Founder stories
Milestone achieved January 2015. Current revenue not tracked.
A mattress cover that heats and cools each side of the bed, tracks sleep through embedded sensors, and adjusts temperature through the night.
How Matteo acquired customers
Tools used to build Eight Sleep
Matteo Franceschetti and three co-founders put a sleep tracking mattress cover on Indiegogo on 27 January 2015. The $100,000 goal was met the same day, and the campaign closed at $1,231,729 from 5,202 backers.
Eight Sleep sells a mattress cover that heats and cools each side of a bed, tracks breathing and heart rate through the fabric, and adjusts the temperature through the night. It is one of the more visible hardware companies of the last decade. What is less visible is that the whole thing was paid for by strangers before a single unit could be manufactured, and that the feature almost everyone buys it for today did not exist in the first product.
Matteo Franceschetti did not come from hardware. He trained as a lawyer in Italy and worked in the finance practices of Freshfields and Allen & Overy, then built two clean energy companies, one in Milan and one in New York. He was also a competitive athlete who tracked everything about his recovery, and by his own account on the campaign page he had lost roughly 900 hours of sleep to restless legs syndrome during his years as an entrepreneur. The product came out of that irritation. He lists Eight Sleep on his own LinkedIn profile as starting in October 2014, which is the anchor used for the day counts below.
He did not build it alone or build it himself. The founding team on the campaign page is four people: Franceschetti as chief executive, Massimo Andreasi Bassi as the engineer and chief technology officer, Andrea Ballarini running operations, and Alexandra Zatarain, Franceschetti's wife, running marketing. Andreasi Bassi built the prototype.
The dates come straight from the team's own Indiegogo page, which is still live. The project was created on 11 November 2014. It went public on 27 January 2015 at 09:11 UTC with a goal of $100,000. The page records the goal being reached at 19:49 UTC the same day, ten hours and thirty eight minutes later. That is the first customer moment and the first hundred thousand dollars in one go, 118 days after the October 2014 start. The campaign closed on 27 March 2015, 177 days in, at $1,231,729 from 5,202 backers. That works out to an average pledge of about $237, so these were not small donations, they were people paying most of a retail price for a product that did not yet exist.
Worth being precise about what that number is. It is pre order revenue, not recurring revenue. Nothing had shipped. The reward tiers on the page still show delivery windows running into February 2016, roughly a year after the money came in. Selling thousands of units you cannot yet make is a validation method and a debt at the same time, and the second half is the part founders underestimate.
The other thing the 2015 page makes clear is how different the product was. The description is a cover that "tracks your sleep, warms up the bed, and connects to other smart products." It talks about auto scheduling, phone control, and integrations with thermostats, door locks and light bulbs. Warming, not cooling. The temperature split between two sides of a bed, which is the reason most current customers buy the Pod, is not in the original pitch at all. The company sold a smart home accessory, then found the real product inside the customer base afterwards.
Eight Sleep went through Y Combinator in the summer 2015 batch, having already been a StartX company at Stanford. Years later the three founders wrote up what they had learned about hardware for the Y Combinator blog, and the list reads like scar tissue. Do not wait for scale to fix unit economics: if the numbers do not work because you are small, "it's because you don't have a business, not because you don't have scale." Move faster than software teams, not slower, because so little is inside your control. Build a technical moat or someone copies the object and undercuts you. Treat customer support as part of the product. Keep backwards compatibility, because you are maintaining hardware in people's homes for years.
That last point has teeth. In October 2025 an AWS outage left Pods stuck at whatever setting they were on, some of them hot. Franceschetti posted a public apology on X within a day and committed to shipping an outage mode, which the company did. A connected bed is a product you can break remotely.
Where it landed: in August 2025 Franceschetti announced a $100M Series D, framed around a billion hours of sleep data and an AI agent that would predict a night before it happens. In March 2026 he announced a further round valuing the company at $1.5 billion, and in the same post said Eight Sleep hit free cash flow positivity the previous year, shipped three new products, reached 34 countries, and published clinical studies on hot flashes and circadian temperature. In August 2026 the company signed an agreement with the Department of Health Abu Dhabi to study sleep apnea, with the stated goal of turning the Pod into a regulated medical device.
One honest gap: Eight Sleep has never published a revenue figure through its founders. There are press estimates and investor summaries floating around, but nothing stated by Franceschetti himself, so no revenue number is recorded here.
Pre-orders are validation and a debt at the same time. Eight Sleep took $1.2M from 5,202 backers in early 2015 and reward delivery windows still ran into February 2016.
The feature that defines your product may not be in your first pitch. The 2015 campaign page sold a cover that warms the bed and talks to smart home devices. Cooling, the reason most people buy a Pod today, is absent from it.
A crowdfunding page is a public record. The campaign start time, the moment the goal was hit and the final backer count are all still readable on Eight Sleep's own Indiegogo listing more than a decade later.
In hardware, fix unit economics before scale. The founders' own advice on the Y Combinator blog is that bad numbers at small volume mean you do not have a business, not that you lack scale.
A connected product can be broken remotely. The October 2025 AWS outage left Pods stuck hot, and the fix was both an apology from the chief executive within a day and an offline mode shipped after.
Non-technical founders can lead hardware companies, but not alone. Franceschetti was a lawyer and cleantech operator. His co-founder and chief technology officer built the prototype.
See what it covers before you sign up: the pattern worth copying, the channel to test, and the risk to avoid when you adapt this story.
$100K ARR
Other
Action checklist
See what it covers first. Create an account only if you choose to unlock.
Inspired by Matteo's journey? Generate a business idea in the Health & Wellness space using AI and real founder data.
Matteo achieved 2 milestones on the path to $100K ARR
$100,000
The journey, decisions, and context behind this milestone
See the complete breakdown: launch strategy, validation methods, startup costs, expert analysis, replication playbook, and more actionable insights.
Upgrade to PremiumInstant access to all founder journeys
Founders with similar journeys or strategies
My journey took me from being a Paris waiter to an $80,000/month solopreneur over seven years of persistence. After 17 failed projects, I found succes...
After selling my previous AI company Headlime for seven figures, I took time off in 2021. I was growing increasingly bored when an idea struck me: why...
We started Pocketed during the pandemic to help startups find grants and funding they didn't know existed. The Idea During COVID, there was sudd...
Get more founder journeys like this delivered to your inbox every week.