Why has Instawork's Pro acceptance rate dropped by 15% over the past month for hospitality shifts?

Introduction

Instawork's Pro acceptance rate for hospitality shifts has dropped by 15% over the past month, indicating a significant issue in the platform's core functionality. This decline directly impacts the platform's ability to match workers with available shifts, potentially affecting both business clients and gig workers. To address this problem, I'll employ a systematic approach to identify, validate, and resolve the root cause while considering both immediate and long-term implications.

Step 1

Clarifying Questions (3 minutes)

• Looking at the timing, I'm thinking there might be seasonal factors at play. Has this drop coincided with any particular events or time of year?

Why it matters: Seasonal trends could explain the shift and inform our solution approach. Expected answer: No significant seasonal events noted. Impact on approach: If seasonal, we'd focus on adjusting for cyclical patterns; if not, we'd investigate other factors.

• Considering user segments, I'm curious about the distribution of the decline. Is the 15% drop uniform across all Pro types, or is it concentrated in specific segments?

Why it matters: Identifying affected segments helps narrow down potential causes. Expected answer: The drop is more pronounced among experienced Pros. Impact on approach: We'd focus on issues specific to experienced Pros if the decline is concentrated.

• Thinking about recent changes, have there been any updates to the shift allocation algorithm or Pro onboarding process in the past month?

Why it matters: Recent changes could directly impact acceptance rates. Expected answer: A minor update to the shift recommendation system was implemented. Impact on approach: We'd scrutinize the update's impact and consider rolling back if necessary.

• Regarding the metric itself, has there been any change in how we calculate or track the Pro acceptance rate?

Why it matters: Ensures we're comparing apples to apples and not facing a data anomaly. Expected answer: No changes to the metric calculation or tracking. Impact on approach: If changed, we'd need to recalibrate our analysis; if not, we can proceed with confidence in the data.