Your data is no longer a moat. What you do with it is.

• Open banking and AI integrations allow customers to export their complete financial history in seconds, handing competitors a view your bank spent decades building.
• Every linked account is a warning your dashboards don’t show, putting deposits and primacy at risk as Big Tech and AI adoption accelerates.
• Passively warehousing data is no longer enough; banks must use live account signals to resolve customer needs proactively before they turn to outside solutions.
• You can launch this operational defense immediately without a core overhaul by auditing external connections and deploying targeted interventions on existing rails.

Every bank sits on an asset no fintech or frontier AI model can build from scratch: years of paychecks, bills, savings goals, and life events. It took decades of trust to build, and it's a powerful competitive advantage. But it has a shelf life.

Consumers now dictate where their financial intelligence flows, and a growing number are diverting it directly to your competitors. Every linked external account is an early warning of silent attrition and lost wallet share. Passive data retention no longer guarantees your primary bank status.

To stop the outward flow of this data, you must deliver the utility these competitors offer, but from a position they cannot match. This means pairing the intelligence you hold with the products you offer to deliver immediate financial value. When you sweep an idle balance into a CD, or structure a timely line of credit to prevent a late fee, you do more than save them money. You eliminate their incentive to export their financial life.

Why is customer data no longer a bank's moat?

Because data is now entirely portable. Open banking frameworks, aggregators, and AI assistants allow customers to instantly share and aggregate their financial footprint. The multi-decade head start you enjoyed is draining into competitor ecosystems.

Fintechs entered the market with a distinct disadvantage: zero historical financial data. Now your customers are fixing that for them, one linked account at a time. Where open banking rules apply, banks must share raw data and while receiving no reciprocal intelligence back. In the US, aggregators and AI assistants extract this exact data via explicit customer consent. The outcome is identical: your challenger applies the intelligence your bank failed to provide, and the money quietly drains from the account.

Three trends are accelerating this erosion:

• Mass API connectivity: Open banking is no longer theoretical, but a real operational threat. In the UK alone, over 16.5 million accounts are directly connected to third-party providers, transforming your proprietary transaction ledgers into shared commodities.
• Frontier AI integration: OpenAI's personal finance integration allows ChatGPT Pro users in the US to instantly connect accounts across 12,000+ financial institutions via Plaid. These AI assistants are now operating and training on live balances and real-time transaction data.
• Fragmented banking portfolios: Customers don't keep their money in one place. The average US retail bank customer keeps deposit accounts at three institutions (J.D. Power). The aggregator connecting these accounts possesses the macro view, leaving your bank blind to the wider picture.

With every connection, customers transfer your historical advantage directly to your rivals. Each link is a signal that your bank must begin tracking and acting on.

Why do customers share banking data with third parties?

Customers expect a tangible return in exchange for sharing their data. Third-party tools deliver this by transforming complex financial history into high-ROI advice: a better rate, a cheaper plan, or a met savings goal. Most importantly, they make acting on that intelligence frictionless.

With each value-added move, these tools replace your bank as the daily dashboard and step into the role of trusted financial guide. Then, they intercept the high-margin moments like credit decisions and wealth guidance, downgrading your bank to the invisible plumbing behind their sleek interface.

This behavioral shift is accelerating. 53% of US consumers asked an AI tool for financial advice at least once in the past three months (J.D. Power).

This isn't a failure of UX design; it's a structural gap. Customer intelligence sits stranded across legacy cores, credit engines, lending ledgers, and front ends that were never built to talk to each other. Your bank already holds the data; it simply lacks the connected architecture to operationalize it before the customer looks elsewhere.

That gap is creating two competitive openings that fintechs and AI platforms are actively exploiting, but where banks can decisively reclaim the advantage:

• Execute natively on the ledger. While third-party apps analyze cash flow and merely recommend moves, your bank can execute them directly on the ledger. By connecting siloed systems, you can automatically sweep excess cash, adjust payment plans, and optimize balances seamlessly within your own app.
• Offer guidance before the customer asks. Rather than waiting for a customer to visit a mortgage calculator or submit an inquiry, use real-time triggers to intervene with the right product. Anticipate the need for a home loan, credit line, or higher-yield account, and present the solution inside the account while intent is fresh.

