- Fintech Brainfood
- Posts
- UX isn't given enough credit as a Fintech Superpower
UX isn't given enough credit as a Fintech Superpower
Great UX reduces CAC, Cost to Serve and increases cross sell. Plus, my take on the Mercury / Choice / De-risking Saga.
Welcome to Fintech Brainfood, the weekly deep dive into Fintech news, events, and analysis.
Hey Fintech Nerds 👋
Welcome back. I had a wonderful birthday with family (thanks for asking), and I am back this week, Ranting about why Great UX in Fintech doesn’t get enough credit. This has been my favorite for a long while.
Revolut got its UK banking license (with restrictions) after a 3-year wait. This is a huge moment of validation for Fintech, which had everything but the license.
Considering bankers used to laugh at the silly little Fintech company doing FX in 2015, they’ve come a long way. How the US Neobanks must look on and wonder what it would be like to get a de novo charter.
Mercury is in the headlines again for de-risking clients in countries like Ukraine and Nigeria. And, while they’ve clearly communicated it poorly, I think there’s plenty of blame to go around here not just Mercury. How could a bank allow this? (More in Things to Know 👀)
The CFPB says tipping a Fintech company is lending. I actually agree.
PS. Ready for Fintech DevCon? I’ll be there with Alex Johnson, Jason Mikula, and countless other Fintech nerds. So grab a ticket and come say hi 👋
Here's this week's Brainfood in summary
📣 Rant: The UX Magic Trick
💸 4 Fintech Companies:
👀 Things to Know:
📚 Good Read: The EU has gone too far
This week’s edition is way too long for Gmail; click the below version to view it online.
Weekly Rant 📣
The UX Magic Trick
Hot take: Upgrading the UX of finance gets unfairly dismissed, yet is the single most impactful thing we can do. The problem with building great UX is it sounds simple in theory, but is incredibly difficult in practice.
Every three months or so, I meet a new product team that came from Big Tech and consumer mobile and wants to design a "frictionless UX," and I facepalm.
This Rant focusses on:
Fintech UX is hard because of the edge cases
UX was always, and is still misunderstood by incumbents
This created the opportunity for disruption in experience
E.g. Shop Pay's frictionless checkout
E.g. Monzo fraud prevention features
E.g. Ramp's "make recurring" feature
Great services all
Focus on performance (like page load)
Remove hidden speed bumps
Solve problems in a creative way
These features yield
Lower CAC
Lower Cost to Serve
More cross sell opportunity
Fintech is undefeated at UX, and UX has an incredible competitive advantage from lowering structural costs.
Fintech UX is hard because finance is all edge cases
Every generation that enters Fintech goes.
Payments are too hard.
I will build an app with no friction.
Oh, I have to build a KYC thing; now I will remove all the friction.
Oh, the card networks are going to ban me if I don't get my fraud rates down.
Oh, my bank wants to de-platform me because I don't have AML processes.
Oh, payments randomly don't work sometimes and are full of weird formatting errors. Customers are complaining.
All of this is 100x harder cross border.
Welcome to the vicious cycle of why UX is hard.
You can observe this pattern in every generation, from PayPal to early Neobanks, to Crypto.
Great UX is incredibly hard.
Great UX comes from solving 1,000s if not 10,000, tiny little issues in infrastructure, process, and compliance that add up to feeling frictionless. There are so many that enumerating them all could create a GPT5 scale training data set.
Nearly all of the complexity of financial services is below the waterline. Take these four buckets for example:
Risk
Regulation
Infrastructure
Economics
To a user, applying for a loan should be simple, supply some information, set a recurring payment, and be done.
Risk: The lender has to check if the user has any links to organized crime terrorist financing if they have a history of fraud, and if their documents are accurate or forged, and that's before they begin to figure out if the borrower will repay the loan in full!
