reads your page three times late at night and never fills in anything. The person making your case inside a buying group you’ll never sit in. The quiet comparison against two competitors, the conversation in a peer group, the budget being lined up a quarter ahead. None of it shows up in your CRM.
What does show up — opens, clicks, downloads — is a thin and often misleading trace. I’ve watched businesses pour attention into the lead who opens every email and never buys, while the one who’ll sign next month looks cold on every metric they track. The visible data isn’t just incomplete. It can point you in exactly the wrong direction, because the easiest things to measure are rarely the things that predict a sale.
So why don’t businesses just engage the 90% earlier?
Because, until very recently, it was too expensive. Engaging people early and individually meant putting a human on them, and the human who could do it well — someone senior enough to be useful in a real conversation — is your most expensive resource. No business can afford to point that person at ninety percent of a pipeline, most of whom won’t buy this quarter and some of whom never will.
So a gate was built. Marketing-qualified, sales-qualified, a scoring threshold a lead had to cross before a salesperson’s time was spent on it. That gate was a rational decision. It protected expensive people from being spread too thin. It also did something quietly costly: it wrote off the ninety percent by design, because looking after them properly was a luxury the maths couldn’t justify.
What changed in 2026
The cost wall came down. Being present with someone early no longer requires your most expensive person, and it no longer requires you to choose which handful of leads are worth a human’s attention.
Part of that is a shift in who does the engaging. Early presence isn’t a sales job — it’s a help job. Depending on the business, and on whether you’re selling to other businesses or to consumers, the people best placed to engage the early phase are often closer to customer service than to sales: geared to be useful, not to close. A buyer who isn’t ready responds far better to genuine help than to a pitch, and help is cheaper to give than a closing conversation.
The larger part is that AI can now hold presence across that invisible middle at a scale no team ever could. And this is the point I want to be precise about, because it’s the opposite of how it’s usually sold. The value of AI here is not that it engages your leads for you. It is that it makes the invisible middle visible. The messy early phase was always full of people deciding to buy. You simply couldn’t see them, so you sent everyone the same email and hoped. AI turns the light on in that room. The engaging is still done by people — it’s just that now your people can see who to show up for, and when.
AI didn’t replace the human engagement. It made the invisible visible, so a human could finally show up where they never could before.
That distinction is the whole thing. It’s not AI alone, and it’s certainly not AI instead of people. It’s two kinds of intelligence doing what each is good at: the machine making sense of a scale of quiet signal no person could read, and the person doing the part that was always human — the actual relationship. That’s the moment nurture stops being something you do blind and becomes engagement. Nurture was the monologue you ran because you couldn’t see who was listening. Engagement is what becomes possible once you can.
What this looks like in practice
It looks less like a campaign and more like a conversation held at scale. You engage the ninety percent early, while they’re still deciding, instead of waiting for the few who declare themselves. You treat the contact as two-way — reading what each person is actually doing and responding to it — rather than pushing one sequence at all of them. You keep the expensive human for the moments that genuinely need a human, and you let a help-focused layer and the machine carry the long, patient, early middle that used to be impossible to staff.
None of that is a longer email sequence or a cleverer drip. It’s a different posture: present rather than waiting, reading rather than broadcasting, useful rather than selling, until the moment a real conversation is worth having — at which point you already know who you’re talking to and roughly where they are.
The ninety percent were never junk, and they were never really lost. They were invisible, deciding in a room you had no light in, and they bought from whoever happened to be present when they were ready. For most of business history that was a matter of luck. In 2026, for the first time, it doesn’t have to be. The bridge between the people who aren’t ready and the few who buy has always been engagement. The only thing that’s changed is that you can finally afford to build it.