80 Photos and videos
Welcome to The Distribution Era.
1
4
283
We were fortunate to back @SayPrimer early, and now paid is rapidly evolving in The Distribution Era.
NEW: How to win at paid advertising in the AI era.
@kcpdelaney (Co-Founder and CEO of @SayPrimer) joins GTMnow to break down the B2B paid squeeze, the B2C-for-B2B unlock, and his 3 contrarian rules for paid in 2026.
He has the data on what's actually working, and it's not where you might assume.
Primer's targeting layer drives 80% match rates on Meta, 50% on Google, and 70% on Reddit for B2B advertisers, vs the 2 to 10% standard you get uploading work emails.
Highlights:
01:44 Why AI made it more expensive
05:04 LinkedIn CPMs went from $20 to $800 in 18 months
05:59 The B2C-for-B2B unlock that Clay and Primer are racing for
07:01 What Primer does: 80% match on Meta, 70% on Reddit
09:19 Why attribution is where startups go to die
17:14 Why doubling down on Google search is "kind of foolish"
23:13 Three philosophical rules for paid: targeting, niche, credibility
26:13 The most undervalued tactic in B2B paid right now
30:34 Why you must push conversion data back into ad platforms
34:21 Ads in ChatGPT within 6 months prediction
1
42
gtmfund retweeted
Jun 4
⚡️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
NEW: Selling to the government takes 3 years before you get a foot in the door. Esper has cracked the GovTech code and shares learnings in this conversation.
@MalekaMomand (co-founder & CEO) joins GTMnow to break down how to successfully sell to governments / risk-averse buyers, what she's actually seeing on the ground with DOGE, and more.
Policy governs how we live and work, yet the average one takes ~20 people to write and was built by passing paper desk to desk.
Esper is the operating system for government policy: the source of truth and system of record for the regulations governing NYPD officers, Tennessee health workers, Arkansas licensing boards, and much more.
Highlights:
00:10 What Esper actually is: an OS for government policy
02:06 NYPD's 3000-page policy book, updated by hand with tape
05:23 How Tennessee used Esper during COVID to expand rural healthcare
08:18 The CKV framework: Esper's ICP for govtech
10:16 Two pieces of advice for new govtech founders
11:27 Why professional services is the most underrated moat
14:03 "Without being spicy": Maleka's unfiltered take on DOGE
16:54 The funding round and the "SaaS apocalypse" debate
22:32 Why founders should read fiction, not productivity books
24:14 The AI take: data quality is the real blocker
2
2
7
469
gtmfund retweeted
Jun 3
We closed a competitive deal on the Friday of Thanksgiving week and the founder said we picked you to lead because you work founder hours. Always on.
Jun 2
I was once pitching in a board room at a top 3 VC firm for a $15M Series A.
12 people in the meeting. One of the GPs fully fell asleep. Out cold for 30 minutes. Nobody acknowledged it. Everyone just kept going.
I kept presenting my Series A slides to an unconscious man in a Herman Miller chair and somehow that was considered normal. That's venture capital.
You might fly across the country to perform for people who may or may not be conscious.
It's a dance.
And sometimes you lead and sometimes you follow and sometimes your partner is unconscious.
If you're raising right now, just know: every founder has a story like this. The process is weird. The power dynamic is weird. You're not crazy for thinking it's weird.
No one talks about it because they want to continue raising. But I'm happy to stick my neck out there.
It is weird.
4
2
22
6,210
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.
3
159
gtmfund retweeted
May 29
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."
1
2
306
gtmfund retweeted
May 29
Great post @gtmfund - it’s why we’re betting big on key GTM leaders early!
1
4
257
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."
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
3
9
2,213
gtmfund retweeted
May 28
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 :-)
4
11
694
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
2
7
2,475
Big congratulations to our portfolio company @withdefault on the Series A and new product launch!
May 27
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.
1
13
2,303
gtmfund retweeted
May 20
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.
May 19
BREAKING: Armada Raises $230M Series B at $2B Valuation co-led by Overmatch, BlackRock & 8090 Industries
The company’s largest & heavily oversubscribed round to date, bringing total funding to nearly half a billion dollars.
PLUS, Armada announces Galleon Forge One, a new modular data center manufacturing facility
“Johnson Controls is working with Armada to rapidly deliver secure modular data centers at scale,” said Joakim Weidemanis, CEO of Johnson Controls.
"Together, we have already deployed units across the United States and around the world, demonstrating the expertise and global reach required to support mission-critical environments. Johnson Controls' differentiated technology, U.S.-based manufacturing strength and Armada's edge computing expertise will deliver the thermal‑critical environments that perform predictably, deploy quickly, and scale with confidence.”
2
2
16
4,265
gtmfund retweeted
May 16
on the cutting edge @gtmfund 🔪
May 16
Thank you @HackItMax for the surprise swag for Emerald LPs @gtmfund. Proud to be an investor in fund III — third time’s a charm. LFG! 🚀
1
3
175
gtmfund retweeted
May 19
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.
May 19
The AI race is the defining race of our time. A little over three years ago, we made a bet that winning this race would require us to bridge the digital divide and deliver modular infrastructure and the latest models, full stack, anywhere in the world.
Today, we are announcing the next step in that journey with our $230M Series B funding round and Galleon Forge One. This new facility will enable us to continuously manufacture and deploy our Galleon modular data centers with speed, scale, and sovereignty.
Thank you to the Armada – our team, investors, customers, and partners – for helping us build the industrial base to win as we scale in our mission to bridge the digital divide.
2
10
693
RT @mrp: Thank you @HackItMax for the surprise swag for Emerald LPs @gtmfund. Proud to be an investor in fund III — third time’s a charm. L…
1