FindING work

Challenge: High Dropout Rate in Job Application Process

Bluecrew faced a significant challenge: a high percentage of prospective workers abandoned the platform before applying for a job. Only 45% of new signups progressed beyond selecting their first job and initiating the onboarding process.

Discovery: Uncovering the Root Causes

To understand this attrition, I conducted a comprehensive investigation, including:

Worker Interviews: I spoke directly with both new and existing workers to understand their experiences and pain points.
Data Analysis: I examined usage metrics to identify patterns and trends in user behavior.
Stakeholder Interviews: I gathered insights from internal teams to gain a broader perspective on the problem.
Usability Testing: I tested early design concepts with workers to validate assumptions and gather feedback.

Key findings revealed that:

Information Overload: Workers faced a daunting list of 100-200 nearby jobs, but typically only reviewed the first 10-20 before abandoning the search.
Lack of Personalization: Bluecrew's platform treated all workers uniformly, failing to account for individual constraints and preferences. Factors beyond wage, such as location, commute, schedule flexibility (gig vs. long-term), required skills, industry, company culture, and even vaccine mandates, significantly influenced job suitability.
Redundant Listings: Similar jobs with slightly different schedules were listed as separate entries. For example, a warehouse needing workers for two different start dates appeared as distinct jobs, unnecessarily multiplying the number of listings and creating clutter.
Insufficient Information at a Glance: Crucial details, like specific workdays, were hidden within job details, forcing workers to click back and forth, creating a frustrating browsing experience.

Goal: Improved Conversion Through Personalization and Streamlining

The primary objective was to increase the conversion rate of new signups completing their first job application ("Get This Job") from 45% to 60%. This would be achieved by:

Personalizing the Job Search: Tailoring the displayed jobs to match each worker's individual needs and priorities.
Reducing Cognitive Load: Grouping similar jobs and decluttering the interface.
Enhancing Information Visibility: Providing key job details directly in the browse view.

Solution: A Tailored and Efficient Job Search Experience

The solution involved a three-pronged approach:

Personalization Engine: To personalize the experience, I first designed a system to capture individual worker preferences based on the discovery phase insights. These preferences are then fed into a relevance algorithm that dynamically adjusts the job listing sort order and determines the timing and content of notifications about highly relevant opportunities. I collaborated closely with the development team to design this algorithm and establish initial preference weightings.
Intelligent Job Grouping: Building upon the relevance algorithm, I implemented a system to group similar job schedules. The most relevant schedule for each position is displayed prominently in the browse view, mirroring the familiar pattern used in online retail (e.g., grouping clothing items by size and color). This significantly reduced visual clutter and was intuitively understood by workers.
Enhanced Browse View: Key job details identified during discovery – including position title, start date, duration (expressed as the number of shifts), shift times, workdays, and wage – were incorporated directly into the browse view. Usability testing confirmed that this dramatically reduced the need for users to navigate between the browse list and individual job details.

Result: Implementation Underway

The redesigned "Find Jobs" experience is currently under development.