co-founder @amplemarket, @fermatslibrary. physicist. yc (w14). 👨🏻💻 in foggy SF 🌁
San Francisco, CA
Joined July 2011
- Tweets559
- Following483
- Followers2.8K
- Likes3.3K
Pinned Tweet
10 tips for Founders doing Sales (from a Founder) - a thread
I have a technical background (I am a Physicist by training) and had 0 experience when I first started doing sales. Here are the 10 most important things I learned since I started doing sales. 👇
Come run with a bunch of AI and GTM enthusiasts 😊
Coffee on us ☕️ sign up here: luma.com/8lrofron
We just launched the new Amplemarket University.
Videos, webinars and playbooks on what actually works for Sales, GTM and Revenue teams. As always amazing work by the @amplemarket team! 🖤
Great weekend binge watching. 😂
the market is catching up to what the @amplemarket team has known for years. 👀
Great insights
“We just gave every employee, even the worst ones, infinite headcount and budget.”
“People are spending so much on tokens because they don't know how to use them.” 👀
Great insights @Carles_Reina - 3 I'd push even further:
Outbound from day 1, but the 2026 version. Templated volume is dead. The new outbound runs on context: who to reach, why now, what to say, and which channel to say it in. Volume without context is just spam with a personalization token.
Enablement isn't a headcount milestone as many leaders think. It's actually a product-velocity problem. When product ships fast, the pitch falls behind within a quarter and reps keep selling the demo they saw on day one. You're not losing to competitors - you're losing to your own old demo.
Not all deals are good deals - a hard lesson for a founder to learn. The wrong ones are just churn you're buying twelve months in advance. And you can't coach a seller into walking away from a deal their comp plan pays them to close. Qualification is a comp problem before it's a training problem - incentives do the coaching.
If I had to restart @ElevenLabs GTM from scratch, these are the things that I would change 🖊️
- Hire quicker. I was alone selling for the first 9 months because I believed we didn't have PMF. Main reasons: the AI tourist external mantra of 2023, and the uphill battle convincing everyone to hire more GTM (Sales development, Sales, Customer Success) because we were still mostly PLG / self-service. The hiring constraint lasted until earlier this year when we boosted our team to meet demand.
- Build an enablement team earlier. Sales enablement has such an important role in an org, and we mostly think about it when we hire a lot of people. The reality is that we need Enablement earlier due to the speed of product iteration.
- Open more markets. The results from Uber v Lyft operating models played a key role in deciding to open markets early, but I would triple down on it if we were to start over.
- Have more big bets. Similar to VCs, you only need 1 big bet to work in order to generate Alpha. We could have executed more big bets if we had had a single person owning & iterating on them. Former successful founder profile with drive and speed.
- Be more vocal about product ROI & our success. People buy from people, and companies buy based on success & ROI. In AI, we've all been slow showing ROI. This should be the key number when you start a conversation.
- Hire senior sellers. In startups, we've been told that culture is everything and someone with +25y experience may struggle to fit in. This is bullshit. The experience, connections and drive +25y sellers bring is crazy. We have a bunch today, but I would hire many more earlier.
- Reduce cash commissions and replace them with more equity. Cash is king, but Equity is queen. Queens also rule countries, so why not make Equity the dominant one in GTM?
- Open more offices. We are a remote-first company, but our customers are everywhere, and employees want to meet each other. I would open more offices to be embedded in markets quicker.
- Fun brand. Brand perception is 33% of the work. Building a fun brand adds quickly.
- Outbounds outbounds outbounds. We waited too long to migrate to an Outbound organisation. I would do it from day 1 now.
- Pipeline construction with more whales. The ideal pipeline needs to have a combo of segments to reduce risk, improve results and provide liquidity. We waited too long to pitch & bring whales.
- Not all deals are good deals. We, sellers, want to close a lot of ARR, but it is important to split early on between good vs bad deals.
Thank you to Tomasz Gmur for asking how I would do GTM again!
People tolerate mediocrity to a degree that's hard to describe. Company culture is mostly just refusing to.
And you can't install that refusal. You can't make people care - you can only hire people who already do.
The best people I work with treat what they do as their own craft. Getting better at it isn't a chore for them - it's the fun part. When the whole team is like that, the quality bar takes care of itself. You stop enforcing it because nobody around you can stand shipping something mediocre.
It shows up in the small things. The bug that annoys one rep, the half-second of friction nobody notices on a demo, the detail a customer feels every day but never mentions. The details - that's where care actually lives. Not in the big launch, but in the thousand quiet things around it.
Craft is just who you hire.
The hardest part of putting AI into a sales team isn't the AI. It's context.
