Each test should verify a single behavior so a failure points to exactly one cause.
Multi-concept tests obscure which behavior regressed and stop at the first failed assertion.
BeforeMerge scans your pull requests against this rule and dozens more. Get actionable feedback before code ships.
A test that checks many unrelated things fails for many reasons and stops at the first failing assertion, hiding later problems. One logical concept per test means a red test names exactly what broke.
// Bad — multiple concepts in one test
expect(parse(input).name).toBe("Ada");
expect(format(other)).toBe("X");
// Good — split by concept
it("parses the name", () => expect(parse(input).name).toBe("Ada"));
it("formats the value", () => expect(format(other)).toBe("X"));