Catch the problems you didn't know your survey had.
It validates the survey you wrote: wording that leans, questions that ask two things, scales that don't fit, and topics your goal needs but nothing measures. Then it suggests fixes, and you decide which to keep.
One PIN, both apps. This Validator and the Survey Analyzer, six months for $49.99 USD.
What it checks
Nine checks, each explained in plain words, with the reason it matters for your results.
Biased wording
Leading and loaded questions, agree-by-default statements, social desirability, and filler like "Overall" that adds nothing.
Two questions in one
"How clear and helpful was it?" One answer to two ideas can't be read. It flags them so you can split or reword.
Scales that don't fit
A "How satisfied…" question with an agree–disagree scale quietly spoils the data. Each section gets one scale, and each choice is explained.
Readability
Grade level for your audience, question by question, so you can see which ones push the reading level up.
Overlap and gaps
Questions that measure the same thing twice, and topics your goal needs that no question covers.
Will it hold together?
Simulated responses estimate whether each group of questions will form a reliable scale, before anyone answers.
What you'll be able to report
Tell it roughly how many people will answer and whether you'll compare groups. When the survey is ready, it recommends how to report the ratings, as topic scores or question by question, and what that gives you, using the Survey Analyzer's own thresholds.
A second opinion
A second AI model reviews every question independently. Each note can show how sure it is: where the reviews agree, and where it's a close call.
Published instruments
Recognises well-known validated question sets and leaves them as written, rather than "improving" them.
How a session goes
What the score means
75 is the line. At 75 and above the real defects are cleared, and the points above it are optional rigour, not problems.
Re-checking the same survey can move the score by a few points, so the app says when a change is just noise.
The file you keep
The saved Word survey travels with your work, to a colleague, a reviewer or your future self.
- Title, date and your survey goal
- A validation summary: score, checks run, and how it got there
- The final questions, ready to load into your survey tool
- Design decisions you own: each trade-off left to you, with the context and your call
Plus the full report as a web page and a findings spreadsheet.
The AI suggests. You decide.
Every change is shown with its reason, and nothing is applied without your tick. Ask how sure it is about any note, or ask about any question. Where a choice is genuinely yours (keeping a sensitive question, say), it says so and records your decision.
What it doesn't do
It doesn't replace collecting real answers, a pilot, or a methodologist on a high-stakes instrument, and it can't promise a response rate. It gives you the checks most teams skip, and a record of why the survey is the way it is.
Ready to check your survey?
One PIN, six months from your first sign-in, both apps. As many surveys and re-checks as you need, with $12 of AI included.