Speaklarity
Speaklarity
SaaS
MVP
From concept to MVP
From concept to MVP
In March 2025, I joined Speaklarity, an early-stage startup armed with a core concept: to create a powerful AI coach for structured professional communication.
In March 2025, I joined Speaklarity, an early-stage startup armed with a core concept: to create a powerful AI coach for structured professional communication.
My mission was to take the product from zero to a fully functional MVP. The challenge was to translate complex AI feedback into a human-centric experience that encourages users to practice, rather than intimidating them.
My mission was to take the product from zero to a fully functional MVP. The challenge was to translate complex AI feedback into a human-centric experience that encourages users to practice, rather than intimidating them.
Research, UI/UX Design, Prototyping, Testing
Research, UI/UX Design, Prototyping, Testing
Product Designer
Product Designer
4 months
4 months
From concept to MVP
In March 2025, I joined Speakiarity, an early-stage startup armed only with a core concept: to create a powerful AI coach for public speaking.
My mission was to take the product idea from zero, through deep research and hypothesis testing, to a fully functional MVP.
Speaklarity
SaaS
MVP
Research, UI/UX Design, Prototyping, Testing
Product Designer
4 months
About Product
About Product
Overview
Overview
Speaklarity is an AI-powered tool designed to help professionals improve their speaking skills and prepare for interviews.
Speaklarity is an AI-powered tool designed to help professionals improve their speaking skills and prepare for interviews.
While the long-term vision includes sales pitches and public speaking, the MVP focuses on behavioural interviews for top-tier tech companies.
While the long-term vision includes sales pitches and public speaking, the MVP focuses on behavioral interviews for top-tier tech companies.
The Problem & Hypothesis
The Problem & Hypothesis
Candidates often fail not because they lack experience, but because they lack structured practice.
Candidates often fail not because they lack experience, but because they lack structured practice.
My core hypothesis was that by providing instant, structural feedback, the product could motivate users to iterate on their stories, turning vague answers into reliable reflexes.
My core hypothesis was that by providing instant, structural feedback, the product could motivate users to iterate on their stories, turning vague answers into reliable reflexes.
what makes the product different
what makes the product different
Speaklarity is a Habit Engine. Unlike competitors, the product acts as a "Gym Coach." The flow was engineered to trigger a specific behaviour: Active Iteration.
Speaklarity is a Habit Engine. Unlike competitors, the product acts as a "Gym Coach." The flow was engineered to trigger a specific behaviour: Active Iteration.
By forcing the user to close the loop immediately, the interface transforms passive learning into muscle memory.
By forcing the user to close the loop immediately, the interface transforms passive learning into muscle memory.
Record
Analyse
Iterate
This behavioral shift creates a "hook" that drives higher retention and ensures the user actually acquires the skill, rather than just reading about it.
This behavioral shift creates a "hook" that drives higher retention and ensures the user actually acquires the skill, rather than just reading about it.
product value
product value
Speaklarity bridges the gap between knowledge and execution. By providing a safe, simulated environment for repeated failure and rapid improvement, we turn interview anxiety into reliable confidence.
Speaklarity bridges the gap between knowledge and execution. By providing a safe, simulated environment for repeated failure and rapid improvement, we turn interview anxiety into reliable confidence.
Research
The Challenge
The Challenge
Context
Context
Professionals at top-tier companies need to ace behavioral interviews.
Professionals at top-tier companies need to ace behavioral interviews.
Professionals at top-tier companies need to ace behavioral interviews.
Problem
Problem
Candidates often know the theory but fail in delivery due to lack of stress-resilient practice. They can't bridge the gap between "knowing" and "speaking."
Candidates often know the theory but fail in delivery due to lack of stress-resilient practice. They can't bridge the gap between "knowing" and "speaking."
Candidates often know the theory but fail in delivery due to lack of stress-resilient practice. They can't bridge the gap between "knowing" and "speaking."
Current Landscape
Users might know how to deliver speech in theory but fail in real life interviews due to lack of practice
Passive learnings
Creates a knowledge illusion but builds no muscle memory.
Voice memos
High friction, zero feedback. Users reinforce bad habits
AI Tools
High utility but high anxiety. Overwhelms users with data, lowering motivation.
Human Coach
Effective but expensive and unscalable.
Core insight
Core insight
To build a sustainable habit, I discovered that the product must satisfy two conflicting needs simultaneously:
To build a sustainable habit, I discovered that the product must satisfy two conflicting needs simultaneously:
Actual Skill Growth
1. Actual Skill Growth
UTILITY
The tool must provide high-quality, actionable feedback to technically improve speech (better structure, less filler words).
The tool must provide high-quality, actionable feedback to technically improve speech (better structure, less filler words).
Perceived Progress
2. Perceived Progress
EMOTION
The user must feel they are winning. High-friction learning requires a "dopamine loop" (immediate validation) to keep motivation alive.
The user must feel they are winning. High-friction learning requires a "dopamine loop" (immediate validation) to keep motivation alive.
#1 Actual Skill Growth Utility
The user must technically improve (better structure, less filler words).
#2 Perceived Progress Emotion
The user must feel they are getting better to stay motivated (dopamine loop).
