BYD sold the most EVs and profit fell 19%. CATL made more selling batteries. 6 EV startup case studies on where electrification actually pays.
6 EV Innovators Who Are Reshaping the Future of How We Move
Case 1
Wang Chuanfu built the biggest electric car company in the world. Even with record sales, his profit fell 19%. Robin Zeng, who makes the batteries inside those cars, earned more than twice as much. Six EV startups, and one strange pattern underneath all of them: where the real money in electric cars actually ends up.
Electric cars were supposed to shift money away from gas engine makers, and toward the carmakers instead. In 2025, that's not quite what happened. The money skipped past the carmakers entirely. The battery maker out-earned the car company by more than two to one. The most well-funded American company in this space made its first real profit not by selling cars, but by selling its engineering knowledge to others. And the two companies that built stations where drivers can swap a dead battery for a full one are still paying off the cost of building them.
The numbers back this up. BYD sold 4,602,436 new energy vehicles in 2025, more than any other company on earth, with revenue up 3.46% to 803.97 billion yuan. Yet profit still fell 19%, down to 32.62 billion yuan, because the company kept less from each sale; its margin narrowed from 19.44% to 17.74%. CATL, the company that makes BYD's batteries, earned RMB 423.7 billion, up 17%, and RMB 72.2 billion in profit, up 42%, on roughly half of BYD's revenue, while holding 39.2% of the entire global battery market for a ninth year in a row. NIO turned its first profitable quarter in Q4 2025, earning 282.7 million yuan, right after a full year of losses totaling 14.9 billion yuan, itself a 33.3% improvement on the year before's 22.4 billion yuan loss. And Fisker Inc., after building 5,000 Ocean SUVs and delivering just over 900 of them by September 2023, filed for bankruptcy on June 17, 2024, the second time the founder's company has shut down this way.
Every number here is real, backed by actual financial records. These six electric vehicle startups are ranked by one simple thing: how close, or how far, each founder sits from the physical car itself. And here's the twist. The farther away they get, the better their returns look- the real lesson inside EV engineering, EV innovations, and smart EV startup thinking, worth understanding before making any costly unit economics mistakes of your own in this space.
Your key takeaways are:
The six companies in this article are lined up in a specific order, based on how close each one actually gets to building the finished car. It starts with BYD, a company that makes almost everything itself, the batteries, the parts, the car. It ends with Fisker, a company that designed a car but paid someone else to actually build it. Read them in that order, and something interesting shows up. The further away a founder gets from physically building the car, the better their business results tend to look.
Here's the part that might surprise you most. The company making the most money per dollar of sales in this whole group isn't a car company at all. It's the battery supplier. CATL grew its profit by 42% in the very same year BYD's profit dropped 19%. That single comparison is really the clearest example in this entire article.
And that pattern sets up everything that comes next. The real, lasting profit in electric cars hasn't ended up in the car itself. It's sitting further back, in the battery cell and in the overall design, not in the finished car sitting in someone's driveway.
Case Study #1: Wang Chuanfu, BYD
The battery maker who built a car company around his own cells
Snapshot
Founder | Wang Chuanfu |
Company | BYD (founded 1995, Shenzhen, China) |
Model | Vertically integrated battery, component and vehicle manufacturer |
Peak metric | 4,602,436 new energy vehicles sold in 2025 |
Philosophy | Own every layer that touches the cell |
In 1995, BYD was just a small battery company in Shenzhen, China. By 2025, it had become the biggest electric car maker in the entire world, selling more cars than anyone else. That jump is the whole story here, before we even get to the surprising twist about its profits.
Starting out as a battery company and then trying to build cars is incredibly hard. There's no network of dealers selling your cars. No brand people already trust. No history of crash tests proving your cars are safe. No relationships with the companies that supply car parts.
The usual path here would be simple: sell your batteries to the big, established car companies, and accept whatever smaller cut of the profit they're willing to give you, forever. That's the ceiling most ev startups quietly accept, without ever pushing past it.
