About Us

The tool currently has 11 built-in rules, each written in advance and tied to one specific situation — for example, "check-cashing costs are above a set threshold, and the person has an accepted form of identification." Every rule looks only at the answers relevant to it and simply checks whether its condition is met. Nothing is estimated or guessed at this step; it is a yes-or-no check against the information entered.

If a person's answers satisfy more than one rule, more than one possibility is shown — the tool does not pick just one and hide the rest. Each possibility is produced independently by its own rule, so the full list reflects everything that applies, not a single "best guess."

Possibilities are displayed in order of the estimated annual cost each one relates to, largest first, calculated only from the figures entered. That ordering is arithmetic on a person's own numbers, on a single scale, and is not a judgment about which option is better or more advisable for them. Possibilities that have no dollar cost attached — gathering identification, or speaking with a caseworker — appear at the end of the list, which reflects only the absence of a fee to measure against, not lower importance.

An earlier version of this tool ordered possibilities by an internal 0–100 "match score" and displayed that number to the user. That has been removed. Each rule computed its score on its own scale, so the scores were not comparable between rules — one remittance rule could never exceed 58 while others could reach 100 — which meant the resulting order partly reflected internal constants rather than anything about the person's situation. Ordering by cost avoids that problem entirely.

Some possibilities address the same underlying cost — for example, two different account-opening possibilities might both target check-cashing fees. When that happens, only one of them is credited with that savings, and the other is marked as already included. This keeps the "illustrative annual reduction" figure shown on the calculator from double-counting the same dollar under two different possibilities. Because both possibilities in such a pair are measuring the identical fee, the choice of which one carries the figure is made by a fixed internal tie-break rule; it is not a statement that one option is better than the other.

If too many questions are left blank, or if the answers given don't clearly match any of the specific rules, the tool does not guess at a possibility. Instead, it shows a general note suggesting a conversation with a financial counselor, along with a plain explanation of why — for example, that a number of questions were left unanswered.

Each possibility shown includes a "Why this appeared" section, listing the specific facts from a person's answers that caused that rule to apply — for instance, that check-cashing was reported and the estimated annual fees crossed the relevant threshold. These explanations are pulled directly from the same rule that produced the possibility; they are not written separately or added afterward.