Author: Francesco Andreoli
Compiled and organized by BitpushNews
Everyone is counting funding rounds, but no one is counting the fallen "graveyards." So I did both.
Six months ago, I began counting the number of neobanks, because I realized no one could accurately say how many neobanks were still operational. Industry analysts who sell PDF reports for $4,000 couldn’t tell me, the venture capitalists (VCs) funding them couldn’t tell me, and the founders competing with them couldn’t tell me either.
As of July 2026: 368 verified active digital banks. I track each one; the data is public and available at neobankbeat.com.
But the number that truly changed the way I view this industry isn't 368—it's the lists I had to delete to arrive at that number.
368 verified active digital banks
Rise is real. Let’s start here.
Summing up the user numbers self-reported by each company in the dataset: approximately 1.46 billion people worldwide are using these digital banks we track. This is not a market forecast or a TAM figure on a PowerPoint slide—it’s actual reported customer numbers.
Moreover, its geographic distribution will surprise anyone who reads only Western fintech media:
Of these 1.46 billion users, 817 million are in Asia. Just WeBank alone serves over 400 million people—more than the combined user base of all digital banks in the U.S. and Europe.
Nubank has 131 million customers, more than the total number of users of all digital banks in the United States.
Revolut, the jewel in Europe's crown, has over 50 million users. While impressive, this pales in comparison to the numbers in Asia.
The marginal vitality of this industry has also shifted. Of the digital banks founded and still operating in the 2020s, 30% are Web3-native self-custody applications where no company holds your account balance; in contrast, only 4% of companies from the 2010s operated this way. Regardless of your view on cryptocurrency, entrepreneurs have already voted with their feet.
So yes: the rise is real. 368 companies, divided into three distinctly structured waves (254 traditional challenger banks, 58 hybrid fiat/crypto apps, and 56 Web3-native apps), are supported by a foundation of 106 infrastructure providers, backed by 219 investors—all laid out clearly on the map.
Now, we get to the part that no one ever includes in a fundraising pitch deck.
Of these 368 companies, only 127 hold full banking licenses.
Read that sentence again. Two out of three “banks” you see in app stores aren’t banks at all. They’re renting their right to operate from sponsor banks, e-money license holders, or issuing institutions you’ve never heard of. Their customers almost never know exactly which side of the line they’re on.
Banking license
This is not a technical detail. It is a core structural risk across the entire industry, and it has already accumulated a chilling list of casualties:
WaveCrest (2018): Visa cut off a card issuer, causing dozens of cryptocurrency card projects to die overnight.
Wirecard (2020): A payment processor revealed a €1.9 billion financial hole, leading to the freezing of banks that had simply built their services on its platform.
Synapse (2024): The collapse of a banking-as-a-service intermediary revealed to ordinary Americans that “FDIC insurance” was not what they thought—because the ledger tracking who owned what money was where the failure occurred.
Ready (2026): Same script, new lead.
When a real bank fails, deposit insurance pays out; when the infrastructure of a digital bank collapses, customers receive nothing but a number in the queue for bankruptcy proceedings.
The death of this industry is silent. That is the real bad news.
This is exactly what I hadn’t anticipated while maintaining this dataset: the work of removing names has never stopped.
Who is in control?
This month alone, five entities were removed from the list through delisting, mergers, or quiet transitions into other businesses—no press releases, no post-mortem reports. Digital banks don’t collapse in the dramatic fashion of FTX. The apps simply stop updating, customer service stops responding; then one day, the domain redirects to a partner’s homepage, and hundreds of thousands of customers either migrate or vanish.
No one writes obituaries for digital banks. Fintech media only report on launches and funding rounds because that’s where the advertising dollars and exclusive access lie. As a result, this graveyard remains invisible, and every new founder who walks in steps into the same five traps, believing they’re the first to discover them.
That’s why we meticulously track exiters just as we do entrants. The value of failure data exceeds that of funding data. You learn nothing from press releases.
“But AI will fix the economic model, right?” Will it really?
Every digital banking pitch deck now claims to use AI. Therefore, we audited all 368 companies individually, cross-referencing regulatory filings, regulatory disclosures, and real production environment evidence—rather than marketing pages.
Only 67 passed the audit—just 18%. The remaining 300+ are either still testing, exploring, or claiming partners’ models as their own.
AI digital bank?
More interestingly: the leaders in AI are not the big-name players you’d expect—they’re credit institutions in emerging markets like Nigeria, the Philippines, Mexico, and Bangladesh, where credit bureaus are useless; being able to build risk models for borrowers with no credit history isn’t a “gimmick”—it’s the only reason these businesses can survive. The West talks about AI banking; the Global South is delivering it, because they have no other choice.
Look at the infrastructure segment on the chart: 106 providers power 368 consumer-facing brands. Within that segment, a small number of sponsoring banks, BaaS platforms, and card processors each support dozens of brand logos above them. This high level of concentration, invisible to consumers, is precisely the breeding ground for the next Synapse.
This is the real picture of this industry in 2026: on one hand, an extraordinary and tangible rise that is changing the world—1.5 billion people gaining access to banking services through applications, many for the first time in their lives.
On the other hand, all of this is built on an underlying infrastructure that most customers have never heard of, and two-thirds of the companies building these digital banks couldn't withstand a single quarter of performance decline from their upstream providers.
Both of these things are true at the same time. That’s what makes it interesting.
My three predictions that I’m ready to be proven wrong on
The licensing gap will narrow from both ends. Well-capitalized unlicensed players will acquire or win banking licenses; weaker players will end up on the "deleted list" by 2027. The middle ground will vanish.
The first AI risk control failure will occur within two credit cycles. Most of these 67 production models have never experienced a real economic recession in their current form. Some will soon pay the price to learn things never included in their training data.
The next wave of customers isn't human. Seven companies are already building the banking rails for AI agents—wallets operated by agents, cards issued by agents, machine-to-machine payments. It looks just like Web3 native did in 2021: tiny, quirky, yet profoundly structurally transformative.
All of the above conclusions can be reproduced from public datasets. No paywalls, no email locks, based on the MIT open-source license:
Complete ecosystem map (including all logos and names):neobankbeat.com/blog/neobank-ecosystem-map/
Current State of Digital Banking Report (Monthly In-Depth Analysis): neobankbeat.com/report/
Original data: neobankbeat.com/data.json
Twitter: https://twitter.com/BitpushNewsCN
BitPush Telegram community: https://t.me/BitPushCommunity
BitPush TG subscription: https://t.me/bitpush