Wang decided to do something different. He kept the battery, and built everything else around it himself, keeping the profit at every single step, instead of handing it over to someone else's company.
The scale he reached was huge. In 2025, revenue reached 803.97 billion yuan, up 3.46% from the year before. Sales hit 4,602,436 new energy vehicles, up 7.73%, and cars shipped overseas reached 1.05 million units, 1.4 times more than the year before. Those numbers put BYD ahead of nearly every other name among electric vehicle startups anywhere in the world.
BYD also spent 63.4 billion yuan on research and development, up 17%, nearly double what the company kept as actual profit for the entire year. It's a business growth strategy built entirely on owning every single piece of the puzzle, instead of relying on someone else's company for any of it.
Results: Highest-Volume Manufacturer in the World, With Falling Margin
Thirty years passed, from 1995 to 2025, turning a small parts business into the world's biggest electric vehicle maker, covering everything from batteries and computer chips to motors, assembly, and even shipping cars overseas.
But here's the twist. Record sales and record revenue showed up right alongside a 19% drop in profit, exactly what happens when a company grows huge mostly by cutting its prices.
Lessons & Playbook
Owning the most expensive part of your product gives you the strongest position in any hardware business.
Building everything yourself turns someone else's profit into your own profit, but it doesn't protect you from having to lower your prices.
Selling the most cars, and making the most money, are two completely different races, and a company can win one while losing the other.
The Bitter Truth: The 2025 numbers tell the story of a company that grew right into a price war it helped start in the first place.
Net profit fell 19%, down to 32.62 billion yuan, and the company kept less money from each sale too, as its margin shrank from 19.44% down to 17.74%.
BYD's own yearly report actually explains why, in plain language. It says ongoing price wars and intense competition are squeezing car companies' profits, BYD describing, in its own official documents, the exact same market conditions it did more than any other company to create.
BYD sold more cars than ever, and still made 19% less money doing it. That's the real warning here. A company can look successful just by looking at how much it sells, while actually getting weaker underneath.
PrometAI helps founders build financial plans that watch both things at once, how much you're growing, and how much you're actually keeping, so a story like BYD's doesn't sneak up on you too.
Case 2
Case Study #2: Robin Zeng, CATL
The supplier who let everyone else build the cars
Snapshot
Founder | Robin Zeng |
Company | CATL, Contemporary Amperex Technology (China) |
Model | Battery cell and energy storage manufacturer supplying the industry |
Peak metric | 39.2% global power battery market share, ninth consecutive year first |
Philosophy | Sell to every carmaker rather than compete with them |
Out of every five electric car batteries on earth, roughly two of them came from one single company.
That kind of power wasn't the starting point though. A company that makes battery cells sits in a tough spot. On one side, it needs raw materials it has no control over. On the other side, it's selling to giant car companies big enough to demand whatever price they want.
Normally, that squeeze leaves a company with thin profits and constant price pressure coming from both directions at once. It's the exact trap most companies working in ev engineering never manage to escape.
Getting big enough changed everything. Once one single company makes roughly two out of every five batteries sold worldwide, something interesting happens. The usual power dynamic flips. Instead of big car companies pushing CATL around, CATL becomes too important to push around at all.
The numbers show just how far that flip went. In 2025, revenue reached RMB 423.7 billion, up 17% from the year before. Profit hit RMB 72.2 billion, up 42%. Battery sales climbed to 661 GWh, up 39%.
CATL held the world's largest share of the battery market, 39.2%, for the ninth year running. On top of that, its energy storage business, the batteries used to store power for the electric grid, made up 30.4% of the entire global market, ranking first in the world for five years straight. Very few companies working on EV innovations have ever gotten anywhere close to numbers like these.
Results: Profit Growing Two and a Half Times Faster Than Revenue
In 2025, profit grew two and a half times faster than revenue, 42% growth in profit compared to just 17% growth in revenue. When profit outpaces revenue by that much, it usually means one thing: the company can charge more, and customers have no real choice but to pay it.
