Tag: banking

The pressure on banks has always been structural. High operating costs, tight margins, and layered regulatory complexity have defined the industry for decades. What has changed is the pace at which technology is reshaping what banks can do, and how sharply the gap between leaders and laggards is widening.

“For banks, technology is increasingly seen as a strategic enabler, underpinning trust, resilience, and growth,” says Will Moroney, Chief Revenue Officer at Temenos. Rather than focusing only on IT spend as a line item and the conversation is moving toward technology as a driver of competitive positioning. For many institutions, that shift in framing is itself the first challenge.

The Technology Trends Redefining the Future of Banking report, produced by Temenos in collaboration with Bain & Company draws on industry research and insights from Temenos Value Benchmark data spanning more than 200 banks and 100,000 data points. Its assessment shows that banks that embed intelligence into a modern core are pulling ahead, while those running on legacy infrastructure are finding agility slow, change expensive, and innovation harder to deliver.

Nearly a third of legacy banking applications lack comprehensive software documentation, creating hidden operational risk, and that’s before accounting for the data duplication and batch-processing limitations that constrain AI and analytics.

Framing modernization as a phased journey

Banks are moving away from “big bang” core replacements. Complete overhauls tend to be costly, high-risk programs. Instead, progressive modernization within a composable architecture  allows banks to upgrade components independently without destabilizing existing operations. . This approach is actually one of the top five predictors of the success of a modernization program, according to the report.

Moroney’s client conversations reflect this trend. “We frame modernization as a phased journey,” he says. “Progressive modernization spreads investment over time, reduces risk, and delivers value earlier and more often.” Rather than a multi-year, all-or-nothing commitment, banks can prioritize higher-priority components, lending, payments, customer data, while leaving other systems unchanged until needed.

The architectural underpinning matters here. A cloud-native, composable core enables incremental progress, providing the elasticity and integration capability to break apart monolithic legacy implementations without triggering system-wide disruption. Investment is also distributed more efficiently: as components go live sooner, ROI accelerates relative to a traditional replacement cycle. 

The AI readiness gap is wider than banks would like to admit

Generative and agentic AI represent a genuine step-change in what banking technology can do, but realizing that potential requires foundations that most institutions have not yet built. Banks  need strong data environments, before deploying these capabilities at scale.

“The reality is most banks are still using fragmented legacy environments, so those foundations are only partly in place,” explains Maroney. The playbook, as he describes it, starts with modernizing the core platform and the data environment, and putting strong controls around how AI accesses information.

The data challenge alone is significant. On average more than a fifth of bank data is duplicated, with banks in the bottom quartile seeing duplication rates above 52%, according to the report. That level of fragmentation raises costs, reduces accuracy, and limits the effectiveness of analytics and AI before either has even been deployed. 

The regulatory dimension has a compounding effect, too. Banking operates with very low tolerance for errors, and every AI-driven decision must be both predictable and auditable. For Moroney, these requirements are the design parameters that shape how a responsible deployment is built.

Shifting the board conversation from IT cost to business outcomes

The way banks justify technology investment has changed meaningfully over the past two years. Boards and senior stakeholders are less receptive to infrastructure arguments and more focused on evidence of business impact. This is a direct response to competitive pressure, particularly from digital-native players who have demonstrated what a modern technology platform can deliver in terms of speed, personalization, and cost-to-serve, according to Moroney.

“Trying to respond on platforms that weren’t designed for real-time decision-making or modern security standards only slows progress and drives up cost,” he says.

Digital banks achieve significantly higher front-office productivity, with top-quartile institutions serving over 6,000 customers per front-office full time employee, compared to an average of around 4,300. For corporate banking, on average, only 13.8% of products are both originated and transacted digitally. The upside available to institutions that close that gap is considerable. 

Data governance as a foundation for growth

Data mesh architecture can be a key enabler of the “intelligent bank”: A decentralized but well-governed approach that keeps data clean, accessible, and compliant across all business lines rather than siloed within individual functions. This is where the link between modernization and revenue becomes most tangible.

“Most banks still can’t get full value from data because it’s fragmented, hard to access, and often duplicated. The Temenos Value Benchmark puts duplicate data at about 21%; this raises cost and hurts accuracy. That makes it harder to use data for analytics, AI, or true personalization,” he says.

The connection to revenue runs through hyper-personalization and cross-sell effectiveness. Average products-per-customer rate across retail banking sits at 2.59, a figure that represents significant headroom for banks that can use data to anticipate customer needs and surface relevant offers at the right moment. With propensity models identifying customers likely to disengage and next-best-interaction logic enabling proactive outreach, better data architecture translates directly into wallet share, retention, and cross-sell conversion. 

The hidden cost of not modernizing, as Moroney frames it, is forfeiting these opportunities.

Agentic AI and the shift from reactive to proactive compliance

The payments and financial crime space offers one of the clearest illustrations of where AI is already delivering measurable impact, and where legacy infrastructure is most visibly straining. Payment volumes continue to grow, real-time transaction expectations are rising, and the compliance surface area is expanding accordingly.

In watchlist screening, AI agents can assess and clear low-risk alerts automatically, freeing human investigators to focus on genuinely complex or high-risk cases. In payments processing, agents can detect and repair broken transactions in real time, reducing manual intervention and increasing throughput.

“We’re already seeing this create real-world impact,” Moroney says, pointing to a production AI agent in financial crime mitigation delivering material reductions in false positives and significantly less manual investigation work. 

The emphasis on production matters: banking’s risk-aversion in this space is entirely appropriate. “Banking has zero tolerance for hallucinations. The key is deploying AI safely, predictably, and auditably, with the right partners,” says Moroney. The goal is to modernize without introducing operational or regulatory exposure.

The balance between the urgency of modernization and the discipline required to execute it responsibly is mission critical. The institutions getting it right are those treating technology as a long-term strategic asset, investing in the foundations before the capabilities, and measuring every step against the business outcomes that justify the journey.