Raise money from those who have been there, done that. Let us help you grow revenue from zero to IPO.
Media platform @GTMnow_ | GPs @hackitmax & @PaulGTM
⚡️Loved recording this episode with @sophiebuona !
We talk about how to win first government customers, the secret to incredible customer retention, and my take on SaaSpocalypse 🔥
Thanks @gtmfund for being fantastic partners on our journey
The story of enterprise software is the story of an evolving moat.
Each era opens with one cost as the binding constraint – capital, deployment, distribution, or building itself – and the companies that own that era recognize and weaponize that constraint for their benefit.
banger piece worth reading
"The alpha of early adoption has never been wider and will widen further before it narrows. Becoming AI-native in your own GTM today produces the widest disparity of performance we've ever seen between teams that adopt early and teams that don't. The aperture for what's possible in building a GTM engine will also continue to widen. The limiting factor is no longer the GTM technology stack you use, but your imagination and your team's operational execution. This phenomenon has always existed for early adopters of technology, but it's never been this acute. If you're not ahead of it, you're getting left behind faster than ever. The alpha of early adoption is the difference between making a category and being absorbed by one.
Welcome to the Distribution Era. We’re just getting started."
If he were starting a data company today, this is what @ttunguz would build:
Images and video.
The reason behind this is in the math.
"An image is probably 1000X to 10000X bigger than a text file. And a video is two or three orders of magnitude larger than that."
Text is already straining the infrastructure we have.
Custom video is coming, robotics is coming. Both run on data volumes the current stack cannot move.
"We need much bigger infrastructure to be able to support it."
@HackItMax The moat had to move because the previous moat collapsed... AI is the first wave that meaningfully eroded the cost of building the product itself.
Welcome to The Distribution Era.
Episode 11 of the @GTMnow VC Series dropped this week with @ttunguz, GP at @Theoryvc.
Before we get into the conversation with Tomasz, @PaulGTM and I talked through what we're seeing across public and private markets right now. A few things we covered:
1. The picks-and-shovels for the picks-and-shovels.
Public and private markets are blurring, and secondary liquidity is everywhere. The names running 2x to 10x recently sit in energy, chips, and memory. Capital is moving down the AI value chain, away from pure software and into the layers that feed it: energy, data centers, GPUs, memory, frontier model labs, and then the application layer on top.
2. NVIDIA's data center revenue was flat until GPT launched. Then 17x in three years.
That's the cleanest single chart for how fast this cycle moves once a real demand catalyst arrives. Jensen has said NVIDIA needs roughly 1,000x more energy and compute capacity to keep up with where inference is going over the next decade. The buildout keeps waterfalling deeper into the economy than most people expect.
3. Image and video are the next memory wall.
A single image is 1,000 to 10,000 times larger than a text file. Add robotics and computer vision data, and today's infrastructure constraints will look trivial. We have a few portfolio companies building at the software-hardware-AI intersection, and the gap between what's needed and what exists is still very wide.
4. The Salesforce / Anthropic math.
Salesforce reportedly spending $300M with Anthropic sounds enormous until you do the math: roughly $15k to $20k of incremental AI cost per employee. Fully burdened headcount at a company like Salesforce runs $150k to $500k+ for individual contributors. If even a fraction of that work gets automated or augmented, the ROI is obvious. This is the math every CFO and head of engineering is running right now.
5. Marketing to agents is here, running in parallel with marketing to humans.
Full agent-to-agent commerce is probably further out than the hype suggests, and parallel discoverability is already happening. Buyers are using agents to research vendors before any human call. That changes how marketing teams structure information, how content is packaged, and what "SEO" even means when the first reader is a model. Companies like Noble, Avara, and Tavis are building for this layer in our portfolio.
Excited to announce Theory leading a new round into one of our port cos, but that'll have to wait a few weeks :-)
NEW: The AI data center buildout will be the fifth largest infrastructure project in history this year. Bigger than either world war.
The General Partner at @Theoryvc, Tomasz Tunguz (@ttunguz), joined GTMnow to break down the $575B infrastructure bet, the fusing of the data and AI stacks, and why there is now a second buyer in every enterprise deal.
For every $1 hyperscalers make from AI, they are spending $12 on infrastructure. Data centers hit 3.5% of US GDP this year and could reach 6 to 7% by 2030.
Highlights:
00:35 Why AI data centers are the 5th largest infrastructure project ever
01:26 "The demand for inference is infinite": the road to 2030
02:17 The $575B game of chicken: how the infrastructure bet plays out
03:48 Why the data stack and the AI stack have fully fused
05:33 What founders need to know about positioning now
06:37 Why product market fit is no longer binary
07:43 The new buying committee: marketing to humans and agents
12:03 AI agents that can influence and change your decisions
18:25 The underrated pattern behind successful founders
19:53 Tomasz's hot take: AI will rewrite org design entirely
Today, I'm incredibly excited to announce @withdefault's Series A led by @8vc and our new product.
We spent the last three years building the tools and orchestration layer companies like @owner and @AirbyteHQ have used to build infrastructure across their go-to-market functions.
Today, we're launching the first real-time data layer for go-to-market, a powerful new revenue agent, and a suite of tools for revenue teams and their agents.
It's Day 1.
Exciting announcement from @gtmfund port co Armada. One of our first checks from Fund II was in the pre-seed in 2023 - Jon, Dan, and team have executed flawlessly since.
Congrats to the @armada_ai team on announcing a $230M Series B at a $2B Valuation!
Plus, Galleon Forge One - a new facility that will enable them to continuously manufacture and deploy their Galleon modular data centers with speed, scale, and sovereignty.
A great 50-person dinner in San Francisco last night!
Video of part of cocktail hour.
This community is what makes @gtmfund so special.
Thank you to our sponsor partners: Nooks // Glyphic // HockeyStack.