CATL led the world in battery market share for the ninth year in a row, and led the world in energy storage shipments for the fifth year in a row too.
Lessons & Playbook
Supplying an entire industry can actually beat trying to compete inside it directly, and by a wide margin.
Once your share of a market gets big enough, the usual power balance between buyer and seller flips completely.
When profit grows faster than revenue, that's the clearest sign a company finally has real pricing power.
The Bitter Truth: The exact same position that creates those huge profits is also the position that governments start paying close attention to. When one company supplies a critical part to most of an entire industry, all from a single country, the more successful it becomes, the more political attention it draws.
At the same time, every customer has a real reason to fund a competitor, precisely because CATL is doing so well. That 42% profit growth is actually the strongest reason any car company could give itself internally to start building its own batteries instead. Being this dominant gives you the money to defend yourself, but it also gives everyone else a reason to come after you.
CATL's dominance shows how business success and real strategic risk can grow out of the exact same set of numbers.
PrometAI helps founders map out this kind of risk, relying too heavily on one customer or one market, into their business plan, before a customer's reason to build a competitor turns into a real threat.
Case 3
Case Study #3: William Li, NIO
A decade of losses to prove a charging model nobody else would fund
Snapshot
Founder | William Bin Li |
Company | NIO (China), multi-brand electric vehicle maker |
Model | Premium electric vehicles with a proprietary battery swapping network |
Peak metric | First profitable quarter in Q4 2025, net profit of 282.7 million yuan |
Philosophy | Infrastructure first, even when it is the expensive answer |
One quarter, NIO lost 7.1 billion yuan. A year later, in that exact same three-month stretch, it earned 282.7 million yuan instead. That turnaround is the good part of the story. Getting there was much harder.
Most electric cars get charged by plugging them in, which is simple, and any charging company can offer it without much extra work. NIO chose something completely different. It built stations where drivers swap out their empty battery for a full one, in just a few minutes.
But that idea needs identical battery packs across all your cars, stations you build and own yourself, and a huge amount of money spent years before you have enough cars on the road to make it worthwhile. Almost every competitor picked regular charging instead, since it needs none of that setup.
So William Li's company, NIO, ended up paying for this entire extra system almost completely on its own, a bet most ev startups would never even attempt to make.
Here's the key thing about a battery-swapping network. It only starts making financial sense once enough people are actually using it. That meant the real fix for NIO's huge losses wasn't to give up on the idea. It was to sell more cars.
And that's exactly what happened. Across the whole year of 2025, NIO delivered 326,028 vehicles total, across its three different car brands, up 46.9% from the year before. Total revenue reached about 87.5 billion yuan, with 76.9 billion yuan of that coming from actual car sales, up 32% compared to 2024.
The turn into profit happened fast, once enough cars were finally on the road. In the fourth quarter alone, NIO delivered 124,807 vehicles, up 71.7%. That quarter brought in a net profit of 282.7 million yuan, compared to a 7.1 billion yuan loss in that exact same quarter the year before, proof that the expensive infrastructure bet was finally starting to pay for itself.
Results: A 7.1 Billion Yuan Quarterly Swing Into Profit
Going from a 7.1 billion yuan loss in one quarter, to an actual profit in the same quarter a year later, is the clearest sign yet that this battery-swapping idea can actually work, once it reaches a big enough scale. It just took years of spending real money first to get there.
Deliveries across all three brands grew 46.9% year over year, in the very same year the company finally turned its first profitable quarter, a milestone very few EV startup founders ever get to see after losing this much money for this long.
Lessons & Playbook
Infrastructure-heavy businesses lose money until enough people are actually using what they built, and then everything changes.
Owning your own special system protects you as long as people keep using it, but becomes a burden the moment they stop.
Making a profit in one quarter and making a profit across a whole year are two completely different achievements.
The Bitter Truth: The full year still ended with a net loss of 14.9 billion yuan, though that was actually 33.3% better than the 22.4 billion yuan lost the year before. Two years in a row of losses that huge, stacked on top of everything that came before them, is the real cost of betting on infrastructure like this. One good quarter doesn't erase that.