Regulation: If you're lending, you must make adequate disclosures (about the APR and repayment terms). In the US, this is the Truth in Lending (TILA) Act. Then there's the Equal Credit Opportunity Act and Fair Housing. Things that make sense on a spreadsheet (like not lending to low credit score customers) create an unfair social bias. The web of regulation and compliance is unbelievably complex.
Infrastructure: The basic tech infrastructure of finance is mostly from the 1970s through the 1990s. Data from credit bureaus and payments made via ACH can create delays, inconsistencies, and errors. This leaves the lender (or any Fintech company) figuring out if the data they received is accurate or reliable. Or worse, if a customer misses a payment, is a fraudster, or has some genuine infrastructure error.
Economics: Lending is easy, but getting paid back is much harder. So you have to price that risk at a level where you still profit, but you're not scaring away borrowers by being too expensive. You could improve your lending accuracy by spending much more time underwriting customers up front, but this adds cost. When you add cost, you can erode the margin of winning that customer.
Everything becomes a balancing act.
Get enough data to decide if you can lend, but ensure it's not biased. Price at a level that's not too high, not too low.
Do all of this while managing a complex web of regulation and unreliable third-party infrastructure.
Incumbents use this difficulty to justify their compromised UX. They position it as a feature, not a bug. "We may not be slick, but we're safe and profitable."
Or, worse, they dismiss UX entirely as a "nice look and feel" and something for designers to worry about.
Either of these perspectives miss the point.
UX was and still is misunderstood by Incumbents
Banks often reduce UX to "great look and feel," people scoff when you suggest that UX is a moat or competitive advantage. If you ever hear someone at an incumbent saying, "We have that feature too," that is an immediate sign that they have not understood how hard great UX is.
We wouldn't be so dismissive for Apple, so why do we do it for financial services?
Apple goes to great lengths in marketing to show the tiny details that go into its crafted experience. This is because crafting an amazing experience is genuinely difficult. Anyone can copy and paste a feature, but very few know why the feature was developed in the first place and why it's delightful.
I spend half of my life showing bankers tiny nuances of the best Fintech companies to receive this response "🤯🤯🤯." Usually because they've never used that product or experienced the delight.
So when they say, "We have that feature too." It's true in the same way that both Taylor Swift and Simon Taylor breathe air and have Taylor and the letter S, but one is a billionaire, and one is not.
A great user experience is fundamentally about solving a previously difficult problem. It sounds so incredibly easy to say, but it's so very difficult in practice. That's not something you can outsource to a design agency.
By all means, use good design partners, but you must understand why a feature exists.
Everyone wants great UX, but very few are willing to put in the work. Banks: your UX is bad and you should feel bad.
This creates an enormous opportunity for Fintech companies to disrupt. It's usually best to start with examples of how Fintech experiences differ from incumbents.
Fintech's opportunity is UX-driven disruption
I like many things about Fintech companies, but more than anything, I like their culture of wrestling with complexity to build something better.
Here are some things I regularly point at and go, "Oh, that's neat."
Shopify Pay
Monzo's fraud controls
How Ramp does recurring expenses
1. Shopify Pay (Shop).
Shopify Pay looks like any other branded checkout button. Instead of clicking on PayPal or Apple, you can click "Shop." The first-time user experience promises to remember you for express checkouts in the future.
The second time, the user experience is pure magic. 🧙♂️
💻 Hit checkout: When you hit checkout, Shop recognizes a returning user and sends them an SMS or app push notification to validate a payment.
🤳 Quick 2-Factor Authentication: Users check their phones and enter the number they received via SMS (or their app) on the checkout page.
🚢 Shipping and card defaults stored: The default card and shipping information is used (for payments nerds it’s “card on file everywhere with strong customer authentication as a UX feature not bug.”
Users can opt out of all of this.
💡 Why does this work? This is aggressively frictionless. "I assume you want to use the card and address you used last time; we got this." What's special is all the hidden work that makes this possible. All of the tiny user devices, behavior, fraud, risk, page load, and thousands of micro-optimizations make this possible.
