Dashboard Design

B2B Enterprise

HR Tech

Shipped

Turning Status Chaos Into a Real-Time Dashboard

ROLE

Product Designer

TEAM

+1 Product Designers | Product Manager | Development Team | CEO

TIMELINE

2.5 Weeks

WHAT I DID

User Research | Wireframing | Prototyping | Status Flow Mapping | Stakeholder Management

GOAL

A Dashboard to Replace Manual Daily Reporting

Customers had no real-time visibility into interview progress. Admins spent 2+ hours daily compiling reports that weren't part of their core role.

CHALLENGE

Statuses & Metrics Were Too Ambiguous to Visualize

Over a few days, I noticed the same questions surfaced in every meeting:

CHIEF TECHNICAL OFFICER

What's the difference between 'In Progress' and 'Ongoing'?

FOUNDER / CEO

Isn't 'Interview Conducted' and 'Completed' the same thing?

DESIGN

I Built a Shared Status Vocabulary & Designed Around It

View Prototype

01

DEFINED STATUSES

Defined a shared status vocabulary across admin and customer views.

02

SPLIT LIVE VS. CUMULATIVE METRICS

Separated live pipeline from cumulative outcomes.

03

SURFACED UNHAPPY PATHS

No-shows, Incompletes, Drop-offs, Status Changes

IMPACT

40+

Statuses Streamlined

Unified terminology across the platform, resolving recurring confusion for customers and internal teams.

2+

Hours Saved Daily

Admins returned to core duties; reports now generated in a few clicks.

PROCESS

I Worked With Product, Engineering, and with the Founder to Define Our Metrics

Stakeholder Interviews

Surfaced recurring confusion around status terminology.

Artifact Analysis

Identified which metrics and statuses customers actually tracked.

Status Audit

Mapped every existing status across three portals.

Ideation & Stress Testing With AI

Used Figma Make for dashboard ideation, ChatGPT for label alternatives, and Claude to pressure-test the final status taxonomy.

Claude

chatGPT

Figma Make

MVP SCOPING

Deciding What Not to Show To Meet Deadlines

Several metrics were cut from the concept dashboard because the underlying data didn't exist.

STATUS WORK

Building a Vocabulary that Could Hold the Product's Complexity

01

Labels that Explain What the User Needs

OPTIONS CONSIDERED

Pending Interview

Ongoing

In Progress

Scheduled

Scheduled

The interview has been set. A waiting period until the interview takes place.

OPTIONS CONSIDERED

Interview Conducted

Completed

Awaiting Report

Report Pending

Awaiting Report

Signals that the interview has concluded and the report is under way.

Completed

Signals that the interview and report is done. The request is fully resolved.

Master Templates

Dropped

AI-generated questionnaires made the efficiency value of Master Templates null.

Objectives

Dropped

This field was left blank in 99% of questionnaires. The objective was always to assess the candidate’s skills.

Question Bank

Kept

A fail-safe for reliable, expert-vetted interview questions in case AI outputs fell short.

Answer Guides

(Previously Ideal Answers)

Changed

AI needs a guide to gauge what a right answer looks like. Furthermore, ideal answers negate that a question could have multiple right answers.

Master Templates

Dropped

AI-generated questionnaires made the efficiency value of Master Templates null.

Objectives

Dropped

This field was left blank in 99% of questionnaires. The objective was always to assess the candidate’s skills.

Question Bank

Kept

A fail-safe for reliable, expert-vetted interview questions in case AI outputs fell short.

Answer Guides

(Previously Ideal Answers)

Changed

AI needs a guide to gauge what a right answer looks like. Furthermore, ideal answers negate that a question could have multiple right answers.

02

Collapsed Granular Statuses

Incomplete: No Show

BILLING IMPACT

Candidate No Show

Flat fee applies during invoicing. Flags candidate reliability.

Incomplete: Dropped Off

Incomplete: Withdrew

Incomplete: System Bug

Incomplete: Interviewer Emergency

Incomplete: Candidate Emergency

QUALITY IMPACT

Incomplete

Specific cause logged under review details.

03

Different Status Labels for Different Users

UNDERLYING STATE

ADMIN SEES

CUSTOMER SEES

Scheduling interview between candidate + interviewer

Availability Recieved

Interviewer Request Sent

Interviewer Declined

Interviewer Requested Reschedule

Interviewer Cancelled

Matching Interviewer

Interview Confirmed

Interviewer Accepted

Scheduled

Working on the Feedback Report

Report Received

Sent for Changes

Report Updated

Approved

Awaiting Report

Interview & Report Submitted

Sent to Company

Completed

Master Templates

Dropped

AI-generated questionnaires made the efficiency value of Master Templates null.

Objectives

Dropped

This field was left blank in 99% of questionnaires. The objective was always to assess the candidate’s skills.

Question Bank

Kept

A fail-safe for reliable, expert-vetted interview questions in case AI outputs fell short.

Answer Guides

(Previously Ideal Answers)

Changed

AI needs a guide to gauge what a right answer looks like. Furthermore, ideal answers negate that a question could have multiple right answers.

Master Templates

Dropped

AI-generated questionnaires made the efficiency value of Master Templates null.

Objectives

Dropped

This field was left blank in 99% of questionnaires. The objective was always to assess the candidate’s skills.

Question Bank

Kept

A fail-safe for reliable, expert-vetted interview questions in case AI outputs fell short.

Answer Guides

(Previously Ideal Answers)

Changed

AI needs a guide to gauge what a right answer looks like. Furthermore, ideal answers negate that a question could have multiple right answers.

PRIORITIZING METRICS

Screenshots Showed What Customers Actually Tracked

01

Candidate Quality and Volume Trends

02

Visibility Into Mid-Flow Disruptions

03

Preferred Benchmark Report Data

Matching the Customers' Needs

01

Surfaced Problem Statuses

Broke down the "In Progress" tile to surface disruptions that affected operations

02

Comparison Rates Graph

Showed candidates sent, interviewed, and pass rate over time in a single view.

REFLECTION

What I Learned

Diagnose Before You Design

The real problem is rarely the one in the brief.

Indirect Signals Are Still Research

Meeting patterns and screenshots can replace formal user interviews when access isn't available.

Work shown with permission from respective companies.

© 2024 Hritika Dharaskar.

Create a free website with Framer, the website builder loved by startups, designers and agencies.