The core risk hasn't gone away either. A special network like this is only valuable as long as the cars using it keep selling well. The moment they stop, that same network turns into a costly burden instead. And Li has already promised to keep spending more money on this network in 2026, which is consistent with his approach, and expensive too.
NIO's years of accepted losses, before finally hitting one profitable quarter, show exactly why business plans built around heavy infrastructure need to plan for many years ahead, not just quarter by quarter.
PrometAI helps founders figure out exactly how long a big, expensive bet like this needs to run, before enough people actually start using it to make it worthwhile.
Case 4
Case Study #4: Horace Luke, Gogoro
He solved battery swapping at scale, on two wheels, and left under investigation
Snapshot
Founder | Horace Luke |
Company | Gogoro (Taiwan), electric two-wheelers and swapping network |
Model | Vehicle sales plus battery swapping subscription |
Peak metric | More than 800 million cumulative battery swaps |
Philosophy | Match the swapping model to the vehicle that suits it |
He built the one battery-swapping network on earth that actually turns a profit. Then he had to resign, under investigation.
Getting there meant tackling a problem almost nobody else wanted to touch. Across most of Asia, scooters, not cars, are how people actually get around every single day. But almost none of the money pouring into electric vehicles went toward scooters. There was another problem too.
Scooter riders can't just plug in and wait around for hours to charge. They're riding constantly, all day, and they have nowhere to park a scooter overnight next to a charger. It's exactly the kind of unglamorous, overlooked problem most of the money in ev engineering skips right past, the kind an ev engineer with no funding still has to solve anyway.
Gogoro's big idea was simple, but clever. Separate the battery from the scooter completely. Then, instead of just selling the scooter once, sell access to a whole network of swap stations, charged monthly, like a subscription. Suddenly, a one-time hardware sale turned into steady money coming in every single month.
And the network really took off. It passed 800 million total battery swaps, across more than 2,700 swap stations, serving nearly 700,000 riders. By 2025, total subscribers had grown to 665,000.
Then came the real proof it was working. Adjusted profit hit a record $59.9 million in 2025, up from $44.7 million the year before, making Gogoro the only battery-swapping network of its kind anywhere in the world that's actually profitable at this level. It's the kind of steady, recurring revenue most vehicle companies never even come close to building.
Results: The Only Profitable Swapping Network at Scale
665,000 subscribers. 800 million total swaps. It's the biggest, most real-world proof that battery swapping can actually work, for any kind of vehicle, anywhere.
Adjusted profit grew 34% year over year, from $44.7 million to $59.9 million, even though total revenue for the year actually dropped 9.4%.
Lessons & Playbook
The exact same idea can work brilliantly on one type of vehicle, and completely fail on another.
Charging people a monthly subscription makes it much easier to fund expensive infrastructure than just selling vehicles ever could.
Being profitable on paper, and truly profitable overall, are two very different things when your business depends on real, physical equipment.
The Bitter Truth: Luke stepped down from every leadership role he held, CEO, Chairman, and Director, effective September 13, 2024. Gogoro admitted it had found problems in its supply chain that led it to accidentally use imported parts in some of its scooters, breaking a Taiwan government rule tied to which vehicles qualify for purchase subsidies.
The money problems haven't gone away either. Full year 2025 revenue fell 9.4%, down to $281.5 million, with a net loss of $80.8 million, though that loss was smaller than the $122.8 million lost the year before. A business built heavily around subsidy-eligible sales found out the hard way that qualifying for those subsidies is something you're legally required to get right, not just a nice selling point.
Gogoro did almost everything right. It built a real network. It made real money. It grew a huge number of loyal subscribers. But one small rule, about which parts could legally qualify for government subsidies, ended up costing the company its founder and its trust.
PrometAI helps founders catch problems like this early, by checking for risks like this while building the business plan, not after the damage is already done.
