Features How it works Languages Sign in Get started

Looking for a Splitwise alternative? Here's what changes when your group speaks 3 languages

Splitwise works fine until half your group thinks in different currencies and writes notes in different scripts. Here's what to look for instead.

Splitwise has done more than any other app to legitimize tracking shared expenses between non-couples. It's the default for a reason: simple math, friendly UI, decent pricing.

But if you've ever been in a group chat that mixes English, Arabic, Hindi, and Spanish — say, an extended family scattered across continents — you've felt the rough edges. Here's what they are, and what a real alternative needs to do differently.

The problems

1. The interface itself isn't fully translated. Even when the labels are localized, the categories ("groceries", "rent") often stay English-only or get translated awkwardly. People type their expense titles in their own language, and the interface fights them.

2. Currency conversion is bolted on, not native. Splitwise lets you pick a default currency per group, but mixing currencies inside the same group is clumsy. If your aunt in Lebanon paid for the family WhatsApp group in LBP and your cousin paid for the shared dinner in USD, the math gets messy fast.

3. There's no concept of asymmetric trust. Splitwise assumes everyone in the group is equally responsible. In real families, some people pay everything and forgive everything, while others meticulously track every $5. The app treats all participants as peers, even when the actual social dynamic is hierarchical.

4. RTL languages are second-class. Try using Splitwise in Arabic on a phone with right-to-left layout. The amounts still align left, the icons stay LTR, the experience feels like a translation overlay rather than a native UI.

What a real alternative needs

Native multilingual UX from day one. Not bolted on. The whole interface, including category names, debt types, payment methods, and notification copy, should localize cleanly. RTL scripts should rearrange the layout, not just the text.

Multi-currency without ceremony. Each debt should remember the currency it was created in. Settling can be in any currency, with the FX rate either auto-fetched or manually entered. The total view should show both per-currency balances and a single converted total.

Bilateral approval for changes. When two people share a ledger, every modification — adding a debt, recording a payment, forgiving the balance — should require both sides to approve. This guarantees the records can never drift, which is the actual reason people stop trusting expense apps after a few months.

A "loose" mode for unlinked counterparts. Not everyone you owe money to wants to install an app. The system should let you track debts solo, share a polite invite link if you want them on board, and auto-link when they sign up later. Forcing both sides to be on the platform from day one is what kills network effects.

No payment processing. Counterintuitive, but important. The moment an app handles money, it needs KYC, banking partners, fees, regulatory friction. Most international families don't want to send money through a US app — they have their own working channels (Wise, Western Union, hawala, in-person cash). What they want is a shared agreement layer, not a payment rail.

What we built

Solvory is our take. It's free for personal use, ships in 16 languages with full RTL, lets you mix currencies inside one ledger, and has bilateral approval for every change once two users link their accounts. It does not touch money — settling is on you and your counterpart, however you usually do it.

That's a deliberate choice, not a feature gap. Coordination is the hard problem. The actual transfer was already a solved problem before any of these apps existed.

If you're shopping for a Splitwise alternative because the multilingual or multi-currency story is what's breaking down, give Solvory a try — it's the niche we're built for.

If your group is one currency, one language, one continent, you'll probably stick with Splitwise. That's fine; tools should fit the use case.

Read next