Customer Experience · November 16, 2026
Capital One Customer Experience: What It Gets Right and Wrong
Capital One topped J.D. Power's banking satisfaction study for six years running, then faced a CFPB lawsuit over savings rates. Here's what that split reveals about CX.
Capital One spent six consecutive years at the top of J.D. Power's U.S. National Banking Satisfaction Study, the industry's most closely watched measure of how customers feel about their bank. In January 2025, the Consumer Financial Protection Bureau sued the same company, alleging it had quietly denied millions of savings customers a higher interest rate while advertising that rate to new customers. Both things are true. Neither cancels the other out.
That contradiction is the most useful thing in banking CX right now, because it exposes a blind spot in how most companies think about customer experience: satisfaction scores measure how a journey feels in the moment, not whether the institution is being straight with you. Capital One built genuinely excellent, friction-removing experiences in the parts of banking people interact with daily — cards, apps, branches, customer service calls. At the same time, it built a savings product structure that depended on customers not noticing a better rate existed next door. The lesson for CX leaders is not "Capital One is good" or "Capital One is bad." It is that ease and integrity are measured separately, and a company can win on one while quietly losing on the other.
What does Capital One get right in customer experience?
Capital One's strength is operational: it has spent two decades systematically removing the friction that customers simply tolerate at most banks. The clearest example is fees. In 2021, Capital One became the first major U.S. bank to eliminate overdraft fees, non-sufficient-funds fees, and overdraft-transfer fees entirely, a decision that was widely covered at the time and that competitors spent the following two years scrambling to match. It also built a visible physical anchor for a largely digital bank through its Capital One Cafés — branded café-branch hybrids where customers can bank, get free coaching on credit and budgeting, or simply sit and work, with no obligation to open an account.
Neither move is a gimmick. Both target a specific, well-documented source of customer resentment — unexpected fees and sterile branch experiences — and remove it rather than apologise for it afterwards. That distinction matters more than it sounds. Richard Thaler's concept of sludge, the friction organisations deliberately or carelessly leave in place because it benefits them financially, is the right lens here. Overdraft fees are a textbook case of institutionalised sludge: banks earned billions from them for decades precisely because reversing a fee required a customer to notice, call, and ask. Capital One didn't streamline the complaint process for overdraft fees. It deleted the mechanism that generated complaints in the first place. That is the difference between reducing friction and removing it.
- Fee elimination as default design: removing overdraft and NSF fees changed the default outcome for every customer, not just the ones who complained — a far stronger intervention than better dispute handling.
- Café network as low-pressure touchpoint: a space with no sales obligation lowers the psychological cost of walking in, which matters for customers who avoid banks until a problem is already serious.
- Consistent satisfaction performance: six consecutive years leading J.D. Power's national banking study is not a single good quarter — it reflects a sustained operating discipline across card, auto finance, and retail banking lines.
How does Capital One's use of digital tools change the experience?
Capital One was also early to put a conversational layer over account servicing, through its Eno virtual assistant, which flags unusual charges, forecasts subscription price increases, and lets customers freeze a card from a text message rather than a phone queue. The behavioural logic is simple: every alternative to calling a contact centre is also an alternative to waiting on hold, and wait time is one of the most reliably negative moments in any service journey. Moving resolution upstream, before a customer notices a problem, is a stronger CX move than resolving it faster once they call.
This is where J.D. Power's satisfaction methodology genuinely rewards Capital One, because the study weighs channel convenience, digital tools, and problem resolution heavily. A customer who never has to speak to anyone because the app already caught the issue will, almost by definition, score the experience well. It is a legitimate win. It is also a narrower measure of experience than most executives assume, because it says nothing about pricing fairness, disclosure quality, or what happens to the customers the product was never really designed to serve well. That gap is where Capital One's record gets more complicated.
What does Capital One get wrong?
The clearest answer sits in the CFPB's January 2025 complaint. The regulator alleged that Capital One marketed a product called 360 Performance Savings to new customers at a materially higher interest rate, while quietly freezing the rate on its older 360 Savings product and never telling existing customers that a better version of the same account existed, even as rates rose sharply across 2019–2024. The CFPB put a figure on the harm:
The bureau's complaint alleges Capital One "cheated consumers out of more than $2 billion in interest payments" by failing to tell 360 Savings customers about the higher-yield 360 Performance Savings account, according to the Consumer Financial Protection Bureau's January 2025 press release.
Strip away the legal framing and what's left is a behavioural-design problem, not a pricing problem. Capital One did not lie about the interest rate on either account. It simply relied on inertia: once a customer has money sitting in an account that isn't actively hurting them, they rarely go looking for a better one, even when the same bank offers it under a different name a click away. That reliance on customer inertia to protect a margin is the mirror image of the fee-elimination decision described above. One used behavioural economics to help customers. The other, as alleged, used it to quietly disadvantage them. The two sit inside the same company, often in the same product category.
This is not a uniquely Capital One failure. Dormant-account and legacy-product pricing gaps exist across the banking industry; several large U.S. and UK banks have faced similar scrutiny over "zombie" savings rates that lag far behind rates on newer products from the same institution. What makes the Capital One case instructive is the scale of the contrast: a bank disciplined enough to engineer away overdraft fee friction appears to have been comparatively undisciplined about engineering away savings-rate friction, because one cost the business revenue and the other protected it.
What does the 2019 data breach reveal about trust as a CX dimension?