Case 5
Case Study #5: RJ Scaringe, Rivian
The American entrant whose first gross profit came from software, not vehicles
Snapshot
Founder | RJ Scaringe |
Company | Rivian Automotive (Irvine, California) |
Model | Electric trucks and vans, with a software joint venture |
Peak metric | First full year of positive consolidated gross profit, $144 million in 2025 |
Philosophy | Build the vertically integrated technology, then sell it twice |
$144 million in profit. That sounds like a big win, right up until you find out where it actually came from.
Building cars from scratch is brutally expensive at first. Costs only start dropping once you're building a lot of cars, but you can't afford to build a lot of cars until your costs come down. It's a chicken-and-egg problem, and the only way through it is with huge amounts of borrowed or invested money, billions of dollars worth.
It's the exact same wall almost every one of these electric vehicle startups slams into, long before they ever reach real scale.
Rivian found a clever way around this. It took the wiring system and the software running inside its trucks, the invisible stuff most people never think about, and turned that into a product of its own. It sold that technology to Volkswagen, through a strategic partnership between the two companies.
That partnership paid off fast. Software revenue jumped to $1,557 million, with $576 million kept as profit, compared to just $7 million the year before. Almost all of that jump came from the Volkswagen deal. Rivian's total revenue reached $5,387 million in 2025, up 8% from the year before.
Add everything together, and Rivian's total profit reached $144 million for the whole year, a swing of more than $1.3 billion compared to 2024. That turned a company that had always lost money into one showing an actual profit on paper, a milestone very few EV startups ever reach, let alone through a shortcut quite like this one.
Results: A Positive Gross Line Funded by a Partner's Budget
For the very first time in the company's history, Rivian's total profit turned positive, an improvement of more than $1.3 billion in just one year, almost entirely thanks to that software deal with Volkswagen.
In fact, the software business alone made $576 million in profit, more than Rivian's entire company-wide profit of $144 million.
Lessons & Playbook
Technology you build for your own product can quietly become a whole second product you can sell separately.
A company's total profit can hide a struggling core business underneath it, so it always pays to look closer, one piece at a time.
Money from government environmental credits is real, but it depends on political decisions, not on anything the company controls.
The Bitter Truth: Now take the software money away, and the whole picture flips. Car sales alone actually fell 15%, down to $3,830 million, and the car business itself lost $432 million for the year. In just the last three months of the year, that car business swung from a $110 million profit the year before, to a $59 million loss, largely because the company earned $270 million less from selling environmental credits than it had before.
For the whole year, Rivian lost $3.64 billion overall. It delivered 42,247 vehicles in 2025. And a company selling forty thousand vehicles while losing money on every single one is leaning heavily on a partner's budget, and on a government credit system it has no real control over.
Rivian's total profit number looked good on the surface. But underneath it, the actual car business was still losing money. One good part of the company was covering up problems in another part.
PrometAI helps founders build financial plans that check each part of the business honestly on its own, so one strong piece can't quietly hide a weak one.
Case 6
Case Study #6: Henrik Fisker, Fisker Inc
A designer who reached production twice and liquidated twice
Snapshot
Founder | Henrik Fisker |
Company | Fisker Inc (La Palma, California) |
Model | Contract-manufactured electric SUV sold direct to consumers |
Peak metric | 5,000 Ocean SUVs manufactured by September 2023 |
Philosophy | Outsource manufacturing and compete on design |
Twice, he actually got a car company all the way to building real cars. And twice, that same company had to shut down and sell everything off because it ran out of money. Same founder both times, same name on the company both times.
Building your own car factory costs an enormous amount of money. So instead, Fisker took a different route. It paid another company, called Magna, to actually build its cars for it, using Magna's own factories.
That saved Fisker from having to spend billions building a factory of its own. But it also meant Fisker's success now depended completely on a partner it didn't fully control, a trade-off plenty of ev startups make, without always thinking through what that might cost them down the road.
Fisker actually did something most electric vehicle startups never manage to do. It built a real, finished car, and delivered it to real, paying customers, on two different continents.
