Conditional questions that can’t show the wrong thing

Brief Builder lets freelancers build a client questionnaire where some questions only appear when an earlier answer makes them relevant. Branching like this is easy to get subtly wrong. These are the rules it follows.

Five states, not two

A rule that decides whether a question is shown is usually treated as true or false. Brief Builder evaluates it as one of five states: true, false, pending (the answer it depends on isn’t given yet), depends-on-hidden (it depends on a question that is itself hidden) and broken.

A question is shown only when its rule is definitely true. A hidden or unanswered question can therefore never expose the wrong branch.

Only look backwards

A rule may only depend on questions that come earlier in the brief. A reference to a missing or later question is treated as broken, never as true, and it is flagged in the Review step so the freelancer can fix it before sending.

Make deletions explicit

Field names and option values are locked once created, so a rule can’t silently stop matching. Deleting a question, option or section that rules depend on lists everything affected and offers three choices: keep them hidden until fixed, delete them too, or make them always visible. Nothing changes without the freelancer seeing it.

When the medium can’t branch

The brief goes to the client as a fillable PDF, and PDF viewers can’t reliably hide fields. So the PDF prints branching as short hints instead, and the last step, Scenario, lets the freelancer describe the client’s situation: sections that don’t apply are left out of the PDF altogether.

What we take from it

Treat “not known yet” and “broken” as states of their own, show conditional content only when its condition is certain, and when the output format can’t follow your logic, tailor the output rather than pretending it can.

Start here

Building something like this?

Tell us about the product or the problem. We’ll tell you honestly whether we can help and what it would take.