Browse Problems

151 problems found

+ Create Problem
support

Identify feature requests that multiple customers are asking for

5 customers this week mentioned 'bulk export'. 3 mentioned 'dark mode'. Need to aggregate these requests and surface: 'Bulk export: 5 requests this week, 23 total, high-value customers'.

1 solution0·Models: gpt-4
support

Route tickets to the right team based on technical complexity

Tier 1 support can handle password resets. Tier 2 handles API errors. Engineering handles database corruption. Need routing that understands: 'Error 500 when calling /api/webhooks' goes to engineering, not tier 1.

1 solution0·Models: gpt-4
support

Extract actionable bug reports from rambling customer messages

Customer writes 3 paragraphs about their frustration. Buried in there: 'When I click Export on the Reports page, nothing happens'. Need to extract: feature, action, expected result, actual result - and ignore the emotional venting.

1 solution0·Models: gpt-4
support

Generate context-aware auto-responses that don't sound robotic

Customer asks: 'Why was I charged twice?' Auto-response needs to: (1) acknowledge the specific issue, (2) explain next steps, (3) sound human. Generic 'We received your ticket' responses make things worse.

1 solution0·Models: gpt-4
support

Detect SLA breach risk before the deadline hits

SLA says 'respond within 4 hours'. It's been 3 hours and the ticket is still unassigned. Need alerts that predict: 'This ticket will breach SLA in 45 minutes unless someone picks it up now'.

1 solution0·Models: gpt-4
support

Classify support tickets by churn risk, not just sentiment

Sentiment analysis says 'negative', but that's not enough. Need to know: Is this 'frustrated but fixable' or 'already decided to cancel'? Looking for prompts that detect churn signals like 'looking at alternatives', 'cancellation policy', or 'export my data'.

1 solution0·Models: gpt-4
finance

Calculate lifetime value (LTV) that accounts for actual churn patterns

Generic LTV formulas assume constant churn. Real SaaS has: high churn in month 1, stabilization by month 6, enterprise customers who never churn. Need LTV calculations that use cohort-specific churn curves, not averages.

1 solution0·Models: gpt-4
finance

Identify tax-deductible expenses from mixed personal/business transactions

Indie founders use personal cards for business expenses. Need to scan transactions and flag: 'This $47 Uber to the airport for a conference is deductible' vs 'This $47 Uber to the grocery store is not'.

1 solution0·Models: gpt-4
finance

Flag high-risk transactions before they become chargebacks

Certain patterns predict chargebacks: first-time customer + high-value purchase + VPN + disposable email. Need to score transactions in real-time and flag the risky ones before fulfillment.

1 solution0·Models: gpt-4
finance

Calculate SaaS revenue recognition for annual plans under ASC 606

Customer pays $12,000 upfront for annual plan. Can't recognize it all in month 1 - need to spread it over 12 months per ASC 606. Need prompts that handle: partial months, mid-month starts, upgrades, downgrades, and refunds.

1 solution0·Models: gpt-4

Showing 4150 of 151 problems

PreviousNext