The second balance
Naming, placing, and teaching a card wallet that couldn’t be merged.
Endl keeps money in two places. The main balance lives in its own self-custodial wallet and handles remittances, stablecoin movement and fiat transfers, while the money a team spends on cards sits somewhere else, inside the card-issuing partner’s smart-contract wallet.
The balances could have been merged the way some competitors do, by parking the whole treasury inside the card provider’s contract. The team refused, because a customer’s entire balance behind a third party’s smart contract was a risk nobody wanted to pass on. So the split stayed and the problem became design rather than infrastructure. A founder who isn’t deep into crypto opens the app and sees two balances. Which one is my money, why is one empty, and how do I actually spend?
- Role
- UX research, design
- Method
- Competitive study, 5 products
How I studied it
I picked five products living with the same tension, created accounts on each, and used them until I understood how they teach a new user what the second balance is and how to fill it. Phantom and Cypher are crypto wallets that ship a card, Brex is the business banking platform closest to what Endl was building, Slice is an Indian fintech that handles two balances unusually well, and RedotPay is the crypto card that took the path Endl couldn’t.
Brex
The closest match to what Endl was building.
Brex has almost the same shape as Endl’s app, a bottom bar with Home, Cash and Card, and two balances that are never added into one number. The Cash tab shows an available balance with account activity, while the Card tab calls its number “Available to spend” and lists merchant purchases under it.
The detail I kept coming back to was Home. Brex never merges the balances there. It shows a Cash block and a Cards block, each previewing its own number and linking out, and moving money between them has its own named action called Transfer. Home summarises and points, it never flattens.
Phantom
They solved it with a single word.
Phantom is a self-custodial wallet that added a debit card, so its situation mirrored Endl’s, on-chain tokens on one side and spendable money on the other. Their answer was to call the card balance Cash and place it at the top of the token list, away from the volatile assets. Funding it is called Sell Crypto, a verb that explains the relationship on its own. They also show the empty state honestly, a near-zero Cash line that makes you wonder what it is, which turned out to be exactly the hook I wanted.
Cypher
They made the opposite bet and hid it.
Cypher keeps the balances on fully separate screens. Home shows only the holdings and the card money appears nowhere on it. The Card tab labels its number “Available to spend”, the same phrase Brex arrived at independently, and funding is called Load Card, a flow with a token selector, a quote and an auto-load option.
The trade-off is a clean front door with no curiosity hook. Someone who never opens the Card tab never learns the second balance exists. That works for crypto natives and fails for the non-technical founder Endl serves.
Slice
A fiat product that separates two balances better than most banks.
Slice treats its two balances as different kinds of thing. Savings is framed as stored money with a big balance, while the card screen leads with what you have spent this period rather than a stored amount, even though the available limit was the obvious number to show. Because the two screens present two different objects, the question of why there are two numbers never forms.
RedotPay
One balance, and why Endl couldn’t have it.
RedotPay has one balance for everything. Spending and remittance draw from the same pot because the entire balance lives inside the card provider’s smart contract, so there is no naming problem, no placement problem and nothing to explain.
Endl couldn’t copy it because the self-custodial wallet exists precisely so a business’s money stays in infrastructure the company controls. RedotPay was the benchmark for the simplicity Endl traded away, and every decision below exists because of that trade.
What repeats
Every product avoids the word balance for the card. Brex and Cypher independently landed on “available to spend” and Phantom renamed the whole thing Cash, because a second figure labelled balance next to the first reads as broken. They all treat the two balances as different kinds of money too, one you hold and one you spend, with Slice pushing that furthest by framing the card around activity.
Funding always gets its own verb, whether that is Sell Crypto, Load Card, Transfer or Add money. The action that moves money into the second balance carries most of the teaching, so it is named for what it does.
Where they split is whether the second balance appears on home at all. Brex and Phantom show it, scoped and labelled, and let the empty state spark curiosity, while Cypher hides it and bets on intent. For a non-technical founder, visible and clearly labelled is the only option that creates the moment of wondering what it is.
The call was Card Funds
Three names made the shortlist, Available to Spend, Cash, and Card Funds. The first is a description rather than a name and turns clunky inside buttons. Cash is short but collides with future cash-out and cash-back features and skews consumer. Card Funds explains itself on first read, stays scoped to the card forever, and shares no words with Account Balance, so the pair reads as two siblings instead of two copies. Account Balance is the money you hold and Card Funds is the money your team spends.
Placement: both balances live on home at different sizes in different containers. Account Balance stays the hero at full width, and Card Funds sits below it as a compact strip with a card icon, the label, the amount and an Add button. The hierarchy explains the relationship before any copy does.
The strip has two states. At zero it expands into a nudge card with a one-line explanation and an Add Card Funds button, and once funded it collapses into the compact row and gets out of the way, so the education lives in the empty state and disappears when its job is done. On the Cards tab, Card Funds is the hero with Add and Withdraw above the team management tiles and card transactions, with the same label on both screens.
What shipped
Left: the old home screen, where “Card Balance” collided with the main balance. Middle: the updated home with the Card Funds strip. Right: the updated Cards tab.
Something unplanned came out of the split. Because Card Funds is a deliberate separate pot, businesses started using it as an organisation-wide spending budget, one universal cap on team card spend sitting on top of the per-card limits. The constraint that couldn’t be removed became a control the customers wanted.
Conclusion
Later I came up with three options: Card Balance, Card Funds, and a slightly different layout. We finalised Card Funds, because it sounded completely different from Account Balance and it was the clearest thing to explain.
All the stakeholders approved it, and it solved the major issues. The B2B users were comfortable with it after that. They read it as a separate balance they deposit into, so they and their team members can use the cards and have the spend deducted from there.