If you were forwarded this newsletter, join 5,491 weekly readers—some of the smartest GTM founders and operators—by subscribing here:
Thank you for reading this hand-rolled, farm-to-table, 100% organic (human-written) newsletter!
Hey y’all!
As I continue down the path of AI and GTM, it’s never been clearer to me that the new playbook is being rewritten in real-time as we all figure it out together. Which makes it incredibly fun and energizing (to me).
One area I’m thinking about a lot is the underlying business models. I’ve spent a decade plus in B2B SaaS. But it’s evolving. And I’m starting to formulate a thesis on where it might go next. Making predictions is always risky. But, by publishing these ideas, I get feedback and can refine/improve my thinking. So, tell me your thoughts in the comments. I want to learn!
Alright, let’s jump into it.
The $100B+ Winner in GTM Tech Won’t Sell Software, It Will Sell Labor
The unbundling era.
In 2004, Uncle Marc rang the gong at the NYSE — signaling Salesforce was ready to take down the (then) incumbent, Oracle. They IPO’d at around $1B.
Today, Salesforce’s market cap is $250B. That growth paved the way for startups to unbundle certain applications to build “best-in-class” point solutions, that have become unicorns in their own right (Salesforce became the incumbent).
Zoominfo built data, Outreach/SalesLoft built email sequencing, Gong built call recording, Clari built forecasting.
The rebundling era
By 2025, these providers have begun to overlap, consolidate, and—in some cases—stagnate (see: PE acquisitions). I wrote about this in November of 2023: Consolidation szn (The race is on...) | 6 sales tech unicorns chasing after the elusive “one-stop-shop”.
These companies are realizing that to get beyond $200M in revenue, they have to rebundle across these products and provide a comprehensive tool for GTM teams (particularly on the prospecting side).
When I squint and think about what’s going to be next, it’s not 'replacing the 80 Outreach seats that a company has in their org. It’s something different.
The AI-era (selling labor, not SaaS)
The GTM company that will break out beyond a $1B valuation (and into $100B territory) will be the one that builds something that looks like labor in the form of swiping a credit card (instead of sourcing+ramping+managing people, equipped with software).
Obviously, this won’t happen overnight. Or even in the next 12 months.
→ The first iteration of this is AI SDRs.
I know this space intimately because I tried to build the first AI SDR back in 2018.
My first company, Hexa.ai, helped generate emails using “AI” (it was not). We were acquired by OutboundWorks, which was a services company that outsourced sales development.
The pitch was easy to sell (instead of hiring a new SDR who costs you $90K all in, give us that budget—actually 30% less, so we’ll call it $6K a month—and we’ll book you the same amount of meetings “per rep.”).
But really hard to deliver. There were so many factors that we couldn’t control, including simply whether or not a company had product-market (aka: a good offer).
AI SDRs today are very similar.
The pitch is simple: “instead of hiring more human SDRs, give us that budget and we’ll automate it for you, using AI.”
But, they’re not delivering (yet). If any of these AI SDR companies were so good that they were actually coin-operated ($1 in gets you $2 out, predictably), they would be ripping right now. Everybody would be using them and using more of them. That’s not happening. They are mostly just pilots. I think they could get there, but they’re not there today.
So, back to my point, the next generation of vendors in the space will be companies that can get budget, not from SaaS spend, but from labor spend on the P&L.
Companies that can unlock this are not going to be $1 billion companies or even $10 billion companies. They’re going to be $100 billion+ companies.
It’s the world of agents that you read on AI Twitter. But for real. Deployed in production, in the real world, at scale, at real enterprise companies.
GTM operators will become orchestrators of AI agents (“labor”), instead of managing people (who use software).
→ Another version of this is consumption-based pricing / token usage.
Fin, by Intercom, is doing resolutions for customer support. You pay $1, or some amount, to resolve an issue.
You can look at Clay in a similar way, where it’s consumption-based (Clay credits).
What I love about consumption-based pricing products is that the incentives are directly aligned between the user (customer) and the product (vendor).
The vendor is only going to build things where tokens are consumed and people pay for those tokens. The more tokens you use, the more value you get. You’re only going to use more tokens as you see more value. It’s a flywheel.
Consumption-based pricing is not labor, not yet anyway. But it’s another step towards that inevitable future.
Maybe SaaS companies can expand into this new future. Maybe. I’m skeptical.
It’s much more likely to come from “AI-native” companies that are built from the ground up with all the scaffolding and architecture required to build in this new way.
Selling labor, instead of software, is going to be a major unlock over the next decade. And whatever GTM tech provider can pull it off will create $100B+ of enterprise value.
Corners of the internet I explored this week:
Karpathy on Dwarkesh’s pod has been making the rounds on tech twitter — he presented bearish (realistic?) on AI, which was interesting. Judge for yourself: “Andrej Karpathy — We’re summoning ghosts, not building animals.”
I also went back and listened to the Gwern interview, which is fascinating: “Gwern — Anonymous writer who predicted AI trajectory on $12K/year salary”.
And I have tabs open to read Gwern’s blog this week (incredible design).
Palmer Lucky on Bro Jogan (banger pod, controversy and all). I was hooked when they started by talking about float tankssensory deprivation pods (ha). Seriously though, I’m a big fan of them (it’s like forced meditation / can be psychedelic).
The Bros (TBPN) with the EXCLUSIVE on the latest development with the Microsoft<>OpenAI relationship. Truly, a generational run. Bullish on media. The attention economy in the AI-era is going to melt faces.
I continue to be AI-pilled. One of the godfathers of SaaS sales, Jason Lemkin, lays it out bluntly in this 30-second YouTube short (tldr: “You don’t need to learn AI… you need to do AI”).
That’s it for this week!
As always, thank you for your trust and attention — I do not take it for granted.
See you next time,
Brendan 🫡