If you have a returning user, you've seen them make a successful transaction with that account before. Verifying with SMS or push notification is a "step-up verification," but instead of being an additional step, it's used as the only step. 👏👏👏
2. Monzo Fraud prevention
For international readers who don't have an account, keep in mind that Monzo's primary use case is the everyday spend account. A sort of money hub, and router. It might not get the full salary payment, but it's the best place to manage money in, out, one-off payments, or anything else.
Because it works.
Because edge cases are handled elegantly. If you look at screenshots of Monzo's page flows, they look like any other at first glance.
But here's an example of a feature detail I love.
They just announced upcoming fraud prevention features for users.
📍 Known locations or "leave your money at home". Places you're comfortable moving large amounts of money but fraudsters are unlikely to access
🤝 Trusted contacts. If a large transfer is requested your trusted friend must approve. Hugely helpful if your phone is stolen or for vulnerable populations.
🤫 Secret QR codes. Print out a QR code and keep it in a safe place to scan when needed.
If these aren't available, the fallback is a step up KYC. It's one thing to have these types of signals on the mobile device; it's another to package them well.
💡 Why does this work? It's not hitting a call center or dealing with the problem after the fact. It's reducing customer support volume and building trust with them.
3. Ramp recurring expenses.
Many of you will have used Ramp, but most bankers haven't. On top of having expense cards that just work, the company has 1,000s of tiny little features that save time and effort. Examples like
📧 Forward receipts from personal emails: Some receipts appear in your personal inbox. Ramp removed the step of forwarding receipts to work email
⚡ Auto-suggesting an expense memo with AI: Based on the transaction data and your company role, Ramp will suggest an expense memo like "Marketing team travel" or "Conference travel."
♻ Make recurring: Subscriptions that hit your card as a transaction every month require a memo. The "make recurring" button recognizes the repeat transaction and just applies the same memo.
Ramp expenses suggested reasons. When complete it auto loads into the next transaction
You could write entire book chapters on each of these features.
And yet.
None of these features is the killer feature by themselves.
The sum is much greater than the parts. The lesson is in being the type of organization that would create these features before they exist.
That's the innovation.
There are 10s of thousands of companies I could have name-checked here. These just happen to be at the top of my mind / semi-canonical.
What all of these services have in common
There's a bunch of intangible stuff you will never see in a screenshot but you can feel as a user, like
Page load times
Removing hidden speedbumps
Creative problem-solving with digital tools
1. Page load times matter
As Kaz Nejatian so eloquently put:
Not all fast software is great, but all great software is fast. Every piece of software you like to use is very fast. Everything just happens. No delay
Each 100ms improvement in page (or app screen) load time can increase conversion by 1%. Conversely, Google found every second delay in page loads on mobile reduces conversion by 20%.
💡 Faster page load times increase conversion: Page loads are like lap times. The best CEOs obsess over that as an input they can control, which yields higher conversion output. Speed matters whether that's onboarding a new user, making a payment, or at e-commerce checkout.
So why can't we make everything go super fast?
I refer you to the payments doom loop at the top of this post. Removing friction is not as simple as it sounds. There are countless things beneath the waterline of the UX experience iceberg. Every bit of technology, firewall, config, code, fraud rules, 3rd party, and SDK you use adds up.
These hidden speed bumps are worth exploring.
2. Removing hidden speed bumps creates delight.
Your existing customers don't always notice bad UX. Some might complain about being stuck in a call center doom loop or unable to resolve an issue, but it's rarely a UX complaint. Most incumbent banks will log customer support calls, but they focus their KPI on reducing the cost of each call, not the cause of each call.
The clue to hidden pain is in user support tickets. Most UX-obsessed companies are also obsessed with customer support tickets. An Anecdote Jason co-founder of Monzo once told me about their early days was that the top 3 reasons for customer support calls were
Did this payment work?
What's that transaction?
I don't recognize this thing on my statement?
