A dentist doesn't sell information. She sells crowns and cleanings and the occasional root canal, and if you asked her whether she runs an "expert business" she'd look at you strangely.
Now suppose she writes a short guide — what actually causes the pain you're feeling, and the four things people try first that make it worse. She gives it away. People read it, and some fraction of them arrive at her office already convinced they need the thing she was going to recommend anyway. She hasn't changed what she sells. She's changed what her patients believe before they walk in.
That's the second type of expert business, and Brunson argues it's the one most people miss because the book has "Expert" in the title and they assume it isn't for them.1
The first type is the obvious one. You take what you've learned, package it, and sell the package — courses, coaching, consulting.
Brunson's pitch for it is that it's the cheapest serious business to start. No capital, no inventory, no premises. What you need is something you're genuinely obsessed with and the ability to tell a story about it that makes someone else care.2
That second requirement is doing more work than it looks. The barrier to entry isn't money and it isn't knowledge — it's narrative competence, which is why the rest of the book is largely about storytelling rather than about business.
The second type keeps your existing business exactly as it is and bolts an information product onto the front of it.
Brunson's own case is ClickFunnels. The company sells software. But nobody arrives at the software cold — they arrive having bought a book or a course first, which explains why they'd need the software. By the time the software is offered, the customer has already been, in his word, indoctrinated.3
The financial consequence is the part he's proudest of, and it's genuinely clever arithmetic: because the information product is sold at a profit, the cost of acquiring a software customer isn't merely low, it's negative. You get paid to acquire the customer, which means acquisition stops being budget-limited.4
He lists four claimed effects: complicated sales get easier because the prospect already knows why they need the thing; you're positioned as an expert rather than a commodity, so price resistance falls; acquisition becomes effectively unlimited; and growth accelerates.5
Same tactics, different strategy — that's Brunson's framing.
It's roughly right, and it undersells the difference badly.
In type one, the information is the value exchange. If it's bad, the customer got nothing.
In type two, the information is a belief-installation device for a purchase that happens later. Its job isn't to satisfy; its job is to make the real offer feel necessary. Those are different design briefs, and they pull in different directions — the most satisfying free guide is the one that solves your problem completely, which is precisely the guide that removes the reason to buy anything else.
Brunson never names this tension. It sits underneath the whole model.
Take the ClickFunnels claim apart, because it's the load-bearing number.
Competitors and venture capitalists, Brunson says, couldn't understand how his cost to acquire a customer was "better than free."4 The mechanism: customer buys a $47 information product, which more than covers the ad spend that found them. Later, the same customer starts a software subscription. The subscription's acquisition cost has already been paid — by the customer.
If it works, it breaks the normal constraint on growth. Most companies can only spend as much on acquisition as their funding allows. This one funds acquisition out of acquisition.
Two things to hold alongside it.
The number is unaudited. It's self-reported by the person selling the method — no accounting shown, no timeframe, no churn figure.6 Churn is exactly where a model like this fails invisibly: if the information buyers convert to subscribers who cancel in month three, the acquisition was never free, it was deferred.
The claim is doing double duty. It's a business mechanism and it's the proof-of-concept for the book you're currently holding — which is itself the front-end information product in the funnel it describes.
That second point isn't a gotcha. It's the honest structure of the argument: the evidence for the model is the existence of the artifact making the claim.
You already own a business. Nothing about it is going to change. Here's the move.
Think about the last customer who bought from you after a long, tiring conversation. Not the easy sale — the one where you spent forty minutes explaining before they'd commit.
Write down what you had to explain. Not the features. The thing they didn't believe yet — that the cheap option would cost them more, that the problem wasn't what they thought it was, that the timeline was real.
Now notice that you have that conversation constantly, and you have it one person at a time, and it's the same conversation every time.
That conversation is the product. Not a brochure about your service — the argument you make before the service becomes obvious. Record it once. Badly is fine. Put it where people find you before they call you.
The test isn't whether it's polished. The test is whether the next person who calls you skips the forty minutes.
