Distribution Beats AI
Key Points
- Distribution, not AI breakthroughs, is the primary driver of long‑term economic advantage because making software easier with AI doesn’t solve the increasingly complex challenge of getting it into users’ hands.
- Sam Altman (as cited) argues that a billion‑user platform with solid distribution will be far more valuable in five years than the most advanced AI model, emphasizing that reach outweighs raw technology.
- Salesforce exemplifies the power of distribution: its entrenched, industry‑wide presence creates a network effect that makes it extremely difficult for competitors to displace, regardless of product hype.
- Investors like Gary Tan prioritize a founder’s distribution strategy over product features, recognizing that sustainable growth hinges on how effectively a solution can be delivered and adopted.
Sections
- Distribution Beats AI in Value - The speaker argues that while AI eases software creation, distributing software remains increasingly difficult and more economically valuable than AI breakthroughs, citing Sam Altman's preference for a billion‑user platform over the most advanced AI model and highlighting the critical role of distribution for both consumers and enterprises.
- Distribution Is the Real PMF - The speaker warns that the AI boom has flooded the market with tens of thousands of tools, but without existing customer distribution—true product‑market fit—most founders will fail, creating an imminent crisis.
- AI Disrupts Distribution Playbook - The speaker outlines how LLM-driven search is upending traditional funnels, making customer acquisition and visibility uncertain and leaving B2B marketers without a clear distribution playbook.
Full Transcript
# Distribution Beats AI **Source:** [https://www.youtube.com/watch?v=ptyysvuXTlU](https://www.youtube.com/watch?v=ptyysvuXTlU) **Duration:** 00:09:16 ## Summary - Distribution, not AI breakthroughs, is the primary driver of long‑term economic advantage because making software easier with AI doesn’t solve the increasingly complex challenge of getting it into users’ hands. - Sam Altman (as cited) argues that a billion‑user platform with solid distribution will be far more valuable in five years than the most advanced AI model, emphasizing that reach outweighs raw technology. - Salesforce exemplifies the power of distribution: its entrenched, industry‑wide presence creates a network effect that makes it extremely difficult for competitors to displace, regardless of product hype. - Investors like Gary Tan prioritize a founder’s distribution strategy over product features, recognizing that sustainable growth hinges on how effectively a solution can be delivered and adopted. ## Sections - [00:00:00](https://www.youtube.com/watch?v=ptyysvuXTlU&t=0s) **Distribution Beats AI in Value** - The speaker argues that while AI eases software creation, distributing software remains increasingly difficult and more economically valuable than AI breakthroughs, citing Sam Altman's preference for a billion‑user platform over the most advanced AI model and highlighting the critical role of distribution for both consumers and enterprises. - [00:03:04](https://www.youtube.com/watch?v=ptyysvuXTlU&t=184s) **Distribution Is the Real PMF** - The speaker warns that the AI boom has flooded the market with tens of thousands of tools, but without existing customer distribution—true product‑market fit—most founders will fail, creating an imminent crisis. - [00:06:17](https://www.youtube.com/watch?v=ptyysvuXTlU&t=377s) **AI Disrupts Distribution Playbook** - The speaker outlines how LLM-driven search is upending traditional funnels, making customer acquisition and visibility uncertain and leaving B2B marketers without a clear distribution playbook. ## Full Transcript
It is cold, it is rainy, it is dark
outside and we are going to talk about
distribution and why distribution
matters more than AI. So just get
excited, get the cocoa or if you're in a
hot place, I don't know, get an ice
coffee. It's coming from a long-term
perspective that I've shared here that
other people have noticed that
distribution is something that is not
getting disrupted by AI the way a lot of
the software production chain is. It is
getting easier. It is getting cheaper to
make software with AI. It is not getting
easier to distribute to people. In fact,
arguably it's getting harder because
there's more and more software scaling
up. And what's bringing this back to the
top of my mind is an interview that Sam
Alman gave. And one of the things like
he dropped a lot of things, but one of
the things he mentioned is that if he
had to pick in five years, knowing what
Sam Alman knows now about how incredible
AI is going to be between having a 1
billion user site where you have that
regular traffic coming in and you've
already acquired them or having the most
advanced AI model. He thinks the billion
user site is going to be more
economically advantageous to have. Sam
Alman himself, the model maker, picks
distribution. He picks
distribution. And it's not just for
consumers. Distribution matters for
enterprise. The reason why, even though
Clarna made a bunch of headlines and
like trumpeted that they were going to
leave Salesforce, blah blah blah. The
reason why that didn't turn into a route
for Salesforce is because Salesforce has
an incredible distribution advantage.
Salesforce is baked in everywhere you
go. If you walk away from it, it is an
extremely intentional decision and it
requires a degree of focus from an
organization that is by definition rare
because most organizations are focused
on fighting fires or one central
objective that is tied to their core
economic value. They are not focused on
their SAS stack and whether they can
sort of clean it up. So the fact that
Salesforce is there, that they have a
distribution advantage and they have a
network effect and they've got people
who have worked for Salesforce with
Salesforce everywhere in the
industry, that's irreplaceable. That is
why no matter how many pitch decks I see
that are like, we're going to beat
Salesforce, I never bet against
Salesforce. It's just the distribution
advantage is too good. It's not about
the product. I know they'd like you to
believe it's about the product when they
put Matthew McConna out there and he's
sitting in the rain in a Paris cafe,
which was a real Super Bowl ad. And it's
like, AI will help fix the rain for
Matthew McConna. No, AI will not help
fix the rain. Uh, and I don't care about
their product. I care about their
distribution.
