Pressure-test a draft survey for leading questions and bad scales before it goes out to real respondents

Review this draft survey instrument for wording problems before it goes out to respondents. I want a question-by-question audit, not a general commentary on survey design.

SURVEY GOAL
Deciding whether to keep or drop the optional onboarding webinar based on perceived usefulness.

DRAFT QUESTIONS (numbered, with response scale noted)
1. Did you find the onboarding webinar helpful and well-organized? (Yes/No) 2. How much did you enjoy using the product in your first week? (1-5 scale)

RESPONDENT POPULATION
New customers within their first 30 days, mostly small business owners with limited time to respond.

KNOWN SENSITIVE TOPIC
Whether they've considered canceling their subscription.

For every question, check it against four specific failure types and only flag the ones that actually apply — don't pad the audit with a pass/fail note on every category for every question:

1. LEADING PHRASING — wording that signals a preferred answer (e.g. "How much did you enjoy..." assumes enjoyment happened).
2. DOUBLE-BARRELED — asking two things in one question so a single answer can't be interpreted (e.g. "Was the onboarding fast and clear?" — fast and clear can diverge).
3. SCALE MISMATCH — a response scale that doesn't fit the question's actual range of possible answers, including scales missing a genuine "does not apply" option where one is needed.
4. SENSITIVE-TOPIC HANDLING — if Whether they've considered canceling their subscription. applies to a question, check whether the phrasing and answer options let a respondent answer honestly without visible judgment, and whether an opt-out is available.

For every flagged question, give the specific rewrite, not just the diagnosis — a respondent-facing survey needs the fixed wording ready to drop in, not a description of what's wrong with it. If a question has no issues, just list it as clean; do not manufacture a minor note to justify commentary on it.

After the question-by-question audit, do one pass across the whole instrument for question-order effects — whether an earlier question could prime how New customers within their first 30 days, mostly small business owners with limited time to respond. answers a later one — and flag only genuine ordering risks, not hypothetical ones.

OUTPUT FORMAT
A table: question number, issue type (or "clean"), and rewrite where applicable. Followed by a short list of any order-effect risks across the instrument as a whole.