How fast are banks losing primacy?

Faster than your current IT roadmap can respond. Primacy dictates where the direct deposit lands and who gets the first call for a mortgage. The velocity of new primary account acquisition favors digital competitors, and these challengers are effectively converting casual users into primary relationships.

• New primary relationships: US online banks and fintechs already control 20% of millennial and Gen Z primary checking accounts, but they captured 28% of new originations. These cohorts now drive two-thirds of all new primary checking openings (J.D. Power, September 2026).
• Primacy at challengers: Revolut's primary-account user base surged 45% in 2025, driving a 66% spike in customer balances to $67.5 billion (2025 annual report). In the UK, Curinos reports that 34% of customers abandoning a Big Seven bank moved to a digital challenger. Crucially, 73% of those switching to challengers were changing primary banks for the very first time since 2020.

Traditional banks still hold the bulk of global accounts. However, the primary status that took decades to win is shifting to digital competitors with every new relationship.

What does losing primacy cost banks?

The cost appears slowly, then all at once. It begins with the signals executive dashboards routinely miss:

1. The authorized link: A customer links their primary checking account to an external budgeting tool, an investment app, or an AI assistant. On internal reports, the relationship appears healthy. The account remains open and the payroll direct deposit still lands twice a month.
2. The silent deposit drain: Accounts stop growing, followed by a quiet drain as automated tools sweep idle cash to better yields. With $23 trillion, roughly a third of all global consumer deposits, sitting in zero-interest accounts, the vulnerability is massive. McKinsey estimates that shifting just 5% to 10% of these balances to competitive yields would slash bank deposit profits by 20% or more.
3. The severed advisory relationship: The customer preference for guidance turns to third-party tools causing your bank to miss lucrative opportunities. Challengers intercept the high-margin opportunities formerly owned by relationship managers, from mortgages and credit cards to wealth management. Meanwhile, your customer view goes stale while their model gets richer.

These quiet defections compound until they surface directly across your P&L:

• Collapsing Non-Interest-Bearing Deposits (NIBD): The zero-cost funding base that anchors retail profitability evaporates as idle balances migrate to higher-yield competitors.
• Spiking deposit betas: Facing tighter liquidity to support the loan book, institutions are forced to raise deposit rates defensively or rely on costly wholesale funding to maintain balance sheet size.
• Compressed Net Interest Margin (NIM): Cost of funds outpaces asset yields, accelerating margin compression and directly eroding Net Interest Income (NII).
• Interchange fee leakage: As day-to-day spending shifts to linked challenger cards and digital wallets, swipe volumes and transaction fee revenues steadily bleed away.

When cheap core deposits vanish and margins compress, the market reprices the institution. Analysts penalize the earnings volatility and the loss of primary customer relationships, contracting your price-to-earnings and price-to-tangible-book multiples. McKinsey estimates that banks failing to adapt their operating models face a $170 billion, or 9%, contraction in global profit pools over the next decade.

What started as a quiet API connection ends as sustained multiple compression on your stock.

How can banks turn customer data into an advantage?

By operationalizing it directly inside the account. Banks must capture behavioral signals, prescribe useful interventions, and execute them natively with explicit customer consent. This applied intelligence, not the data itself, is your new competitive advantage.

Challengers started with a data problem that open banking solved. They earned the right to aggregate financial lives simply by delivering immediate utility. Traditional institutions now face an intelligence problem: they hold the data, but deliver no utility with it.

Yet incumbents retain the decisive advantage: decades of transaction ledgers, the primary settlement rails, and an irreplaceable trust premium. According to McKinsey, 62% of consumers prefer their primary bank for AI-driven financial services, compared to just 19% for big tech. Aggregators and frontier models can read a customer's data and recommend moves, but only your institution has the structural authority to settle, structure, and execute directly on the balance sheet.

