Prompt: Create a Customer Survey
Jay Banlasan
The AI Systems Guy
tl;dr
Design surveys that customers actually complete and that produce insights you can act on.
Most customer surveys fail twice. First, nobody completes them because they are too long. Second, the results are useless because the questions are vague. "How satisfied are you on a scale of 1-10?" tells you nothing about what to change.
This prompt designs surveys that get completed and produce actionable data. Use it to prompt create customer survey ai drafts that respect your customers' time while giving you real insights.
The Prompt
Design a customer survey for the following purpose:
SURVEY GOAL: [e.g., "Understand why customers churn after 3 months" or "Identify the most valued feature for pricing tier decisions"]
AUDIENCE: [Who is taking this survey]
RELATIONSHIP: [How long they have been customers, what they have purchased]
DESIRED OUTCOMES: [What decisions will this data inform?]
DISTRIBUTION METHOD: [Email, in-app, post-purchase, etc.]
Design the survey with these constraints:
1. Maximum 8 questions (people abandon after that)
2. Maximum 3 minutes to complete
3. Mix of question types:
- 2-3 scaled questions (1-5 or 1-7, not 1-10)
- 2-3 multiple choice with an "Other" option
- 1-2 open-ended questions (placed at the end)
- 0-1 ranking questions (max 5 items to rank)
For each question provide:
- The question text
- Question type and options
- Why this question matters (what decision it informs)
- What a concerning pattern in responses would look like
Also provide:
- SURVEY INTRO: 2 sentences explaining why their input matters and how long it takes
- INCENTIVE RECOMMENDATION: Whether to offer one and what type
- FOLLOW-UP PLAN: What to do with different response patterns
Rules:
- No leading questions ("How much do you love our product?")
- No double-barreled questions ("How satisfied are you with our product and customer support?")
- Scale questions must have labeled endpoints ("Not at all satisfied" to "Extremely satisfied")
- Open-ended questions must be specific enough to get useful answers ("What is one thing we could improve?" not "Any other thoughts?")
- Every question must connect to the stated survey goal. No "nice to know" questions.
The Incentive Question
Small incentive (10% discount, free resource) typically doubles completion rates. But it also attracts responses from people who do not care about the survey and just want the reward. For high-stakes surveys (product decisions, pricing research), skip the incentive and accept a lower completion rate with higher quality responses.
Acting on Results
The follow-up plan is the most neglected part. Before sending the survey, define: "If 60%+ say X, we will do Y." Without pre-committed actions, survey results get discussed in meetings and then forgotten.
The Thank You
After submission, show them you listened. "Based on your feedback, here is what we are changing." This closes the loop and makes them more likely to complete future surveys. Customers who feel heard become advocates.
Build These Systems
Ready to implement? These step-by-step tutorials show you exactly how:
- How to Create an AI Chatbot with Order Lookup - Build a chatbot that looks up orders and provides real-time status updates.
- How to Create Automated Review Request Campaigns - Ask happy customers for reviews automatically at the right moment.
- How to Build a Customer Self-Service Portal - Create a portal where customers resolve common issues without contacting support.
Want this built for your business?
Get a free assessment of where AI operations can replace overhead in your company.
Get Your Free Assessment