So they put in an enormous effort to ensure that every time a user initiated a payment, the push notification was instant. Often, long before the in-store checkout or e-commerce payment had completed, Monzo would notify a user of success. They also added merchant logos to transactions and cleaned up the transaction data, making it much clearer.
These features are fairly standard today, but in 2016, they were novel.
Great UX is a form of prevention that is better than a cure for a problem. Making a payment recurring means a user saves time, and fraud controls prevent a user from calling in distress about a scam.
The compounding effect of these moments of delight is one that users start to genuinely love.
💡 User delight reduces CAC and increases cross sell: If users deeply enjoy using a product they tend to tell their friends. To this day, 66% of Monzo users join via word of mouth, which is similar to Mercury. Nubank's fully loaded cost to acquire a customer is $9. That compares with a minimum of ~$250 for most incumbent banks (before switching incentives). Users who love a product are more likely to expand into it (the cross-sell unlock).
💡 Removing hidden speed bumps reduces the Cost to Serve: If the product has prevented a problem that isn't a customer support ticket. If the user isn't having problems, they enjoy the product. Which creates a flywheel effect for word of mouth.
3. Creative problem-solving uses technology in novel ways
Everyone in financial services has the same constraints, but only a few imagine new ways to solve those problems. While you can argue that the Fintech companies win by having a higher risk tolerance, that's not true or fair.
Some Fintech companies unquestionably have had wider risk appetites than some large banks.
However, the great ones differentiate in how they solve a problem.
It's much easier to copy a feature once someone else has already created it, but it's much harder to come up with it in the first place. There's a strong correlation between those who innovate on UX fastest and those growing the fastest.
💡 Creativity comes from knowing what technology can do. Monzo combined the mobile push notification, focusing on building its own authorization engine to ensure it got that data to the customer as fast as possible. Shopify knows that a second authentication factor adds security to a transaction, but by skipping straight there, it also removed friction at checkout.
Technology understanding unlocks better UX.
Yes. Price will always matter. But the price is dictated by cost. Technology and UX are the greatest cost levers.
Great UX yields a differentiated cost structure, which drives the competitive advantage.
Summary: Moving from good to great UX tangible business benefits
Every company wants more revenue, lower costs, and customer retention. Those are outputs, but it's often very hard to draw a direct link between a great user experience and those outcomes.
Great UX reduces CAC.
By increasing conversion
And increasing word of mouth / advocacy
Great UX reduces Cost to Serve
By avoiding customer support
And automating what would have been analog processes
Great UX increases Cross Sell %
If it focuses on the inputs that lead to a great UX, any organization can reduce its cost of acquisition and cost to serve and become much better at cross-selling to existing users.
Those inputs are making speed an intentional goal, focussing on customer pain and speedbumps as a sign of what to improve and then how to get better at creatively solving those problems with the tools mobile and digital technology has.
Incumbents want these things but don't put in the work. So they'll wait to see those features appear in a Fintech company and then copy and paste without understanding why that feature exists or how the company came up with it.
That opportunity is still wide open for Fintech companies.
Neobanks are just getting started. The best incumbent (JP Morgan) is still building branches to grow (and it's a strategy that works). The opportunity to remove hidden speed bumps and win at UX is just starting.
For the Fintech companies that focus on the inputs of performance and UX, the outputs of lower costs, better retention, and more revenue will follow.
Removing friction in UX is counterintuitive.
Removing friction in Fintech UX comes from engaging with complexity.
That's the competitive advantage of a Fintech Nerd.
ST.
How would you rate this weeks edition?I crave feedback. |
4 Fintech Companies 💸
1. Hebbia - The AI platform for finance, law and compliance
Hebbia helps asset managers, law firms, private credit, and real estate funds use their internal documents to figure out what to do next. For example, a credit team could look at "what terms did we offer companies with similar financials in the past." Law firms can figure out key negotiation levers, and real estate firms can summarize their property portfolio.
