Refined Innovation Execution Can Help Banks Regain Ground Lost to Fintechs
By Steve Cocheo, Senior Executive Editor at The Financial Brand
Simple Subscribe
Subscribe Now!
The rate of revenue growth in corporate and investment banking is flagging in the face of competition from fintechs and other non-banks. While traditional institutions are attempting to beef up their innovation in products and services in this space, in many cases the reality is falling far short of the intent, including implementation of AI.
Key insight: Both technological and human reasons lay behind stalling AI implementation, according to the inaugural edition of the Capgemini Research Institute’s World Corporate and Investment Banking Report.
This lack of traction comes at a point when 82% of corporate and financial institution customers worldwide told the firm that they prefer to interact with their providers via digital channels whenever possible. More and more clients want bank systems to integrate with their systems. This includes the ability to embed banking functions and services.
The dilemma: “Corporate and investment banking has reached a point where incremental progress no longer moves the needle,” the Capgemini report says. “Scattered pilots, isolated digital upgrades, and surface-level process fixes can’t offset slowing growth, rising competition, or rapidly shifting client expectations.”
Yet many bank executives don’t always trust AI solutions because of deficiencies in the way their organizations are training AI models, according to the study.
Need to Know:
- While corporate and investment banking revenue rose at a compound annual growth rate of 6.5% from 2022-2024, Capgemini projects a deceleration to a CAGR of 5.4% over the next five years.
- The firm projects that future rate will be 7% in the Asia-Pacific region, 4% in Europe, and in-between in the U.S., at a projected 6%.
- Only 23% of corporations and financial institution clients worldwide believe that their banks meet their needs completely — encouraging movement to competitors like fintechs.
What Clients Say is Missing from Corporate Customer Experience
Overall, bank customers told Capgemini that, despite banks’ attempts to match their wants, they face four elements of friction:
- 92% see only limited integration of bank services with their treasury platforms and other internal systems.
- 89% say banks’ offerings lack personalization and flexibility.
- Nearly as many — 87% — say banks make it hard to track transactions and don’t enable consolidated reporting, with data frequently fragmented and subject to divergent definitions in different places in the bank.
- 68% say their banks don’t provide advanced analytics nor do they provide forecasting abilities.
The study reports that 60% of bank clients feel indifferent or dissatisfied with their bank’s digital interactions.
Facing a shortfall from their banks, the study says, customers have been seeking out fintech solutions that unite streams from among multiple providers. This puts the fintechs, rather than banks, in the pilot seat.
Key point: “As third-party platforms handle daily digital interactions, banks risk being seen as commodities, rather than strategic partners,” according to Capgemini.
Read more: AI Won’t Deliver ROI Until Banks Redesign How Work Gets Done
What’s Holding Banks Back from Meeting Those Needs?
As the study makes clear, it’s not as if banks don’t understand what corporate customers want. They do, and, although conservative cultures delay things in some cases, many are putting technology in place to make it possible to deliver.
However, four barriers have been getting in the way:
1. Fragmented data constrains delivery.
Customer data resides in separate business silos inside the same bank, “with inconsistent definitions, lineage, and access rules,” the study says. As a result, even with information under the same roof, human employees lack a complete view. Beyond that, when institutions attempt to train AI models, they are often learning based on an incomplete picture. The results:
- “Only 30% of banking executives say they can give their clients proactive strategic insights.”
- “Without unified data foundations, pilots remain isolated and enterprise impact stays limited.”
2. Legacy systems still hold things up.
The study found that banks devote 43% of their IT budgets to operating and maintaining old platforms — a challenge they’ve been dealing with for decades. “Only 29% goes toward innovative or transformative technologies, like distributed ledger technology, AI, and tokenization.”
The result is a patchwork, often leaning on manual workarounds.
3. AI model reliability and governance issues degrade delivery.
The study found that just short of half of the banking executives surveyed report problems with reliability of their AI models. This ties in with many institutions still not hewing to central governance of AI efforts, the firm says.
Key point: In the absence of that governance, “models get stuck in limited production, with extensive human-in-the-loop checks, leading to uneven benefits. Without reliable, explainable models and consistent governance, AI remains an experiment rather than an operating advantage.”
4. Culture, lack of talent and lukewarm attitude can hold things back.
Two out of five banking executives surveyed indicated that their institutions are short on skilled tech and data talent.
Key point: “Without clear accountability and incentives, teams treat technology or AI as an add-on rather than as a core part of how work gets done — leaving innovation fragmented and too siloed to deliver enterprise-wide impact.”
Read more: How BECU’s AI Financial Advisor is Moving Beyond Product Answers to Customer Handholding
Get Your Act Together Before Trying to Make AI Your Innovation Base
The key point of the study is that corporate and investment bankers must build the right data and technology structures before attempting to use AI to pursue innovative products and processes.
It’s essential for firms to simplify and stabilize their front-to-back journeys before they apply AI, because sustainable value can’t be created by bolting intelligence onto fragmented, human-centric processes,” the report states.
Efforts at AI adoption are exposing the weaknesses in current structures, according to the report. This includes duplicate sets of data and unrestricted flows of unstructured documents into large language models, according to the report.
However, the report also makes the point that institutions can’t move forward only by improving technology and tech structures.
“Cultural inertia remains one of the strongest barriers to transformation,” the report says.
Two tips from Capgemini for helping employee behavior to be effective:
- Insist on innovation adoption. The report suggests making every business unit run an innovation experiment at least once every quarter.
- Create innovation “fast lanes.” This measure is intended to jumpstart adoption, removing internal bureaucratic blockages.
One bright spot, per Capgemini: the way Bank of America has extended its AI-based virtual assistant, Erica, to corporate banking:
“By automating approximately 40% of corporate client inquiries, the bank has improved response times, streamlined onboarding, and delivered more consistent client support.”
Read next: How AI is Permeating Work Throughout TD Bank, and Lessons Learned