Competitive Analysis
Competitive Analysis
Key takeaways
Guidance over Data
1.Guidance over Data
Users don't want a list of 20 improvement suggestions, they want a clear path forward. The UI must filter the "noise" and highlight the highest ROI actions to prevent analysis paralysis.
Users don't want a list of 20 improvement suggestions, they want a clear path forward. The UI must filter the "noise" and highlight the highest ROI actions to prevent analysis paralysis.
Emotional Safety
2.Emotional Safety
Feedback is scary. To encourage practice, the interface must shift from a auditor to a supportive partner, prioritising perceived progress over raw critique
Feedback is scary. To encourage practice, the interface must shift from a auditor to a supportive partner, prioritising perceived progress over raw critique
Usability Testing
MVP VALIdation
The initial MVP was launched with a technically flawless flow.
The initial MVP was launched with a technically flawless flow.
Dasрboard
Dasрboard
Dasрboard
Recording
Recording
Recording
Analysis
Analysis
Analysis
However, my usability testing revealed a critical gap between function and adoption. Users completed the tasks but refused to return.
However, my usability testing revealed a critical gap between function and adoption. Users completed the tasks but refused to return.
Data signal
I asked users to record a story, review the AI feedback, and improve their score.
I asked users to record a story, review the AI feedback, and improve their score.
User task
User task
User task
Record
Record
Analyse
Analyse
Re-record
Re-record
Result
Result
Result
85% drop-off
85% drop-off
ROOT CAUSE ANALYSIS
Recording
Recording
Analysis
Analysis
Goal
Goal
Record a STAR answer
Record a STAR answer
Record a STAR answer
Understand performance
Understand performance
Understand performance
Thought
Thought
What do I say? Will the AI even understand my context?
What do I say? Will the AI even understand my context?
What do I say? Will the AI even understand my context?
I feel stupid talking to a blank screen. What was the question again?
I feel stupid talking to a blank screen. What was the question again?
I feel stupid talking to a blank screen. What was the question again?
Whoa, too much info. I have an AI-revised script, tone advice, and 20 grammar fixes. What do I do first? Do I memorise the new text or just fix the old one?"
Whoa, too much info. I have an AI-revised script, tone advice, and 20 grammar fixes. What do I do first? Do I memorise the new text or just fix the old one?"
Whoa, too much info. I have an AI-revised script, tone advice, and 20 grammar fixes. What do I do first? Do I memorise the new text or just fix the old one?"
Emotion
Emotion
🫨
🫨
🫨
Anxiety
Anxiety
Anxiety
😲
😲
😲
Overwhelm
Overwhelm
Overwhelm
MVP Fiction
MVP Fiction
The minimalist interface removed distractions but also stripped away essential visual anchors.
Users felt lost without clear instructions and awkward talking to a void
The minimalist interface removed distractions but also stripped away essential visual anchors.
Users felt lost without clear instructions and awkward talking to a void
The minimalist interface removed distractions but also stripped away essential visual anchors.
Users felt lost without clear instructions and awkward talking to a void
The system provided detailed fixes and revised versions, but lacked hierarchy.
Users suffered from сognitive overload, seeing "everything everywhere all at once" demotivated them from starting anywhere
The system provided detailed fixes and revised versions, but lacked hierarchy.
Users suffered from сognitive overload, seeing "everything everywhere all at once" demotivated them from starting anywhere
The system provided detailed fixes and revised versions, but lacked hierarchy.
Users suffered from сognitive overload, seeing "everything everywhere all at once" demotivated them from starting anywhere
I conducted a qualitative analysis that identified two psychological barriers blocking the habit loop.
I conducted a qualitative analysis that identified two psychological barriers blocking the habit loop.
Strategy
Recording Strategy
1.Recording Strategy
To shift the focus to a video-call simulation to reduce performance anxiety.
To shift the focus to a video-call simulation to reduce performance anxiety.
Analysis Strategy
2.Analysis Strategy
To replace comprehensive reporting with a single-focus task to cure analysis paralysis.
To replace comprehensive reporting with a single-focus task to cure analysis paralysis.
Recording screen
the Problem
In the first iteration, I removed all distractions. However, testing revealed three critical issues:
Trast Gap
Users doubted the AI would understand their context without manual input
Lack of direction
Users didn't know what to say or where to start.
Low Intensity
Speaking to a blank screen felt too casual. Users lost focus, rambled, and didn't take the practice seriously.
To fix this, I shifted from a passive recorder to a video-call simulation
To fix this, I shifted from a passive recorder to a video-call simulation
redesign
I redesigned the screen to mimic a real interview environment. This forces the user to "posture up" and manage their presence, while giving them full control over the topic.
I redesigned the screen to mimic a real interview environment. This forces the user to "posture up" and manage their presence, while giving them full control over the topic.
Context and control
1. Context and control
Users can select a preset question or write their own.
Users can select a preset question or write their own.
This builds trust (ensuring the system "knows" the topic) and solves the "blank page" paralysis by giving a clear starting point.
This builds trust (ensuring the system "knows" the topic) and solves the "blank page" paralysis by giving a clear starting point.