🧠 I hadn't heard of this company until their Series B raise, but I was surprised they already have major funds and law firms as clients. Interestingly, the output is presented in a table format and tends to cite its sources well. It's another way to package GenAI. We're still very early in figuring out how it shows up, but this example is becoming a pattern in complex business workflows. This feels different from most people's experience of LLMs, which is either directly related to the models or the Microsoft co-pilot style "baked into the product."
2. Niva - Global KYB
Niva helps lenders, Fintech companies, and banks onboard business customers through a Know Your Business (KYB) process. This includes collecting documents, checking them against government databases, and screening for any sanctions risks. They balance reducing the friction at onboarding (down to 10 minutes) with detailed checks like taxpayer status or screening for active litigations.
🧠 There's a massive need for this product outside the US. Companies like Middesk have become a default in the US, making complex KYB processes more automated and effective. Once you leave the US, it's a zoo of local providers or a vacuum. Niva is focussing on LATAM markets initially, where there's massive growth in Fintech opportunity. This will help companies who want to expand into LATAM as much as the local ones.
3. Tapi - Adyen for LATAM
Tapi provides a combination of checkout and recurring payments (billing) to companies in Argentina, Colombia, Peru, Chile and Mexico. It supports LATAM native payment methods like airtime cash in/out and local payment networks.
🧠 There isn't yet a dominant Stripe or Adyen for LATAM. LATAM looks more like Europe because each market has a fragmentation of local payment methods, laws, and regulations. Someone like a Dlocal helps, but the products that make Stripe so successful are things like checkout and billing. Stripe and Adyen have struggled to offer the scale and experience to Tapi's home countries, which users expect in the US or Europe because they lack the conversion and experience in local payment methods. That is Tapi's opportunity. Its focus markets are growing big in their own right but are often second in line for Fintech innovation. What made Adyen stick initially was the modern products plus the early lead in local European payment methods. LATAM markets haven't had that until now.
4. Trustap - Escrow for marketplaces
Trustap allows buyers and sellers to transact without the worry of being scammed. Users define if they're a buyer or seller, how postage will work and can enter pricing or postage information into the downloadable app. Once complete delivery is tracked, the app shows sellers delivery complete, and buyers that the goods are en route.
🧠 Why isn't this more common? Marketplaces have a massive fraud risk. The flywheel of a marketplace requires attracting good sellers to get repeat buyers. But fake reviews, goods not delivered, fake listings have created a zoo, where 40% of users are now wary of marketplaces. Escrow could help fix this. (Funnily enough, for cash markets like Germany, some sort of escrow is quite common, and much of Klarna's history is rooted in this type of use case from the early days of e-commerce). I wonder how long it is until this is a Stripe feature? Because it feels like a no-brainer now I've seen it (and the fraud risk in marketplaces).
Things to know 👀
"Digital banking startup Mercury is no longer serving customers in certain countries, including Ukraine, the company confirmed to TechCrunch." This follows FDIC concerns that Mercury's bank Choice had opened Mercury accounts in "legally risky countries." This issue caught fire when Fuel Finance Alyona Mysko's social posts went viral after their account was shut down. Mysko later noted after talking with Mercury founder Immad Akhund, the issue appears emblematic of a problem in the wider banking systems where banks do not differentiate between Russia and Ukraine.
🧠 The core issue here is with the bank partner, yet the press is about Mercury == bad. That seems wrong. It appears Mercury's bank had no issue with accounts in certain countries, and then later did. I won't pretend every Fintech company is a compliance angel, but lets also not make them the scapegoat.
🧠 Look, I'm not saying Mercury is blameless in this situation; I’m saying this situation should never have been allowed to happen. I was calling out Fintech companies for their checkbox approach to compliance back in 2022. They also apparently allowed registered agent addresses during KYB (rookie error). They should have been better from the start. They had warning and could have communicated this better. All of this should be caught by the sponsor bank, or better yet, the Fintech company should be directly accountable to regulators.
