Watch the full Revolt! session on YouTube
At Revolt!, Sangram Vajre, co-founder of GTM Partners and Terminus, made five predictions for go-to-market in 2027. He made them with no slides and no hedging.
The number that stuck with me: only 20 cents of every dollar spent on AI tokens ties to a business outcome, according to Accenture. Sangram thinks that bill, along with NRR and how you go to market, will decide who's still standing in 2027. The full recording is below.
Last year, 78% of sellers missed quota.
14% of reps drove 80% of new revenue.
The other 86% weren’t less talented, they were working a worse list.
Lusha fixes the list.
Every morning your reps open up to 10,000 accounts ranked against your ICP,
with the reason on every row and the buying group one click away.
Set the ICP once, share it, and it works in Lusha Workspace, Claude,
ChatGPT, your CRM, or any agent you build.
What you’ll learn
Why AI tools are following the martech boom, from fragmentation to consolidation
How 120% NRR doubles a business with no new customers
Why AI cost needs its own line item next to sales and marketing
What happens when you ship product faster than customers can adopt it
Why how you go to market is the moat that lasts
Session Highlights
00:57 - Prediction 1: AI tools are the new martech
01:39 - 15,000 martech tools, almost zero growth
03:04 - AI is scaling, growth isn't
06:44 - Prediction 2: NRR is the new growth engine
07:55 - 120% NRR doubles you with no new customers
09:37 - How Bending Spoons grows on NRR
11:44 - Usage-based vs seat-based NRR
12:27 - Prediction 3: Token maxing is the new tax
14:56 - 20 cents back on every AI dollar
16:27 - 30x the cost of a chatbot
18:28 - Prediction 4: Your next product isn't on your roadmap
20:51 - 200 customers, only 20 using the product
23:06 - Prediction 5: GTM is the new moat
25:56 - Service as a software
32:31 - Q&A: the one metric to watch
Key Takeaways
1. AI tools are the new martech
More than 15,000 martech tools existed in 2026, and Scott Brinker's State of Martech shows the landscape flat for the first time in 15 years. New products launched and almost as many vanished. Content marketing lost more than 175 products in one year, the largest category contraction on record. Sangram expects consolidation next, and a new stack in 2027 for agents, workflows and orchestration. He sees fractional GTM leaders building the layer between AI models and the operators who use them.
2. NRR is now the growth engine
Sangram asked HubSpot's CEO which one metric she watches as a public company. She said NRR without blinking. At 120% NRR, a business doubles every 3.8 years without adding a single new customer. Benchmarkit puts the median around 100%, which he calls a treadmill, and below 80% a company is in a spiral. At $50M ARR, 75% of growth comes from expansion. Usage-based companies run about 108% NRR against 98% for seat-based.
3. Token maxing is the new tax
GTM Partners built a super agent for its 150-plus certified partners and watched its AI bill climb. Accenture found only 20 cents of every dollar spent on tokens ties to a quantifiable business outcome. An orchestrated agent interaction costs about 30 times a simple chatbot exchange, and AI spend can jump from $1,000 to $50,000 a month when ten people use it badly. Gartner expects 40% of agentic AI projects to be canceled by the end of 2027, mostly on cost. His advice is to give AI cost its own line item and manage tokens the way teams once managed cloud spend.
4. Your next product won't come from the roadmap
One $20M CEO he works with ships product every week, and her customers have started churning quietly because they can't keep up. Another company has 200 customers and only 20 using the product. Sangram predicts companies that keep shipping faster than customers adopt will close down faster. Bending Spoons buys companies like Evernote, stops product development, raises prices and keeps the customers who stay. NRR goes up and the business grows.
5. Go-to-market is the moat that lasts
Features get copied in days. Sangram watched one email tool raise $28M on auto-drafting, then saw another tool he uses ship the same feature two days later. The moats he trusts are brand, ecosystem and network effects, plus what he calls service as a software: headless products where an operator designs the workflow and the customer approves the output. HubSpot's moat is the 20,000 to 30,000 people who show up to Inbound. OpenAI just posted a $400,000 role for a head of events.
"Go to market is the business."
If you're enjoying these, reply and tell me which of the five predictions you'd bet against.
Catch you next week,
Jared
P.S. Sangram shares GTM Partners research on LinkedIn every week. Worth a follow. Also, always growing our RevGenius community; you can join here to follow along.
Upcoming events
Revolt! 2026 Demo Day 1
Wednesday, November 4 | 12 PM ET / 9 AM PT
Revolt! covered the ideas. Demo Day shows you the products behind them. ZoomInfo, Airspeed, and TitanX each walk through their product live, with time for your questions in the chat.
Revolt! 2026 Demo Day 2
Thursday, November 5 | 12 PM ET / 9 AM PT
Three more live demos from Goldcast, Syft, and Firmable, focused on turning content into pipeline and finding the accounts worth working.
Know The Buyer, The Tech, The Pitch
Tuesday, November 10 | 3 PM ET / 12 PM PT
Your reps know the buyer's name, title, and company size before the first call. They usually don't know what software that company runs. Chester Thompson (VP Americas) and Alec Van Cleve (Enterprise Sales Manager) of Firmable on closing that gap before the pitch.
Want to get your brand in front of 100,000+ revenue leaders?
Partner with us and put your name at the center of the conversation.
