Survey Intelligence Analyzer: تبدیل پاسخهای خام به بینشها و برنامههای عملی
Why this prompt matters
Survey data has a short shelf life. Leadership makes decisions based on gut feel when analysis is slow or shallow. A thorough synthesis that flags the top pain points, quotes real users, and ranks recommendations by impact turns survey data from a checkbox into a strategic asset.
What we use it for
You ran a 200-person customer satisfaction survey after your product launch and have a spreadsheet of responses — NPS scores, feature ratings, and 80 open-ended comments. You need a structured analysis ready for a board meeting in two hours.
Prompt
Act as a Senior Research Analyst specializing in survey data synthesis and stakeholder communication. CONTEXT: I have conducted a survey on the topic: [SURVEY TOPIC / PRODUCT / PROGRAM NAME] Survey size: [NUMBER] respondents Survey period: [DATE RANGE] Primary audience for this report: [EXECUTIVES / PRODUCT TEAM / HR / OTHER] SURVEY DATA: [PASTE YOUR SURVEY QUESTIONS AND RESPONSES HERE — include question text with each set of responses. For quantitative questions include totals or percentages. For open-ended questions include the actual responses or a representative sample.] TASK: Produce a complete Survey Intelligence Report with the following sections: 1. EXECUTIVE SUMMARY (3–4 sentences: who responded, top finding, top risk, top opportunity) 2. QUANTITATIVE HIGHLIGHTS - Key metrics with percentages - Notable score changes vs benchmark (if applicable) - Distribution anomalies worth flagging 3. SENTIMENT ANALYSIS - Overall sentiment: positive / neutral / negative with percentage breakdown - Sentiment by segment (if data allows): [SEGMENT VARIABLE e.g., department, plan tier, region] - Strongest positive signal - Strongest negative signal 4. KEY THEMES FROM OPEN-ENDED RESPONSES - List the top 5 themes with frequency count and representative verbatim quotes for each - Flag any themes that appear in both positive and negative contexts 5. SURPRISING OR COUNTER-INTUITIVE FINDINGS - Results that contradict assumptions or prior surveys 6. PRIORITIZED ACTION PLAN - 5 specific recommendations ranked by: (Impact × Urgency) / Implementation Difficulty - For each: what to do, who owns it, and a suggested success metric CONSTRAINTS: - Do not invent data or fabricate quotes — only use what is provided - Flag explicitly when a finding is based on a small sample (under 10 responses) - Avoid vague recommendations like "improve communication" — every action must be specific and measurable - Keep the executive summary to 4 sentences maximum OUTPUT FORMAT: Use the section headers above. Use bullet points within sections. Bold key findings. Include a section break between each major section.
Result
Executive Summary
Survey: Q2 2026 Customer Satisfaction Survey — CloudBridge Pro (N=187, April–May 2026, primary audience: Product and GTM leadership)
Overall satisfaction sits at 71% positive — 8 points below the industry benchmark of 79% for B2B SaaS platforms at this price tier. The strongest positive signal is onboarding quality (88% satisfaction), driven by the new in-app guided setup flow launched in March. The most urgent risk is API reliability: 34% of power users flagged downtime or latency issues as their top complaint, with three verbatim responses specifically mentioning considering alternatives. The highest-impact opportunity is expanding the native Salesforce integration, mentioned unprompted by 41 respondents.
Quantitative Highlights
- NPS Score: 32 (down from 41 in Q4 2025) — driven by detractor growth in the Enterprise tier
- Feature satisfaction: Reporting dashboard 62% | Onboarding 88% | API 54% | Mobile app 48% | Support response time 79%
- Renewal intent: 81% "definitely" or "probably" renewing — but only 61% in the 50–200 seat segment
- Distribution anomaly: Satisfaction scores are bimodally distributed — 43% rate overall experience 9–10/10, while 28% rate it 1–4/10. The middle is thin, suggesting a polarized user base.
Sentiment Analysis
- Overall: 71% positive / 16% neutral / 13% negative
- By tier: Starter 84% positive | Growth 73% positive | Enterprise 58% positive
- Strongest positive signal: Onboarding and customer success responsiveness — "best onboarding I've experienced in 10 years of SaaS" (VP Operations, 80-seat account)
- Strongest negative signal: API stability — "We've had three incidents this quarter that caused missed SLAs with our own clients" (CTO, 200-seat Enterprise account)
Key Themes from Open-Ended Responses
Theme 1 — API Reliability (n=63, 34%): Latency spikes, occasional 502 errors during peak hours. Quote: "The product is great when it works. The 'when it works' part is the problem."
Theme 2 — Salesforce Integration Gaps (n=41, 22%): Bi-directional sync requested; current read-only integration not meeting workflow needs. Quote: "We're manually copying data between systems — defeats the purpose."