🧠 Banks lack nuance with Sanctions because it's better to be safe than sorry for them. The downside of a bad customer experience is much better than the downside of a consent order or a potentially multi-regulatory fine. After Russia invaded Ukraine, every bank began to panic about sanctions. Defensively de-risking potentially good customers is a classic banking move.
🧠 Neobanks and Fintech companies are left to wear the horrible UX of banks de-risking entire countries. If your bank partner says, "This is fine," then later says, "Oh wait, no, it isn't." The Fintech company has to follow the instructions of their bank partner.
🧠 Being de-risked is a horrible experience for anyone involved. It sucks for the customer because it's a rug pull. It sucks for the Fintech because they're not allowed to explain too much about why, because if they did, they'd be in breach of "tipping off rules." This means that if you suspect a client of AML or a risk of sanctions, you cannot explain it too much.
🧠 We desperately need a path to Fintech companies getting charters. I think just about every major Fintech company welcomes direct federal oversight. If there are gaps in controls, let's fix them. But the situation now where tiny banks get it wrong and create an issue 3 years later is the worst-of-all-worlds.
Revolut has secured its UK Banking license, which will allow it to offer overdrafts, loans, and savings products to its more than 9 million UK consumers.
🧠 This is a huge moment for Revolut and UK Fintech. Revolut has talked a big game on the license for at least 5 years. Constant compliance and audit issues created delays, but it seems they finally got over the line.
🧠 A massive congratulations to everyone who's worked there. Notably, their UK CEO, Francesca Carlesi, has been there for just 9 months and appears to have made a massive difference. A well-known, credible UK CEO goes a long way in the eyes of regulators. I’ve called Revolut the boy who cried wolf on getting a licence. Well. I was wrong. They got it done.
🧠 Note this is "with restrictions." The UK regulator (PRA) has a two step process to getting a full banking licence. "With restrictions" allows the bank to begin to mobilize and prove their operations, before moving to the full banking licence phase.
🧠 How many US Neobanks are looking on with envy? The US does not have a multi-phase process for granting de novo charters. Imagine if it did.
Good Reads 📚
Ben Thompson argues that the EU has gone too far in regulating Apple. Its regulations are largely ineffective and annoying. Mandating browser choice didn't stop Chrome from becoming dominant, and having Google and Meta collect data is likely better than sitting with companies with poor cybersecurity controls. The EU is threatening Apple with 10% of its global revenue, when it makes just 7% on the continent. This makes the cost of operating in Europe potentially higher than the revenue a Big Tech company could generate there.
Apple will likely continue operating in the EU but will neuter its products by not offering new AI features in the operating system.
🧠 Regulation can create as well as constrain innovation. At its best, the EU's approach on open banking sparked an idea that has gone global.
🧠 Well-intended policy doesn't guarantee good outcomes. The idea of online privacy or 3rd party access to NFC chips is a good goal. The problem is that whenever I have to deal with a GDPR lawyer, I want to chew concrete instead.
🧠 Law and Regulation should be measured for effectiveness, tested, and garbage collected annually. There's a sense that passing a new rule is an achievement (partially because passing a law is so tough.) The reality is, unless it works, it's useless. There's no customer testing or feedback mechanism baked in for society. Regulators need to think like a product business, except their objective isn't profit maximising.
That's all, folks. 👋
Remember, if you're enjoying this content, please do tell all your fintech friends to check it out and hit the subscribe button :)
(1) All content and views expressed here are the authors' personal opinions and do not reflect the views of any of their employers or employees.
(2) All companies or assets mentioned by the author in which the author has a personal and/or financial interest are denoted with a *. None of the above constitutes investment advice, and you should seek independent advice before making any investment decisions.
(3) Any companies mentioned are top of mind and used for illustrative purposes only.
(4) A team of researchers has not rigorously fact-checked this. Please don't take it as gospel—strong opinions weakly held
(5) Citations may be missing, and I've done my best to cite, but I will always aim to update and correct the live version where possible. If I cited you and got the referencing wrong, please reach out