Customer SuccessGPT-3.5 Turbo

Churn Predictor Using GPT-3.5 Turbo

Optimized prompts and cost breakdowns for running churn predictor with GPT-3.5 Turbo.

About GPT-3.5 Turbo

Provider: openai
Context Window: 16K tokens
Input Cost: $0.5 / 1M tokens
Output Cost: $1.5 / 1M tokens

Best For

Legacy supportSimple classificationQuick responses

Optimized Prompts for GPT-3.5 Turbo

These prompts are tailored to work best with GPT-3.5 Turbo's capabilities

analyze

Analyze this customer's churn risk:
{{customer_data}}

Usage trends: {{usage_trends}}
Support tickets: {{support_history}}

Provide:
1. Churn risk score (1-100)
2. Key risk factors
3. Recommended interventions
4. Urgency level

Cost Breakdown

Estimated costs for running this workflow with GPT-3.5 Turbo

input tokens: ~916 tokens
output tokens: ~927 tokens
cost per request: $0.0040
monthly 1000 requests: $31.12

Rate Limiting Considerations

Consider batching requests during peak hours Check provider rate limits for GPT-3.5 Turbo

Monitoring with Administrate

Track latency and token usage for GPT-3.5 Turbo in Administrate dashboard