Evening, CEO!
Let’s talk about ElevenLabs.
You know them. They are the reason your Instagram feed is full of motivational quotes read by a voice that sounds suspiciously like Morgan Freeman but definitely isn’t.
In about three years, they went from “two guys in a London apartment” to a multi-billion dollar unicorn that basically owns the voice AI market.
Meanwhile, in that same three-year timeframe, I have successfully:
Managed to keep a single fiddle-leaf fig tree alive (barely).
Overthought 4,000 messages.
We are not the same.
But that’s fine. Because today, we are going to crack open the brain of their CEO, Mati Staniszewski, and steal the operating system he used to build this empire.
We are going to look at how he turned specific constraints into massive leverage. Because if a couple of guys can revolutionize audio while sitting in a flat in London, you can probably optimize your Q4 deliverables without having a nervous breakdown.
Here is the playbook.
1. Your “Crappy” Background is Your Secret Weapon
There is a temptation, especially in tech, to assume that if you aren’t in a coffee shop in San Francisco eavesdropping on OpenAI engineers, you have already lost.
Mati proves this is nonsense.
In fact, ElevenLabs exists because Mati isn’t from San Francisco. He’s from Poland.
Now, here is a fun fact about Poland that I just learned: Voiceovers there are insane.
In the US, if you watch a dubbed movie, you get a full cast. The hero sounds like a hero. The villain sounds like a villain.
In Poland, traditionally, they used a “Lector.”
This is one guy. Just one dude. usually speaking in a monotone, flat voice.
He reads the hero’s lines. He reads the grandma’s lines. He reads the alien’s lines.
It is a terrible user experience. It strips all emotion out of the art.
But if Mati had grown up in California, he never would have noticed this problem. He would have taken high-quality dubbing for granted. Because he suffered through the “One Guy” problem, he understood deep in his bones that the world needed AI that could capture emotion, not just words.
The Lesson for Us:
Stop trying to copy the “cool kids” in your industry.
Your leverage comes from the weird, specific, annoying things unique to your situation.
That manual spreadsheet process you hate? That client who asks the same question 50 times? That is your “Polish Lector.” That is the pain point you understand better than anyone else. Solve that, and you become irreplaceable.
2. The “Slider of Shame” (Or: How to Stop Overthinking)
I am a recovering perfectionist.
If I need to post a 15-second Instagram Reel, I will spend 30 minutes debating whether the font “Courier New” conveys enough irony, or if I need to download a new font pack. I end up posting nothing.
ElevenLabs had this problem too.
Early on, users kept asking for a simple feature: A slider to control how fast the voice speaks.
Simple, right? Just give them a speed knob.
But the ElevenLabs team was full of purist researchers. They said, “No! That is crude! The AI model should be smart enough to intuit the correct speed based on the emotional context of the text!”
So they spent nine months trying to teach the AI to guess the speed.
Nine months.
The users were annoyed. The researchers were stressed. And the AI still wasn’t getting it quite right.
Finally, they gave up. They stopped trying to be geniuses and just built the slider into the product interface.
It took a fraction of the time. The users were thrilled.
Mati turned this failure into a heuristic I am now taping to my forehead: The 3-Month Rule.
If the “Research” (the deep, perfect, automated solution) can’t solve the problem in 3 months, the “Product” (the manual, slightly hacky, human intervention) has to take over.
The Lesson for Us:
There is a battle inside your brain between the Architect and the Builder.
The Architect wants to spend six months designing a perfect, fully automated, AI-powered content machine.
The Builder just needs to get a post out by 5:00 PM today.
Don’t let your inner Architect hold the project hostage. If you can’t automate it perfectly today, just build the “Slider.” Do it manually. Use the hacky fix. Ship the B+ work so you can live to fight another day.
3. Paying Commission on the Deal You Killed
This is my favorite part of the story because it hurts my brain in a good way.
ElevenLabs has a sales team. Salespeople like money. They get money through commissions.
One day, a massive competitor—a huge Foundation Model company—came knocking. They wanted to license ElevenLabs’ voices for their demos.
For the sales team, this looked like a jackpot. Huge contract. Massive commission check.
But Mati realized something: Strategy is a Leading Indicator. Commission is a Lagging Indicator.
If they sold their voices to a competitor, they were arming their enemy. They would get a pile of cash today, but they might lose their business tomorrow.
So Mati killed the deal. He said “No.”
Normally, this is the part where the sales team riots and burns down the office because they lost their bonus.
But Mati did something genius: He paid them the commission anyway.
He realized that the sales team did their job. They found a customer. It wasn’t their fault that the strategy required saying no. If he stiffed them, they would stop bringing him opportunities. They would start hiding things.
The Lesson for Us:
In your “Me Inc,” you are constantly negotiating between Short-Term You (who wants the easy win/money/dopamine) and Long-Term You (who wants a sustainable career).
Sometimes, you have to turn down a project that pays well but ruins your reputation.
Sometimes, you have to say no to a “promotion” that is actually just 50% more work for 0% more pay.
When you make those hard strategic choices, you have to find a way to reward yourself anyway.
Don’t punish yourself for making the hard, right choice. If you skip the “easy win” to protect your long-term value, buy yourself the fancy coffee. Take the Friday afternoon off.
Pay yourself the commission, even if you kill the deal.
To Recap:
Embrace your “Outsider” status. Your specific frustrations are your roadmap.
Stop researching, start shipping. If you can’t perfect it in 3 months, use the “Slider.”
Protect the Asset. Don’t sell your future for a quick win today (but reward yourself for the discipline).
I am off to go water my fiddle-leaf fig. It’s looking at me with judgment.
Links:
https://uk.linkedin.com/in/matiii
https://elevenlabs.io
ElevenLabs CEO/Co-Founder, Mati Staniszewski:The Untold Story of Europe’s Fastest Growing AI Startup
Mati Staniszewski at ElevenLabs Summit 2025 in San Francisco
