The research literature is unusually direct about the stakes. Using longitudinal data on 4,574 couples, Dew, Britt and Huston found that financial disagreements were the strongest disagreement type predicting divorce, stronger than disagreements over children, chores, or in-laws, and that they fully mediated the link between a perception of financial inequity and the hazard of divorce.^[5] The operative variable is not how much money a couple has. It is whether the arrangement feels fair, and whether the argument keeps recurring.

That finding has a hopeful counterpart. Olson, Rick, Small and Finkel ran a two-year randomized longitudinal experiment on 230 engaged and newlywed couples, assigning some to merge money in a joint account. Couples in separate-account and no-intervention conditions showed the normal decline in relationship quality over the first two years of marriage. Couples assigned to a joint account did not.^[6] The structure of the money arrangement caused a difference in the relationship.

This is the thesis in one sentence: how shared money is organized changes how people treat each other, which means the tool that organizes it is an intervention, not a spreadsheet.

A joint account is not available to everyone, and not wanted by everyone. Cohabiting couples pool far less than married ones, and the research attributes that to real, rational factors. Dual earning, shorter union duration, no joint children, uncertainty about the relationship's future.^[7] Roommates and family members obviously aren't merging accounts at all. What all of them need is the thing a joint account provides socially: a standing, agreed, unspoken-but-settled rule about who carries what, without merging the money itself.

That's what a persistent split ratio is. It's the smallest possible version of a shared financial agreement.

And it has to be automatic, because the alternative is talking about it. Money is measurably the hardest subject to raise: in a field experiment with 755 employees, Cullen and Perez-Truglia found people are both unwilling to reveal their own pay and reluctant to ask about anyone else's.^[8] Recent consumer research finds that financial stress itself shapes, and suppresses, how couples communicate about money, exactly when communication would help most.^[9] And the sociology of money has held for thirty years that people do not treat money as fungible; they earmark it with social meaning, which is why "you owe me $43.50" is never only an arithmetic statement.^[10]

Splitwise is a group ledger. Its primitive is the transaction inside a group, and its centre of gravity is the trip. A temporary set of people, an even division, and a settlement at the end. It's excellent at that. But percentage fairness there is a per-expense adjustment rather than a standing property of the household, receipt scanning and an ad-free experience live in the paid tier, and the free tier is where most households would sit.

Honeydue is a couples' visibility layer. Its primitive is the bank account: link your institutions, choose what your partner can see, set budgets, get bill reminders, chat in-app. It's genuinely couples-first, which is rarer than it should be. But it's a mirror rather than an engine. It shows you balances and budgets; it doesn't hold a fairness rule and settle against it. It also asks for bank credentials up front, which is the single largest trust ask in consumer finance and a hard first step for a household that has not yet agreed on how it splits anything.

Have Another Cherry takes the ratio itself as the primitive. One agreement, set at setup, applied automatically to every shared expense, for a group that is permanent rather than temporary, with no bank linking required to get value on day one.

She is a Ph.D. candidate in Communication at the University at Albany, SUNY, working with Dr. Alan Zemel in ethnomethodologically-informed conversation analysis, the empirical study of how people produce and repair understanding in ordinary talk, worked out from recordings of real interaction rather than from surveys or self-report. Her dissertation, Grouping Up: The Achievement of Collectivity in and as the Achievement of Instruction, asks how a collection of individuals becomes a group in the first place, and how instruction is the mechanism that does it.

That research is not decoration on the pitch. It is the product thesis. A shared-expense app is a group-formation problem wearing a ledger's clothes. Shared money breaks down exactly where the talk breaks down. The purchase nobody flagged, the resentment nobody named, the running tally nobody wants to be the one to raise. The literature cited above says the same thing from the outside: financial disagreement is the disagreement that ends marriages, the structure of the arrangement causally changes the relationship, and nobody wants to open the conversation. Have Another Cherry is being built by someone whose actual field is the mechanics of how those conversations succeed or fail.

It is also why Premium is heading where it is: a planned-expense preview that shows the projected split before the purchase, paired with AI-generated conversation starters for the discussion that purchase would otherwise trigger afterward, as an argument. Most money software moves the conversation later, to the statement, the reconciliation, the settle-up. This moves it earlier, to the point where it is still cheap to have.

That operating half matters as much as the research. Shipping software into a restaurant means non-technical staff have to succeed on the first attempt, under service pressure, with no training budget and no patience for a second try. It is an unforgiving standard for onboarding, and it is the standard this product is held to.

Have Another Cherry began at her own kitchen table. She and her partner Robert use it daily, specifying and testing features against their own finances. It was designed for one real relationship before it was designed for a market. What turned a personal fix into a company was listening to friends over the summer describe the identical failure in Splitwise: an income differential the tool simply refuses to model.

She scoped, designed and built the first working version herself with AI-assisted development, then brought on Matt Powers of Embr Intelligence on contract in August 2026 to harden the backend and ship production receipt scanning. The app runs on Firebase and is in invitation-only beta. Situated Strategies' second product, the Making Moves Kitchen Kit, launches this quarter.