Theme 3 — Reporting Flexibility (n=37, 20%): Custom date ranges and exportable raw data consistently requested. Quote: "I'd pay more for a real data export."
Theme 4 — Mobile App Functionality (n=29, 16%): Limited to read-only on mobile; approvals and quick actions requested.
Theme 5 — Pricing Transparency (n=18, 10%): Confusion around seat-count billing and overage charges. (Note: small sample — treat as signal, not statistically significant finding.)
Surprising Findings
- Support response time scores increased despite a 40% rise in ticket volume — suggests the March support team expansion is working ahead of schedule
- The cohort most likely to churn (61% renewal intent) is the 50–200 seat mid-market segment — not the smaller Starter tier as assumed
Prioritized Action Plan
- Resolve API reliability (Owner: Engineering, Q3): Implement rate limiting, dedicated Enterprise infrastructure tier, and real-time status page. Success metric: API-related support tickets down 60% by September.
- Launch bi-directional Salesforce integration (Owner: Integrations team, Q3): Scope write-back capability for the top 5 field types. Success metric: 50% of Salesforce-connected accounts using bi-directional sync within 90 days of launch.
- Ship custom reporting + CSV export (Owner: Product, Q3): Flexible date ranges and raw export. Success metric: Reporting satisfaction from 62% to 75%+.
- Mid-market retention program (Owner: CS, immediate): Proactive outreach to all 50–200 seat accounts with renewal in the next 90 days. Success metric: Renewal intent in segment from 61% to 75%+.
- Mobile approvals MVP (Owner: Mobile team, Q4): Add approve/reject actions for top 3 workflow types. Success metric: Mobile DAU up 25%.
جمعآوری دادههای نظرسنجی بخش آسان ماجراست. اما درک صدها پاسخ باز و یافتن الگوهای معنادار در دادههای کمی کار دشواری است. تبدیل این دادهها به توصیههایی که تیم رهبری واقعاً بر اساس آن عمل کند، همان جایی است که بیشتر نظرسنجیها متوقف میشوند.
این پرامپت مانند یک تحلیلگر تحقیقاتی ارشد عمل میکند. شما دادههای نظرسنجی خود (یا خلاصهای از پاسخها) را در آن قرار میدهید و در مقابل یک گزارش تحلیل ساختاریافته شامل نکات آماری برجسته، تفکیک احساسات (sentiment breakdown)، موضوعات کلیدی از پاسخهای کیفی، و یک فهرست اقدام اولویتبندی شده دریافت میکنید. این پرامپت برای نظرسنجیهای رضایت مشتری، نظرسنجیهای مشارکت کارکنان، فرمهای بازخورد محصول، ارزیابی کنفرانسها و تحقیقات بازار کاربرد دارد.
پرامپت
چه زمانی از آن استفاده کنیم
این پرامپت را پس از جمعآوری پاسخهای نظرسنجی استفاده کنید، زمانی که نیاز دارید نتایج را به سرعت برای یک ارائه، بهروزرسانی هیئت مدیره، یا جلسه استراتژی داخلی ترکیب کنید. این پرامپت بهویژه زمانی ارزشمند است که ترکیبی از پاسخهای کمی (مقیاس لیکرت، چند گزینهای) و کیفی (باز) دارید و نیاز به تحلیل همزمان آنها دارید.
چه چیزی این پرامپت را مؤثر میکند
بیشتر افراد از هوش مصنوعی میخواهند که «نظرسنجی من را خلاصه کن». این کار یک خلاصه عمومی تولید میکند. اما این پرامپت مدل را به نقش یک تحلیلگر با خروجیهای مشخص وارد میکند: الگوهای آماری، قطبیت احساسات (sentiment polarity)، خوشهبندی موضوعات، نکات برجسته نقلقولهای عیناً از پاسخدهندگان (verbatim highlights)، و یک برنامه اقدام رتبهبندی شده. هر بخش الزامی است، نه اختیاری. خروجی به اندازهای ساختاریافته است که بتوان آن را مستقیماً در یک ارائه اسلاید یا یادداشت اجرایی قرار داد.
سازگار کردن با دادههای شما
برای مجموعه دادههای بزرگ (بیش از 500 پاسخ)، یک نمونه نماینده یا خلاصه تجمیعشده را قرار دهید — مدل زمانی که با دادههای از پیش تجمیعشده کار میکند، بهتر استدلال تحلیلی انجام میدهد تا با ردیفهای خام. برای نظرسنجیهای کوچکتر (کمتر از 100 پاسخ)، همه چیز را قرار دهید. همیشه متن سؤال را همراه با پاسخها قرار دهید تا مدل بفهمد چه چیزی پرسیده شده است.