According to J.D. Power’s 2025 U.S. Retail Banking Advice Satisfaction Study, 46% of retail bank customers recall receiving financial advice, up from 42% in 2024 and 34% in 2021. Banks are getting better at guidance, yet more than half of customers still can't recall receiving any. That silence is the opening third-party tools step into.

What is proactive banking?

Proactive banking uses real-time transaction data and account activity to anticipate customer needs. It delivers tailored solutions, alerts, and recommendations before issues escalate or requests are filed, operating strictly with customer consent.

To make proactive banking an operational advantage rather than an invasive push, institutions must ground their strategy in four execution principles:

• Treat transaction trails as operational triggers: Stop viewing ledgers purely as retrospective audit trails. Run event-driven rules against streaming cash flows to surface real-time liquidity crunches, idle balances, or borrowing intent before the customer searches for an alternative.
• Prioritize situational utility over product volume: Eliminate blanket cross-sell tactics. Intervene only when an identifiable financial inflection point occurs, whether that means structuring an overdraft-prevention transfer or surfacing a high-yield certificate of deposit.
• Route distress signals to human advisors: Proactivity must read as authentic assistance, never opportunistic upselling. When systems detect indicators of genuine financial strain, the signal should route directly to an empathetic banker who can offer tailored guidance, rather than an automated credit card promotion.
• Close the loop with native, one-tap execution: Third-party aggregators and AI chatbots can analyze spending patterns, but they cannot execute ledger-level changes without friction. When presenting an intervention, allow the customer to authorize and settle the transaction natively within the digital banking experience.

Most enterprise AI initiatives look inward, focusing on back-office automation and employee efficiency. While operational cost reduction protects margins, wallet share erosion destroys them. In the battle for primacy, the intelligence that delivers long-term enterprise value is the intelligence the customer values.

Do banks need a new core to act on customer data?

No. The core can stay the system of record. What most banks are missing is the operational orchestration layer between your back end and your front end, and that can be built on top of what they already run, one use case at a time.

When I talk with banking executives about putting customer data to work, I hear the same concern: connecting data across legacy systems could take years and cost millions. I understand the complexity. But your customers and competitors won’t wait for you to finish your AI roadmap.

When digital initiatives don’t deliver timely, personal guidance, it’s tempting to blame the legacy core. But replacing a core banking system takes years, carries real operational risk, and still doesn’t solve the immediate problem.

Our proposed architecture decouple the stability of the core from the speed of the customer experience:

• System of Record (Preserve Stability): The legacy core, lending engines, and payment rails stay untouched. They process transactions and enforce accounting rules.
• Engagement & Orchestration Layer (Inject Intelligence): Sits directly above the core, pulling customer context from current systems making it accessible across channels and departments, within your permissions, policies, and controls. It monitors real-time triggers (payroll arrivals, recurring bills, idle liquidity), applies decisioning rules, and packages the next best action. A person stays in the loop wherever judgment is needed.
• Unified Digital Front End (Execute Natively): Delivers guidance inside mobile and web channels as contextual utility, letting the customer authorize and execute the solution with a single tap.

Integration work still matters. But it scales one use case at a time, not as a multi-year program you have to finish before ROI shows up.

Modernization paths compared

What does proactive banking look like in practice?

In practice, proactive banking means turning one thing the bank already knows into one useful offer, in the channel and moment the customer needs it.

At our annual banking conference, I demonstrated three real-world examples of banks acting on relationship intelligence without touching their core. These go far beyond basic overdraft alerts; it is what true agentic banking looks like in practice.