Brunson says outright that he nearly didn't call the book Expert Secrets, because he expected people selling physical products and services to see the title and assume it wasn't for them.1
That worry is well-founded, and the reason is a category error most operators make about themselves.
If you sell dental work, or roofing, or legal advice, you think of yourself as selling a service. The information you give away in the course of selling it — the explanation, the reassurance, the walk-through of options — feels like overhead. Unpaid time. The friction before the real transaction.
The type-two move is to notice that the overhead has been the valuable part all along, and that it's the only part that scales.
You cannot roof two hundred houses simultaneously. You can explain to two hundred people simultaneously why the cheap roof costs more over ten years — and the fraction of them who then call you have already had the conversation you'd otherwise be having one at a time, for free, forever.
The service stays local. The argument for the service travels.
The practical fork is whether you already have something that makes money.
If you don't, you're building type one, and the information is your business — which means the quality bar is absolute, because there's nothing behind it.
If you do, you're building type two, and the information is your marketing — which means the quality bar is relational: it has to be good enough to earn trust and incomplete enough to leave a reason to buy. That's an uncomfortable design constraint and worth naming as one rather than pretending it isn't there.
Everything here is self-reported by a party with direct commercial interest. The negative-CAC claim, the "fastest growing SaaS company in the history of the internet" framing, and the four claimed effects all appear without figures, methodology, or third-party corroboration.6 [UNVERIFIED] throughout.
Tension the book leaves open: the free information has to be good enough to build trust and limited enough to preserve the sale. Brunson resolves this elsewhere by arguing that you should teach belief rather than method — break false beliefs in the free content, deliver the how-to after purchase.7 That's a real answer, but it's an answer that makes the free content deliberately non-actionable, which is a meaningfully different product from what "give away value" usually implies.
Open question the book doesn't touch: type two assumes the information and the product point at the same conclusion. What happens when honest information about your field would lead a reader away from your product? The model has no provision for it.
Against the vault's existing creator-economy material, Brunson is unusually indifferent to the audience-first sequencing that corpus tends to assume. The Dan Koe material treats the audience as the asset and the product as downstream of it. Brunson's type two inverts that — the business exists first, and the audience is manufactured specifically to feed it.
The two aren't contradictory so much as aimed at different readers, and the difference is worth keeping visible: one is written for someone with nothing, the other for someone with a going concern who wants cheaper customers. Advice that's correct for the first can be wasteful for the second.
To Four-Metric Funnel Architecture. That page treats a funnel as a measurement structure — the four points where you can see what's happening and intervene. Brunson's type-two model adds a stage that sits before the first metric: the belief-installation step that happens prior to anyone entering the measured funnel at all. Holding both produces something neither says alone — the highest-leverage part of a funnel is the part that isn't in the funnel, because a prospect who arrives pre-convinced changes every downstream metric at once, and no amount of optimizing the measured stages substitutes for it. It also explains a common frustration: operators who tune show-rate and close-rate obsessively while their real problem sits upstream of both.
To Status Signaling. Brunson's second claimed effect is that publishing information repositions you "as an expert instead of a commodity," so people pay more for the same thing available elsewhere. That's a status claim wearing business clothing — the price premium isn't attached to the product, it's attached to the seller's perceived position. The status corpus explains why that premium is payable; the business material explains how it's manufactured deliberately and cheaply. Together: expertise positioning is a status good that can be self-issued, requiring no institution, no credential, and no gatekeeper — only publication. That's a genuinely new observation against a status corpus built largely on goods and hierarchies where the signal is expensive to fake.
Sharpest implication. If information can be sold at a profit and function as advertising, then advertising stops being a cost centre. Every constraint on growth that runs through "we can only afford so many customers" dissolves. That's the actual claim under the folksy framing, and it's why the model spread — not because information products are lucrative, but because they let you buy customers with the customers' own money.
Generative questions.
Does type two survive contact with a field where the honest information doesn't flatter the product? Every example in the book is one where it does.
If the front-end product must be incomplete by design, at what point does "strategically incomplete" become "deliberately withholding the useful part" — and is there any position on that spectrum the model can't rationalize?