This is why Gary Tan, who is an
investor, when he asks people, and he
admits this, when he asks
founders a question after they do their
pitch to him, they run their pitch deck.
This is the most amazing thing. What he
likes to ask is, "Hey, that's great.
Where are your customers?" And that's
not a new insight. Like Gary's been
asking that for a long, long time. But
the the point is there's no substitute
for getting the distribution. And
product market fit is basically a fancy
way of saying a startup acquires a
distribution advantage. That's
it. Either you have the distribution by
having your customers or you do not have
the distribution. And the deal is and
this is we are running into this. It's
it's a cretaceious level crisis for AI
waiting to happen. You have so many
indie founders, so many YC founders who
are just incredibly excited about
building building this year, building,
building last year, building through the
rest of this year, AI tooling out the
wazoo, like as as far as the I can see,
new AI tooling coming up. They all need
customers. They all do. And the
competition has never been more insane.
I remember people saying that the 8,000
marketing tools in the marketing
vertical was insane a few years back.
That's nothing compared to where we're
at with AI right now. It is just nothing
at all. There's like I think 50 70,000
depending on where where you count. Like
it's a ridiculous number. It's in the
five figures. It'll probably be six
figures by the end of the
year. Someone's going to have to go to
the wall. Most of these people do not
have distribution. They have to pay for
tokens somehow, don't they? They're not
all paying for tokens out of their trust
funds. They don't all have trust funds
because we've made software easier to
create. Basically, we've changed the
supply and demand relationship to
software and that has made distribution
more
valuable. Software is simple to create
now. I can create a calculator in two
seconds. That's fantastic. How do I
distribute it? Apple has distributed
their calculator to a billion people.
People use Apple's calculator. And
that's a super simple example, but you
get the idea. Works for chatbots. that
works for any other piece of software.
Distribution matters. So if we think
about that, let's take that a step
further. Distribution matters. We have a
bunch of companies that are going to go
to the
wall. How do we think about the
landscape that emerges? Who wins in that
space? The founders that win in that
space are going to be less technical
than the founders that won in the 2010s.
The builders and founders that win in
that space are going to be probably more
invested in sales net net. Like if you
basically look at founding
relationships, it's either a sales guy
or it's a tech guy most of the time. And
and I say guy because we all know that
like the vast majority of founders are
men. Not a great thing, but it's true.
Um,
and the sales side of things, whether or
not like maybe you're focusing on growth
hacking for consumer, that sort of
consumer sales, maybe you're focusing on
B2B sales if you're in the B2B
space. Someone who has the wizardry to
figure out the funnel for that is going
to
be is going to be on the short path to
riding in the private jets and becoming
the billionaire and all the rest of it.
And I say that because the entire funnel
is also getting disrupted. This is why
this is taking a while to like get
into. Distribution is getting
interesting precisely because the funnel
itself for distribution is being
disrupted by AI. Distribution is not
changing. Acquiring customers remains
really hard. But the tool sets to do
that. People's buying behaviors are
shifting. People are talking about how
chat GPT visibility is becoming the new
SEO. No one really knows what to do with
that yet, but if you search for yourself
in chat GPT, I guarantee you'll be
surprised. Try it in Claude now that
Claude has web search. Uh I was
surprised, frankly, by what Claude
found. And how do you how do you game
for that? How do you make sure that
you're visible in a way that an LLM
understands? Stripe made an effort in
that direction with LLM's.ext. So their
markdown for documents is there so that
it's easy to ingest Stripe's famously
good documentation. If you're an LLM,
there's going to be more efforts in that
direction. Those are efforts at
distribution. Markdown for docs, an
effort at
distribution.
And I think that what makes this whole
space super interesting is that it is
not like the 2010s where we all had the
Uber playbook or we all had a playbook
that we could run. Nobody has the
playbook for nailing distribution right
now. Nobody knows how B2B sales is going
to change when people are more in the
buyer seat because they're doing more of
their research for what they buy in chat
GPT than they are doing on G2. We don't
have the answer for that. There's a lot
of people that are reinventing the
existing pieces of the process and
saying you can send more emails with AI,
but that may not be the way to handle
that. Similarly with consumers. If
consumers are spending more time on Tik
Tok and chat GPT, what does the consumer
funnel look like? If it's consumer
goods, are you putting more into Tik Tok
shop? If it's consumer apps, are you
putting more into
what? How how do they find the app? Are
you sticking with
meta? Are you sticking with paid search?
Are you looking for organic? Is this a
personal productivity tool where you
want to find people on X because they
care about productivity? You have to do
all of that micro segmentation. You can
dig for them on Reddit. Maybe they're on
Reddit. The ability to segment and
figure out who your customer is at a
high level of fidelity and then
understand where they are is the key
whether it's in B2B or B TOC at getting
to that distribution advantage and
standing out in this world that is
rapidly becoming a 100,000 AI app world
where all of them need money and all of
them need distribution and distribution
has never mattered more. Getting the
next model won't solve that. It's up to
someone who can figure out how to
segment the audience and do distribution
well. So, that's where I'm at. Rainy
night, we're talking about distribution.
What could be better? Let me know what
you think about distribution.