• Personalised savings plan. A customer holds €4,000 in a current account with an explicit goal to build a safety buffer. Instead of blasting a generic savings promotion, the bank asks if they would like to explore options. The recommendation directly aligns with their goal, ensuring day-to-day liquidity stays within reach while putting the rest to work.
• Help a first-time buyer at their own pace: Repeat visits to the mortgage calculator, paired with an active first-home savings goal, trigger an offer of support. When the customer indicates they are only exploring, the system shifts to preparation and budgeting guidance rather than rushing them into an aggressive loan application. Trust is built before the transaction even begins.
• Meet financial pressure with understanding: Income and recurring costs begin to drift apart. Rather than waiting for a missed payment or sending automated collection warnings, the bank checks in proactively. When the customer mentions their hours were cut, that context routes immediately to a human banker who can pick up the conversation with empathy, forbearance options, or restructuring.

When an institution helps before being asked, those customers turn to their bank first when a need arises.

And that engagement shows up on the bottom line. At Meriwest Credit Union, members who regularly use their AI digital assistant are 30% more profitable than the average digital user. Rather than a core overhaul, Meriwest layered conversational AI directly over their existing systems.

“Customers would enthusiastically select Backbase again, citing high trust and the platform's flexibility and foundational layers that give banks the ability to try out and test new ideas. Backbase is a Customer Favorite in this evaluation.”\ Source: Forrester, The Forrester Wave™: Digital Banking Engagement Platforms, Q2 2026

Where banking leaders should begin

Transitioning to proactive banking does not require an enterprise-wide mandate or a multi-year IT freeze. Leadership teams can begin operationalizing relationship intelligence immediately with five concrete actions:

1. Measure external account linkages to quantify the risk\
   Pull the numbers on the percentage of accounts linked to outside fintechs, data aggregators, and AI financial tools.
2. Run forensics to understand the cause\
   Isolate the cohort of customers who connected an external app in the last 90 days and analyze their behavior in the 30 days prior. the shared signals (idle balances, repeat calculator visits, or surprise fees) that prompted them to look for an alternative solution.
3. Convert one churn signal into a proactive banking use case\
   Take a common churn signal from your audit and build a proactive, frictionless intervention around it. Treat it as a controlled experiment: measure the impact on deposit retention, product uptake, and external link reduction against a holdout group to prove balance-sheet impact before scaling.
4. Set clear guardrails for human routing\
   Codify exactly where automation stops. Route signals of financial stress directly to an advisor with full context, rather than triggering an automated collection alert or credit offer.
5. Decouple engagement from core modernization\
   Stop freezing digital initiatives for back-office core overhauls. Focus on an orchestration layer that taps existing APIs to trigger real-time actions without disturbing underlying ledgers.

Run the first test internally. Prove the economics on your own balance sheet.

When you are ready to scale, Backbase operationalizes relationship intelligence across your entire institution without overhauling your core.

Frequently asked questions

Is data still a competitive advantage for banks?

Yes, for now. Banks hold decades of customer history that fintechs can't build from scratch, but customers can permission it away with one tap. The lasting advantage is applied intelligence: turning raw data into real-time utility and returns before the customer looks elsewhere.

How are fintechs and AI tools gaining access to customer banking data?

Through account connections that customers initiate. The UK has more than 16.5 million live open banking connections. In the US, ChatGPT users can connect accounts from more than 12,000 institutions through Plaid. These connections feed AI agents the missing context needed to advise on spending, savings, and loan decisions.

How can a bank tell which customers are at risk?

External linkages are the earliest warning sign. A customer who connects their account to a third-party app or aggregator is demonstrating that someone else is delivering more value from their data. Running forensics on what that customer did in your own channels in the 30 days prior reveals the exact unmet need that triggered the search.

What could losing primacy cost traditional banks?

Up to $170 billion in global banking profits according to McKinsey. As AI agents proactively manage money, low-cost core deposits drain into outside yield alternatives, the bank loses right-of-first-refusal on high-margin mortgages and wealth management, and markets re-rate the institution from a relationship franchise to a commoditized utility.

Can banks stop customers from sharing their data with fintechs?

No, and they shouldn't try. Regulatory mandates and consumer expectations give customers ownership over their financial data. Banks defend primacy not by restricting access, but by delivering so much proactive value that customers have no reason to look outside.