Designing the aha moment & increasing

activation by 27%

MRR ↑

Received first annual
subscriptions & interest
from enterprise customers

Received first annual subscriptions & interest from enterprise customers

27%

Boosted first-session
engagement 8 → 35%

Boosted first-session engagement 8→ 35%

4,6

Improved dashboard
user-friendliness (SEQ)

Improved dashboard user-friendliness (SEQ)

Domain

B2B
HRTech
AI

Users

Recruiters
HRDs
Hiring Managers
Candidates

Team

Product Owner
Engineers
HR Consultant
Business Developer
Founding Product Designer (me)

What I did

Design audit
User research
Key product design
UX/UI direction
User testing

Domain

B2B, HRTech, AI

Users

Recruiters, HRDs,
Hiring Managers, Candidates

Team

PM, Engineers, HR Consultant, Business Developer, Founding Product Designer (me)

What I did

Design audit, User research, Key product design, UX/UI direction, User testing

Domain

B2B
HRTech
AI

Users

Recruiters
HRDs
Hiring Managers
Candidates

Team

Product Owner
Engineers
HR Consultant
Business Developer
Founding Product Designer (me)

What I did

Design audit
User research
Key product design
UX/UI direction
User testing

Context & Problem

Potis.ai is an AI startup that helps companies run more structured and fair candidate interviews. When I joined as the Founding Product Designer, the product was already gaining traction,

but there was a critical funnel issue: many users signed up but dropped off before reaching the core value.

Business Requirements

These problems mapped directly to what the business needed:

  • Increase adoption and engagement

  • Improve time-to-market

  • Strengthen positioning for investors

  • Establish a cohesive experience across the website, registration fand dashboard

  • Make the look and feel competitive with the top players in the market

The old dashboard

The old dashboard

Research

To understand why users were dropping off, I combined qualitative research with behavioral data.

Key insight: users couldn't see what to do on the dashboard, or why the product was worth their time.

Qualitative

Qualitative

Conducted 7+ deep-dive user interviews and usability tests

Conducted 7+ deep-dive user interviews and usability tests

Conducted 7+ user interviews and usability tests

Quantitative

Quantitative

Analyzed Hotjar recordings and funnel metrics

Analyzed Hotjar recordings and funnel metrics

Hypothesis

Research showed that users left because they couldn't see the payoff before the work.

If users can explore real candidate results before investing setup time, more of them will reach their first candidate report — the aha moment.

We defined the "aha moment" simply: the moment a user sees their first role with real candidate results.

Redesign, not small fixes

Before the redesign, almost no one reached that moment, so the goal wasn't "make
the dashboard better." It was to get more users there, faster.

First I tried small fixes: tooltips, a reordered menu, new empty-state copy. None moved the numbers. The dashboard had no clear first step, and patches on a screen with no main path only added noise.

Before the redesign, almost no one reached that moment, so the goal wasn't "make the dashboard better." It was to get more users there, faster.

First I tried small fixes: tooltips, a reordered menu, new empty-state copy. None moved the numbers.
The dashboard had no clear first step, and patches on a screen with no main path only added noise.

First I tried small fixes: tooltips, a reordered menu, new copy on the empty state. None of them moved the numbers.
The dashboard had no clear first step, and patches on a screen with no main path only add noise.

Design Principles

These principles guided the redesign and helped users reach value faster.

Deliver value before asking for effort

Users can explore a demo role and see candidate results before creating their own role, reducing upfront effort and accelerating time-to-value

Before

After

Progressive disclosure with single-focus views

Progressive disclosure with
single-focus views

The role results screen surfaces information progressively and gives every view
a single clear focus instead of competing blocks

The role results screen surfaces information progressively and gives every view a single clear focus instead of competing blocks

Before

After

Guided onboarding that keeps users oriented

The onboarding flow is designed to continuously orient users and guide them toward their first meaningful result, reducing uncertainty and helping them maintain momentum throughout the setup process.

A first screen after sign-up

A side panel with weighted steps

Final design

Dashboard: First Value & Activation

The redesigned dashboard meets users at every stage:

  • Active state — roles scale into a clean, scannable grid with status tags, "Create new" always in reach.

  • First session — a pre-filled demo role to explore, with "Create your first role" as the clear next step.

  • Empty state — an encouraging prompt and profile checklist guide users to their first action, not a dead end.

Role overview

The layout splits into two areas: a left panel keeps key actions and team prompts always within reach, while the main area stays focused on role content without competing for attention.

Role results

Results are shown in two views: a table for detailed review and a heatmap for fast comparison. Color intensity reflects skill proficiency, enabling quick identification of strengths and gaps.

Role results — "Started" view

Switching the status filter to "Started rokes" surfaces candidates mid-test, with columns built for that moment.

Switching the status filter to "Started rokes" surfaces candidates mid-test

Design Validation

During testing, I conducted ~15 moderated usability sessions across prototypes and the live product after soft launch.

Users completed core scenarios such as creating a job and reviewing candidate results, which allowed us to uncover friction points, validate assumptions, and stress-test product logic in real conditions.

Users completed core scenarios such as creating a job and reviewing candidate results, which allowed us to uncover friction points, validate assumptions, and stress-test product logic in real conditions.

Results & Impact

The redesign shortened the path between sign-up and value realization. Users reached meaningful outcomes faster, leading to:

27%

Boosted first-session
engagement 8 → 35%

4,6

Improved dashboard user-friendliness (SEQ)

4,1

Improved dashboard
UX (CSAT)

Beyond UX improvements, the redesign strengthened commercial outcomes and generated interest from enterprise customers

MRR ↑

Received first annual
subscriptions & interest
from enterprise customers

WAU

Improved time to Value →
More active users → Growth

What I learned

Before I assumed users needed better onboarding, after I learned that users needed evidence of value. The problem wasn't navigation. It was motivation.

next / Scaling VMS for operational complexity

next / Scaling VMS
for operational complexity

ana sushok © 2026