Root-Cause and Fix a Fourfold Peak-Hour Customer Surge at a Gas Station
Company: Google
Role: Product Manager
Category: Analytics & Experimentation
Difficulty: medium
Interview Round: Other
You own a gas station. Over a short period, the number of customers at peak times has jumped to roughly four times its usual level. How would you investigate what is happening, diagnose the cause, and resolve it?
```hint Problem or opportunity?
Extra customers are revenue, so first work out what the surge is actually breaking and whether it will last. Both answers change which fixes make sense.
```
### Clarifying Questions
- When did the jump start, and has it held steady since or kept growing?
- Does the fourfold figure count customers actually served, or every car that arrives, including drivers who see the line and leave?
- What is the surge breaking today: long lines, fuel running low, an overloaded shop, or cars backing up onto the road?
- Is it every peak window or only some (weekday commutes, weekends), and is it fuel sales, shop-only visits, or both?
### What a Strong Answer Covers
- Validating the surge with the station's own data and ruling out counting errors before acting on it
- Localizing the surge by start date, time window, fuel grade, customer type and payment method
- Separating internal causes (price, promotions, visibility) from external ones (competitor closure, road changes, supply scares) and confirming each with a specific check
- Immediate fixes for safety, fuel supply and service speed, then longer-term choices matched to whether the demand is temporary or lasting
- Metrics for throughput, waiting, lost customers and stock-outs, with guardrails for safety and regular customers
### Follow-up Questions
- If you discover your posted price is far below every nearby station, what do you do in the next hour, and what do you check afterward?
- Suppose the cause is a nearby station that has closed for good. Would you add pumps, and what numbers would you need before committing?
- How would you keep some of the new customers once the cause of the surge goes away?
Overview: A product execution case in which a gas station owner sees peak-hour customer volume suddenly jump to about four times normal. It tests validating the data, localizing the surge, separating internal from external causes, and choosing fixes for lines, fuel supply and capacity that fit whether the demand will last.