Random Interview Question Generator

Why the same questions for every candidate matters

Structured interviews — the same questions, asked in the same way, of every candidate — predict job performance substantially better than unstructured conversation, and they are far easier to defend if a hiring decision is ever challenged. The reason is unglamorous: when every candidate answers a different set, you are comparing impressions rather than answers, and impressions track how much you liked someone.

That is what this tool is for. Draw a set once, use it for the whole panel, and the comparison at the end is between answers to the same question. Drawing a fresh set per candidate is the one usage that undoes the entire benefit.

The category checkboxes have a silent default

Four categories are available — behavioural, technical, leadership and culture fit — and this is worth knowing precisely: if you tick none of them, the tool uses all four. It does not error and it does not return an empty set. So a run where you meant to select "technical" but missed the checkbox looks like a successful draw and quietly gives you a mixed set.

Once categories are chosen, their questions are pooled and shuffled together, so the output is not grouped by category and not balanced between them. Ticking behavioural and technical does not guarantee half from each; a five-question draw can legitimately come back entirely behavioural, because it is one shuffle of a combined pool. Each question is labelled with its category in the output and in the CSV export, so you can see what you actually got.

Use the seed if you want to prove it

The Seed field makes a draw reproducible: the same seed with the same categories and count always returns the same questions in the same order. Write the seed in the hiring notes and anyone can regenerate the exact set later, which turns "we asked everyone the same questions" from a claim into something demonstrable.

Without a seed each draw is fresh and unrepeatable. The question count is capped between one and twenty; asking for more returns twenty.

What the panel still has to bring

• It does not score anything. Structured interviewing needs a rubric agreed before the interviews, and that part is yours.
• It does not know your role, level or industry. A leadership question drawn for a graduate hire is your call to discard.
• It does not check legality. Employment law on what may be asked varies by jurisdiction; review the set before you use it.
• It does not track which candidate got which set. Export the CSV and keep it with your notes.