Texas
Capital Bank
Before the redesign, one question: how far behind is the experience, exactly?
Texas Capital Bank was launching a digital transformation, and Deloitte Digital was positioning to guide it. The kickoff needed proof — not opinion — that the team understood the market. So I ran a heuristic evaluation and competitive teardownof TCB’s consumer online banking against the experiences customers actually compare it to: Chase, Wells Fargo, and Marcus.
The kickoff was the moment of truth.
A transformation this size is won or lost in the first room. The kickoff had to do two jobs at once: show the bank the team genuinely understood its market, and give everyone a shared, evidence-based picture of where the current experience stood. The honest way to do that isn’t a pitch — it’s an audit.
Clawing back credibility
You spend the rest of the engagement earning back trust the kickoff was supposed to establish.
Momentum from day one
The room agrees on the problem before anyone argues about the solution.
Grade the experience against best practice — and against the market.
A heuristic evaluation is where UX experts use established usability principles to gauge a digital experience. Combined with a market comparison, it benchmarks TCB against the experiences customers already use every day — and shows exactly where design choices could make the updated experience stand out.
Define the heuristics
Set the usability principles the review would grade against — the parameters everyone agrees are “good.”
Examine TCB’s touchpoints
Walk the real consumer-banking screens — dashboard, payments, transfers, alerts, errors, help — against those heuristics.
Compare to the market
Grade the same tasks at Chase, Wells Fargo, and Marcus to locate the gaps — and the openings.
Make it legible
I owned the visuals — a scorecard and side-by-sides that let the room absorb a market position in five seconds, not five paragraphs.
Four high-frequency features, graded against the leaders.
We picked the four features consumers touch most — Accounts & Dashboard, Payments, Transfers, and Peer-to-Peer — and pitted TCB against industry-leading experiences. The point wasn’t to shame the current design; it was to find the seams where a better pattern was both obvious and winnable.
Recreated screens (illustrative). TCB’s side navigation is well-organized but eats horizontal space a top bar wouldn’t; leading experiences keep an ever-present top nav. Marcus, by contrast, delivers a single, unmistakable balance and one or two relevant actions.
The right rail is real estate TCB isn't spending
Both TCB and Wells Fargo use a right rail for secondary CTAs and announcements — but TCB’s carries little contextual value. It’s the most valuable space on the page and it’s idling.
Navigation should cost less space
A top-bar navigation delivers an ever-present touchpoint while taking minimal room, freeing the page for information and next-best actions.
Recreated screens (illustrative). TCB's full-page payment flow makes it hard to cancel out of a contextual journey — a user who fails is more likely to abandon the whole site. Chase's modal keeps the surrounding context visible, so users stay oriented and exit the flow without leaving the experience.
Payments should retain context
Modal designs let users keep the context of the action. TCB’s full-page journey raises the cost of a single misstep to “lose the whole page.”
Recreated screen (illustrative). A simplified transfer form with limited fields on one page keeps the user focused; dropdowns minimize input error; a disabled CTA clearly signals that more information is needed before the task can complete; and the right rail can surface currently-scheduled transfers if properly used.
Transfers were the one TCB was already close on
Limited fields, sensible dropdowns, and a clearly-disabled CTA are exactly right. The opening is the right rail — scheduled transfers are useful context that’s underused.
Recreated screen (illustrative) of a Zelle-style peer-to-peer flow. In TCB's current experience there's no way to save a recipient for future use; Wells Fargo makes it easy to send and set up recurring payments, with in-context instructions, distinct alert types, and a clean recipient list.
P2P had no memory
No save-for-future, no recurring. A leading experience treats recipients as reusable and payments as schedulable — and folds in-context help and clear alert types into the same screen.
But watch the CTA count
Multiple competing calls to action can overwhelm — an opportunity to consolidate, whichever bank you are.
Then the same experience, graded against six usability heuristics.
Beyond feature-by-feature, I graded the experience on the usability principles that separate a modern banking app from a dated one — aesthetics, system status, language, errors, ease of use, and help. A handful stood out.
Recreated screens (illustrative). Alerts are a 'set and forget' feature: good experiences balance depth (how many alerts) with hierarchy (collapsible categories) and let users choose the channel. Wells Fargo categorizes; Chase gives one alert multiple delivery channels.
Recreated screen (illustrative). For transactional accounts, distinguishing pending from posted gives users real visibility into money moving in and out — a small status distinction with an outsized effect on trust.
Statuses got lost on a busy page
TCB’s homepage offered a lot of information, but competing CTAs drowned out contextual guidance — and any status apart from “Messages” disappeared. The move: suppress the nav when idle, elevate high-value actions to the top of the right rail, and simplify the language.
Speed beats spinners
Users are sensitive to system speed even with progress indicators. No progress bar beats a fast-loading page — a design finding that’s really an engineering priority.
Recreated screens (illustrative). All three banks clearly state an incorrect username or password — but Chase and Wells Fargo bring the solution into the error itself, offering to reset a password or recover a username. TCB states the problem and stops there.
An error message should carry its own fix
Express errors in plain language (no codes), name the problem precisely, and constructively suggest a solution. Prevent the high-cost errors first — good defaults and helpful constraints beat any recovery flow.
Recreated screen (illustrative). Help lived on TCB's public site, but inside the digital experience a user had to hunt for a branch number or send a message. Leading experiences keep help consistently available and non-intrusive — categorized, multi-channel, in context.
Self-service help wasn't where users needed it
Make help easy to search, relevant to the task, and available through more than one channel — secure message, chat, scheduler — right at the moment it’s needed.
Aesthetics: prize the page, cut the repeats
Repeated labels and instructions diminished the visibility of what actually mattered. Minimal aesthetics aren’t decoration — they make systems easier to use.
Ease of use: reward the expert
Little accommodation for power users. Shortcuts, customization, and personalization let one system serve both the first-timer and the daily user.
A defensible point of view.
Each heuristic closed on concrete moves — not vibes. Consolidate navigation into a single, space-efficient menu. Prioritize content around primary goals and cut the clutter. Make system status and transaction speed front-of-mind. Write errors in plain language and carry the fix. Empower customization and reward expert users. Make help searchable, contextual, and multi-channel. A month isn’t enough to prove all of it with primary research — so I framed it as an evaluation, grounded in real competitive evidence, honest about what I knew versus inferred.
In early-stage strategy, the deliverable is alignment — and a rigorous, well-designed audit is one of the most powerful tools for creating it.
The evaluation fed directly into the kickoff: a scorecard, four feature teardowns, six heuristic findings, and a set of recommendations that gave the room a shared, evidence-based picture of where the experience stood and where it could go. The highest-leverage moment in a big transformation is often the smallest one — the room where everyone finally agrees on the problem.
A 2021 heuristic study of publicly-available consumer banking experiences. Screens here are code-based recreations built to illustrate each finding — not the original confidential deck or any bank’s actual screenshots. Grades are illustrative of the study’s directional read.