Trust failures compound this picture. In 2019, Capital One disclosed that a former cloud-infrastructure employee had exploited a misconfigured firewall to access personal data belonging to more than 100 million customers and credit-card applicants, including Social Security numbers and bank account details for a smaller subset. U.S. banking regulators subsequently fined Capital One for deficiencies in its risk-management practices tied to the breach. The incident predates most of the CX wins described above, but it belongs in the same account, because satisfaction surveys ask customers how they feel about their bank this year — not whether that bank has ever put their identity at risk.
Customer experience leaders often treat security and privacy as an IT or compliance function, separate from "the journey." That separation is a mistake. A breach, a regulatory fine, or a lawsuit alleging deception is a moment of truth exactly like a late flight or a denied claim — it just happens off-screen, discovered later through a news alert rather than a bad call. The emotional arc of the customer relationship takes the hit regardless of which department caused it.
Why can a company score well on satisfaction and still face serious trust failures?
Because satisfaction and integrity are measured on different clocks. A satisfaction survey captures a recent transaction: did the app work, was the agent helpful, how long did the call take. Trust is cumulative and retrospective — it is built or broken by disclosures, fairness, and consistency over years, and it often only surfaces when something goes wrong publicly. Daniel Kahneman's peak-end rule helps explain why Capital One can keep winning satisfaction studies despite this: customers judge an experience disproportionately by its most intense moment and its ending, not its average. If the ending of most interactions — the app resolves it, the café visit was pleasant, the fee never appeared — is positive, the aggregate memory of the bank stays positive, even while a structural issue sits elsewhere in the business, invisible until a regulator or journalist surfaces it.
This is precisely why a J.D. Power ranking, Net Promoter Score, or CSAT figure should never be treated as a full verdict on an organisation's customer experience. These metrics are necessary. They are not sufficient. A mature CX function tracks the felt experience and audits the structural fairness of the business separately, because the two can diverge for years before anyone notices.
What should other CX and banking leaders take from Capital One's record?
The practical lesson is not to copy Capital One wholesale; it is to copy the discipline behind its wins and build in the safeguard its lawsuit suggests was missing. A few moves apply directly to any bank, insurer, or subscription business sitting on legacy pricing tiers:
- Audit every "grandfathered" product line for silent rate or fee gaps. Any time a company sells a new, better-priced version of an existing product without migrating or clearly notifying existing customers, it is running the same structural risk Capital One now faces in court.
- Treat fee and rate transparency as a design decision, not a disclosure checkbox. Burying a better rate in a product comparison page that existing customers never visit is a defaults problem, not an information problem — the fix is proactive notification, not better fine print.
- Measure resolution speed and resolution fairness as two separate metrics. A company can resolve complaints quickly and still be resolving them in its own favour; track both the time-to-resolution and the rate at which disputes are settled in the customer's favour.
- Map the full journey, including the moments nobody designed. A data breach, a legacy pricing tier, and a regulatory notice are all touchpoints, even though no one in product or design ever drew them on a journey map.
- Pressure-test friction removal in both directions. Before congratulating a team for removing a fee, ask whether the business has quietly added equivalent friction somewhere less visible, as the CFPB alleges happened with savings rates.
Banks and fintechs building out these journeys properly need a structured way to see where the gaps sit, rather than relying on a satisfaction score that only captures the parts customers happen to notice. A full CX journey mapping exercise that includes pricing structures, legacy products, and disclosure moments — not just app flows and call centre scripts — would have surfaced the exact gap between 360 Savings and 360 Performance Savings long before a regulator did. That kind of audit sits squarely inside what behavioural-economics-led CX work in banking is for: finding the places where a default is quietly working against the customer, not just the places where a queue is too long.
Where does behavioural economics explain the gap between Capital One's two records?
Every example in this piece reduces to the same mechanism: defaults decide outcomes, and whoever sets the default carries the responsibility for what it does to the customer. Capital One's overdraft-fee elimination reset a harmful default in the customer's favour. Its alleged handling of legacy savings rates, according to the CFPB, left a harmful default in place because resetting it would have cost the bank revenue. Loss aversion, inertia, and the sheer effort of comparison-shopping are not neutral forces in a banking relationship — they are commercial assets, and a company decides, consciously or not, which side of the ledger it puts them on.
Renascence's own view, built on work applying behavioural economics to customer experience, is that this is the single most important test for any CX leader auditing their own organisation: find every place a customer's inertia currently benefits the business, and ask whether that was a deliberate design choice or an accident nobody has revisited since the product launched. Most companies find at least one answer they don't like.
What this means for the next banking satisfaction leaderboard
Capital One will likely keep performing well on the metrics J.D. Power and similar studies track, because the underlying product decisions — fewer fees, faster digital resolution, a less intimidating branch format — are real and durable. But the CFPB case is a reminder that a satisfaction leaderboard measures the experience customers can see, not the one buried in a legacy product's terms. The banks that will matter most over the next decade are the ones that stop treating those as two separate scorecards and start auditing both with the same rigour, before a regulator forces the comparison for them. Renascence works with banks and financial institutions on exactly this kind of structural journey and voice-of-customer audit; if a legacy product gap like Capital One's is sitting undetected in your own portfolio, a customer experience assessment is the place to start finding it before a customer, journalist, or regulator does it for you.
Further reading
FAQ
Questions we get on this topic
Related reading
Writing on how human behavior shapes the experiences brands deliver — at the intersection of behavioral economics and customer experience.
Stay ahead of CX
Get the Journal in your inbox.
Insights, frameworks and event round-ups from the Renascence team. No spam, ever.