A model is only as useful as the context it has. Give it the full picture of an account and it can genuinely help. Give it a fragment and it'll still answer and that answer is often wrong. What's scarce now is context, not intelligence.
Look at how most teams are set up: data in one tool, sequencing in another, the dialer in a third, deliverability somewhere else, the CRM its own island. Add AI on top and you're not giving one assistant the whole picture - you're giving five tools a piece each, and none of them can see what the others see. More AI, not much more outcome.
The answer isn't one more tool with "AI" in the name. It's putting the workflow back in one place, so the AI can reason across the whole thing: who to reach, why now, what to say, whether it landed, what to do next.
That's why we built Amplemarket the way we did - not AI on top of a stack, but one platform where the context was never split up to begin with.
The AI was never the hard part. Owning the context was.
stay tuned we have lots of interesting things brewing at @amplemarket
FUN FACT: the previous record was set by accident.
When Apollo 13 suffered an explosion en route to the Moon in 1970, the crew had to swing around the far side on a free-return trajectory to get home. That emergency detour took them 248,655 miles from Earth but it was never the plan.
Artemis II just broke that record on purpose!
Approaching the near side of the Moon.
The Artemis II astronauts have surpassed the record for the distance from Earth at 1:56 ET (1756 UTC). This record was previously set during the Apollo 13 mission when the astronauts traveled 248,655 miles from Earth. The Moon continues to grow larger and larger in the windows of the Orion spacecraft as the Artemis II mission gears up to observe the far side. The astronauts are predicted to make their closest approach of the Moon around 7:02pm ET (2302 UTC).
ALT In this view of the Moon, the near side (the hemisphere we see from Earth), is visible at the top half of the Moon disk. It is identifiable by the dark splotches. These are ancient lava flows from a time early in the Moon\u2019s history when it was volcanically active. The large crater that appears below the lava flows, dark in the center, is Orientale basin, a nearly 600-mile-wide crater that straddles the Moon\u2019s near and far sides as is partly visible from Earth on the edge of the Moon. In this image, we have a full view of the crater. Everything below the crater is the far side, the hemisphere we don\u2019t get to see from Earth because the Moon rotates on its axis at the same rate that it orbits round us. Credit: NASA
You can now run Amplemarket on @claudeai and @ChatGPTapp! Prospect, enrich, and engage -> build and manage your entire pipeline from a conversation.
Amazing work by the team! 🖤
💯 this!
This from @natfriedman's (brilliant) personal website is a great explanation for why telling founders to go slow and take their time is terrible advice. There's no tradeoff between speed and quality for startups.
Amazing work from the team - as always!!!
Enterprise sales has changed.
Big deals aren’t won with more emails.
They’re won by orchestrating the account.
Introducing Amplemarket for Account-based Selling → unify signals, stakeholders & conversations in one workflow.
🔗 dub.sh/account-selling
It's Mathematically Impossible to Play Tetris Forever‼️
This was proven by John Brzustowski in his 1992 master's thesis at the University of British Columbia. The proof relies on a fundamental flaw exposed by the S and Z tetrominoes (called "right kink" and "left kink" in the thesis).
The Problem Pieces: The S and Z tetrominoes can't be placed on a flat surface without potentially creating jagged "overhangs" or dependencies that force holes or uneven stacks (unlike I, O, T, J, L). They are the only pieces that introduce asymmetry, making it harder to clear lines perfectly when they come in bad sequences.
The Adversary: The proof treats the game like it has an evil opponent controlling which pieces you get. Instead of random pieces, imagine someone actively trying to make you lose by giving you the worst possible piece at the worst possible moment. In real Tetris, pieces come randomly, but over enough games, those worst-case sequences will eventually happen naturally.
The Doomsday Sequence: The adversary (or bad luck over time) uses a diabolical strategy to guarantee your defeat:
1️⃣ Flood you with Z pieces: Send a long run of left kinks (Z pieces) until you're forced into a repeating pattern just to survive.
2️⃣ The fake-out: Give you one more Z piece, but show you an S piece is coming next (remember, Tetris shows you the next piece).
3️⃣ Flood you with S pieces: Then send a long run of right kinks (S pieces) until you're stuck in another cycle.
4️⃣ Another fake-out: Give you one more S piece, but preview a Z piece.
5️⃣ Repeat forever: Keep alternating between these floods of opposite pieces.
This creates protective barriers where interlocked pieces form structures that cannot be dismantled without creating holes. Meanwhile, complete horizontal rows accumulate below these shields - rows that would normally clear but are now permanently trapped.
No matter your skill, this structure fills the well (typically 20 rows high) and ends the game. In an infinite game with random pieces, such sequences are statistically certain to happen.