By September 26, 2023, the company had built 5,000 Fisker Ocean SUVs, and delivered more than 900 of them to customers, in both Europe and the US, real, solid proof that its EV innovation had actually reached real buyers, not just prototypes sitting in a warehouse.
The top model, called the Ocean Extreme, cost $68,999, and could drive an estimated 360 miles on a single charge. The cheaper Sport model cost $37,499. The company hoped to eventually deliver 300 vehicles a day, across both the US and Europe, later that same year.
Results: Production Reached, Delivery Rate Never Sustained
5,000 vehicles built, and more than 900 delivered by September 2023, across two continents. That's a real accomplishment most EV startups never even come close to reaching.
But that big goal of 300 deliveries a day never actually happened at that scale. Nine months later, the company was in bankruptcy, a gap worth understanding closely, so you don't end up making the same startup financial model mistakes yourself.
Lessons & Playbook
Paying someone else to build your product saves you money upfront, but it also hands them real control over your business.
Betting everything on just one car model, built by one outside manufacturer, leaves you no backup plan if sales don't show up as expected.
Announcing a big goal isn't the same as actually reaching it, and that gap is exactly where all the money quietly disappears.
The Bitter Truth: On June 17, 2024, one part of the company, Fisker Group Inc., filed for something called Chapter 11, a legal process companies use when they can't pay their debts and need court protection while they figure out what to do next. Two days later, the rest of Fisker's US companies filed too, and all of it was handled together under one court case.
By that point, the company's stock had already dropped to something called OTC Pink, a much smaller, less regulated way of trading stock, since it no longer qualified for a real stock exchange.
Starting October 17, 2024, the company followed a wind-down plan, basically a plan for shutting everything down in an organized way, funded partly by selling off about 3,300 leftover vehicles to a company called American Lease LLC, for up to $46.5 million.
A company that had actually built real cars on two continents ended up being wound down by simply selling off whatever cars it had left. This was the second time a company with this exact founder's name had reached the market, and then gone bankrupt.
Fisker's second bankruptcy, under the very same founder's name, is a reminder that actually building a working product proves you can do it, but it doesn't prove the business itself can survive.
PrometAI helps founders map out the full path from building a product, all the way to selling it at real, sustained volume, so a big announced goal is always backed by a real financial plan behind it.
Conclusion: Where Electrification Actually Pays
Put these six stories side by side, and a clear pattern shows up. The electric car itself, the actual vehicle, has become the least profitable place to stand in this whole industry.
The company that sat furthest away from actually building the car earned 72.2 billion yuan. Meanwhile, the company that sold the most cars of anyone on earth earned 32.62 billion, and that number actually went down. The American company in this list only turned a profit because of money it earned from a partner, not from selling its own cars. And the two founders who built battery-swapping stations had to fund all of that expensive equipment using money they'd actually lost, for years.
Then there's the one company in this whole list that only made cars, run by a designer who didn't even own a factory. That one ended in a shutdown.
None of this means making cars is a bad idea. It's really just an observation about where the real, lasting profit in electric vehicles has ended up. It's not in the finished car sitting in someone's driveway. It's further back, in the battery itself, and in the underlying design.
These founders didn't just build EV startups. They built real businesses that had to survive real financial pressure: shrinking profit margins, the huge cost of building infrastructure, following government rules, and losses hidden inside numbers that looked fine on the surface. That's the kind of pressure test every serious name in electric vehicle startups eventually has to face.
Are you building an EV or hardware startup of your own? Whether you're deep into EV engineering, chasing your next big EV innovation, or just trying to understand where the real money in EV innovations actually ends up, PrometAI helps founders figure out where the real profit sits in their own industry, before they run out of money trying to prove it themselves.
Start by building a real startup financial model, one solid enough to hold up under the same scrutiny BYD's, CATL's, and Rivian's numbers just went through.
If you're planning to raise money to fund that model, it helps to understand how to get venture capital funding before you actually need it in a hurry. And keep watching the bigger trends in entrepreneurship shaping where profit moves next, because in electric vehicles, it clearly didn't end up where anyone expected.