User Experience Designer
Group+322.png

VocoVision: A/B Test Reduces Bounce 29%, Lifts Conversions 11%

BlazerWorks Design System — Case Study Sections
Case Study — Experimentation & Entry UX

VocoVision: Guided Search Improves Job Discovery

An A/B test exploring how structured, intent-based search can reduce friction in early job discovery. By introducing a guided, sentence-style search experience, users were able to identify relevant roles faster and move deeper into the funnel.

Role
UX + UI Designer
Industry
Remote Education Staffing
Tools
Figma, Mouseflow, VWO, GA4
Engagement
+27%

Increase in user interaction and hero CTA behavior.

Friction Reduction
-29%

Decrease in abandonment during initial form interaction.

Completion
+11%

Improvement in successful form submissions further down the funnel.

 
 
 
A more guided entry experience increased job seeker conversions while reshaping how users move through the funnel.
 
 

Context

Users were landing, but not orienting

The homepage served as a primary entry point, but users struggled to quickly determine where to go next. Job seekers had to interpret multiple paths before taking action, which created hesitation and drop-off.

Core Issue

The experience required users to translate their intent into navigation decisions instead of supporting that intent directly.


Hypothesis

Reduce friction by guiding intent instead of exposing options

We hypothesized that introducing a structured, sentence-style search would help users self-select by job type and specialty earlier in the experience.

If we...

Replace open-ended navigation with guided, intent-based input

Then...

Users will find relevant roles faster and move deeper into the funnel with less friction

 
 
Approach

Designing a guided entry experience

Instead of presenting multiple navigation choices, the experience was reframed as a single guided interaction. The goal was to reduce decision overhead and align the interface with how users naturally think about job searching.

  • Reframed search as a sentence

    Users could construct their intent through structured inputs rather than navigating multiple entry points.

  • Segmented by job type and specialty

    The system guided users toward relevant categories early instead of requiring downstream filtering.

  • Reduced competing actions

    Primary focus was placed on job seeker pathways to simplify the entry decision.

 
 

 
Experiment

A/B testing guided vs traditional entry

We tested a guided, sentence-style search experience against the existing homepage. The variation introduced structured input for job type and specialty directly within the hero.

 

Results

Stronger job seeker movement, mixed downstream effects

Conversions

+11%

Increase in job seeker applies compared to control

CTA Engagement

+27%

More users engaged with primary entry actions

Bounce Rate

-29%

Fewer users left without interacting

Observed Behavior Shift

Users moved through the experience faster, with shorter sessions but higher completion. This suggests reduced friction rather than reduced engagement.

 
 
 

 
Tradeoffs

Improving one path introduced friction in another

While job seeker performance improved, employer conversions decreased. The new structure deprioritized direct access to the request talent flow.

Key Tension

Optimizing for one audience at the entry point can create unintended friction for another.

 

 
Learnings

Guided experiences reduce friction but reshape behavior

Entry UX

Users respond well to guided input when intent is clear but not easily expressed through navigation.

Behavior

Shorter sessions paired with higher conversions indicate reduced friction rather than disengagement.

System Design

Homepage decisions influence multiple user groups and require careful balance between competing needs.

 
 
 

Next Steps

Refining balance between audiences

  • Reintroduce employer pathways more strategically

    Explore placement outside the primary job seeker flow.

  • Test hierarchy on mobile

    Evaluate visibility and engagement of guided search above the fold.

  • Continue iteration on guided search patterns

    Refine how users construct intent across different specialties.