Antony Jenkins spent five years as CEO of Barclays, one of the world’s largest banks, trying to modernise its technology from the inside. He failed. The legacy systems were too deeply embedded, the organisational resistance too strong, and the regulatory constraints too tight to allow a wholesale rebuild while the bank was operating. So in 2016, he left and founded 10x Banking, a company that builds from scratch the cloud-native core banking platform he could not create at Barclays. That trajectory, a senior banking executive concluding that the next generation of banking technology must be built outside the banking system, tells the story of where banking platforms are headed. The global banking-as-a-service market reached $18.6 billion in 2024, according to Global Market Insights, growing at 15.1% annually toward $73.7 billion by 2034.
What Defines a Next-Generation Banking Platform
A next-generation banking platform has four characteristics that distinguish it from the mainframe systems it replaces.
First, it is cloud-native. It runs on public cloud infrastructure (AWS, Google Cloud, Azure) and uses cloud services for storage, computing, and networking. It does not run in a private data centre on hardware the bank owns. Cloud deployment accounts for 67% of the BaaS market, per Global Market Insights, because cloud infrastructure scales with demand, reduces capital expenditure, and eliminates hardware maintenance.
The Boston Consulting Group projects fintech revenues will reach $1.5 trillion by 2030, with embedded finance and digital lending accounting for the largest share of projected growth.
According to CB Insights’ 2024 fintech report, global fintech funding declined 40 percent between 2022 and 2024, pushing the sector toward consolidation and a sharper focus on profitability over growth at all costs.
Second, it processes transactions in real time. Legacy core banking systems accumulate transactions during the day and process them in overnight batch runs. Account balances update once per day. Next-generation platforms process each transaction as it occurs, updating balances within milliseconds. This is not a feature preference. It is a requirement. Customers using real-time payment networks (Pix, UPI, SEPA Instant) expect their balances to reflect payments instantly.
Third, it is API-first. Every function the platform performs, from opening an account to processing a payment to generating a regulatory report, is accessible through an API. This allows banks to integrate with third-party services, build new products by combining existing functions, and expose data to authorised partners through open banking frameworks. Banks globally process over 2 billion API calls daily, handling $676 billion in transaction value, per Coinlaw.
Fourth, it uses a microservices architecture. Rather than a single monolithic application, the platform consists of hundreds of small, independent services (payments, accounts, cards, compliance, notifications) that communicate through APIs. Each service can be updated, scaled, or replaced independently without affecting the others.
The Companies Building Next-Generation Platforms
A handful of companies dominate the next-generation core banking market. Each takes a slightly different approach.
Thought Machine, founded in 2014 in London, built Vault on Google Cloud. Its defining feature is a smart contracts engine that allows banks to define financial products (savings accounts, loans, credit cards) as code. Changing a product’s interest rate calculation, fee structure, or eligibility rules requires changing code, not database configurations. Standard Chartered, Lloyds Banking Group, and Intesa Sanpaolo are among its clients.
Mambu, founded in 2011 in Berlin, built a composable banking platform on AWS. Its architecture allows banks to activate only the modules they need (lending, deposits, payments) and add others over time. Mambu positions itself for mid-market banks and fintech companies that want flexibility without building from scratch. N26, ABN AMRO, and Raiffeisen are clients.
10x Banking, founded in 2016 by Antony Jenkins, built a platform specifically designed for the largest global banks. Its architecture handles the regulatory complexity and transaction volumes that tier-one institutions require. Westpac and Chase UK are among its deployments.
Temenos, the oldest of the group (founded 1993), offers both a traditional on-premise product and a newer SaaS version. With over 3,000 bank clients globally, Temenos has the largest installed base, though its legacy product competes with its own next-generation offering.
How Banks Migrate to Next-Generation Platforms
Migration is the hardest part. A bank cannot shut down its existing systems, install a new platform, and restart. Every account, every standing order, every loan, and every regulatory obligation must transfer without interruption.
Three migration patterns have emerged. The big-bang approach migrates everything at once over a weekend or holiday period. This is the fastest but riskiest option. TSB’s 2018 IT migration, which locked 1.9 million customers out of their accounts, demonstrated what happens when a big-bang migration goes wrong.
The parallel-run approach runs old and new systems simultaneously for a period, processing transactions on both and comparing results. Once the new system has proven accurate, the old system is decommissioned. This is safer but doubles infrastructure costs during the overlap period.
The progressive migration approach moves customer segments or product categories one at a time. New customers go on the new platform immediately. Existing customers migrate in batches over months or years. This is the slowest but lowest-risk approach and has become the most common among large banks undertaking core modernisation.
The Economic Case for Migration
The neobanking market, where next-generation platforms run the entire bank, reached $210.16 billion in 2025, per Fortune Business Insights, growing at 49.30% annually. Neobanks built on modern platforms operate at cost-to-income ratios of 30% to 45%. Traditional banks on legacy systems operate at 55% to 70%. That gap translates directly to pricing power (neobanks can offer lower fees and higher savings rates) and margin (for those that choose to retain the savings).
Speed to market is another economic advantage. A bank on a next-generation platform can design, build, and launch a new savings product in four to six weeks. The same product on a legacy system takes 12 to 18 months. Over five years, that speed difference compounds into a product catalogue that is far broader and more responsive to market conditions.
The cross-border payments market shows the revenue opportunity that next-generation platforms enable. It reached $371.59 billion in 2025, per Fortune Business Insights. Banks on modern platforms can connect to local payment networks in new markets through API integrations, capturing cross-border revenue that legacy systems cannot efficiently process.
What Comes After Current Next-Generation Platforms
The platforms being deployed today are not the final form. Three developments are likely to shape the generation after.
AI-native banking platforms will embed machine learning models into core transaction processing, not as separate add-ons but as part of the platform itself. Every transaction will be assessed for fraud risk, compliance risk, and customer intent as it processes, rather than screened after the fact.
Programmable money, enabled by central bank digital currencies (CBDCs) and tokenised deposits, will allow banking platforms to execute conditional payments automatically. A platform could hold funds in escrow and release them when a delivery confirmation API call is received, or split a payment across multiple recipients according to a smart contract.
Interoperable platforms that communicate with each other through standardised protocols (rather than proprietary APIs) will reduce the cost and complexity of bank-to-bank integration. The current generation of platforms still requires custom integration work for each new connection.
The next generation of banking platforms is already here, deployed at neobanks and an increasing number of traditional institutions. The generation after that is being designed now, in the engineering teams of the companies building today’s platforms and in the research labs of central banks experimenting with programmable money.
