SayPro GPT Prompt Execution Framework
Purpose: Automate and streamline the uploading, processing, and analysis of prompts using GPT to extract 100 qualitative or thematic responses per prompt.
🧾 Step 1: Prepare Prompt Submission Sheet
Use the SayPro GPT Prompt Submission Template, including these fields:
| Prompt ID | Prompt Text | Target Group | Focus Area | Submitted By | Date Uploaded |
|---|---|---|---|---|---|
| 001 | “List 100 user feedback themes…” | Users | Digital Access | E. Mataban | 2025-01-10 |
⚙️ Step 2: Upload Prompts into GPT System
Options:
- Use OpenAI API or ChatGPT web interface for manual entry.
- Automate through a script if using bulk input (e.g., Python + OpenAI API).
Example Code Snippet (Python):
pythonCopyEditimport openai
prompt = "List 100 suggestions from users on improving mobile course delivery..."
response = openai.ChatCompletion.create(
model="gpt-4",
messages=[{"role": "user", "content": prompt}],
temperature=0.7
)
results = response['choices'][0]['message']['content']
📥 Step 3: Collect and Store 100 Responses
For each prompt:
- Ensure exactly 100 structured responses are returned.
- Save results in a CSV or Excel format using the SayPro GPT Response Tracker Sheet:
| Prompt ID | Response No. | Theme/Insight |
|---|---|---|
| 001 | 1 | Users want downloadable content. |
| 001 | 2 | Interface feels cluttered on mobile. |
| … | … | … |
📊 Step 4: Review & Clean Responses
- Eliminate duplicates or vague entries.
- Group responses into categories: UX/UI, Trust, Accessibility, Support, etc.
- Tag top responses with priority levels (High, Medium, Low).
📈 Step 5: Analyze Trends Across Prompts
- Use pivot tables or dashboards to visualize recurring feedback.
- Feed top insights into SCRR reporting formats or focus group moderation guides.
🗂️ Optional Output Formats:
- PDF Report per prompt
- Integrated Focus Group Summary Sheet
- SayPro Monthly SCRR Insight Matrix
Leave a Reply