Users don't want a list of 20 improvement suggestions, they want a clear path forward. The UI must filter the "noise" and highlight the highest ROI actions to prevent analysis paralysis.
Users don't want a list of 20 improvement suggestions, they want a clear path forward. The UI must filter the "noise" and highlight the highest ROI actions to prevent analysis paralysis.
The "Mirror Effect" & Presence
2. The "Mirror Effect" & Presence
Seeing themselves on camera forces users to fix their posture and treat the session seriously, preparing them for real interviews.
Seeing themselves on camera forces users to fix their posture and treat the session seriously, preparing them for real interviews.
The AI Avatar (listener) adds a sense of being heard, preventing the feeling of talking to a void.
The AI Avatar (listener) adds a sense of being heard, preventing the feeling of talking to a void.
Visual Anchor
3. Visual Anchor
The question card remains visible throughout the session.
The question card remains visible throughout the session.
This keeps the user focused on the specific topic and prevents rambling or forgetting the prompt mid-speech.
This keeps the user focused on the specific topic and prevents rambling or forgetting the prompt mid-speech.
Result
Session Completion Rate doubled from 42% to 88%. By explicitly setting the context (Multi-selector), we eliminated the "Trust Gap"—users felt the system understood them.
Session Completion Rate doubled from 42% to 88%. By explicitly setting the context (Multi-selector), we eliminated the "Trust Gap"—users felt the system understood them.
Simultaneously, the "Mirror Effect" and Avatar transformed the experience from an awkward monologue into a manageable conversation. Users stopped stalling and stayed focused until the end.
Simultaneously, the "Mirror Effect" and Avatar transformed the experience from an awkward monologue into a manageable conversation. Users stopped stalling and stayed focused until the end.
Analysis screen
the problem
Initially, I designed an Overview tab to summarize the top 3 focus areas alongside a full transcription. I assumed users wanted a holistic view. However, testing revealed this triggered Analysis Paralysis:
Initially, I designed an Overview tab to summarise the top 3 focus areas alongside a full transcription. I assumed users wanted a holistic view. However, testing revealed this triggered Analysis Paralysis:
Cognitive Load
Even selecting between 3-4 suggestions was overwhelming. Users had to click, read, and decide what to tackle first.
Context Switching
Replacing the video feed with text (transcription) broke the immersion. Users switched from Speaker to Reader mode and lost the drive to re-record.
Demotivation
The screen felt like a static list of flaws rather than a path forward.
To fix this, I shifted from reporting to guided, single step coaching.
To fix this, I shifted from reporting to guided, single step coaching.
redesign
I completely restructured the screen to direct all attention to re-recording. The interface now hides the details and highlights a single, high-impact task, gamifying the improvement process.
I completely restructured the screen to direct all attention to re-recording. The interface now hides the details and highlights a single, high-impact task, gamifying the improvement process.
Gamified Forecast
Instead of just showing past mistakes, the new graph projects the future. It promises: "Fix this one thing, and your score grows by 12%." This creates an immediate dopamine reward hook.
Instead of just showing past mistakes, the new graph projects the future. It promises: "Fix this one thing, and your score grows by 12%." This creates an immediate dopamine reward hook.
One Key Task
To cure overwhelm, I removed the list. The system auto-selects the Highest-ROI Improvement and expands it by default. Users don't have to choose or search, they just read the tip and hit record.
To cure overwhelm, I removed the list. The system auto-selects the Highest-ROI Improvement and expands it by default. Users don't have to choose or search, they just read the tip and hit record.
Persistent Environment
I replaced the static transcription with the paused video feed. This signals that the session isn't over, the "interviewer" is simply waiting. It keeps the user in the flow of speaking, making the transition to re-recording seamless.
I replaced the static transcription with the paused video feed. This signals that the session isn't over, the "interviewer" is simply waiting. It keeps the user in the flow of speaking, making the transition to re-recording seamless.
Result
Re-recording Rate soared from 15% to 65%.
Re-recording Rate soared from 15% to 65%.
By removing the "choice friction" and introducing the "Next Step" Prediction, we transformed the analysis screen from a dead end into a launchpad. Users stopped analyzing their failures and started practicing their improvements.
By removing the "choice friction" and introducing the "Next Step" Prediction, we transformed the analysis screen from a dead end into a launchpad. Users stopped analyzing their failures and started practicing their improvements.
Conclusion
BUSINESS OUTCOMES
The redesign successfully triggered the "Immediate Iteration" behavior defined in my initial hypothesis.
The redesign successfully triggered the "Immediate Iteration" behavior defined in my initial hypothesis.
Completion Rate: 42% → 88%
Completion Rate: 42% → 88%
Creating a "Simulated Video Call" removed the anxiety of the void.
Creating a "Simulated Video Call" removed the anxiety of the void.
Re-recording Rate: 15% → 65%
Re-recording Rate: 15% → 65%
Replacing the wall of suggestions with a single focus task turned passive analysis into active practice.
Replacing the wall of suggestions with a single focus task turned passive analysis into active practice.
KEY TAKEAWAYS
Guidance > Raw Intelligence
Guidance > Raw Intelligence
My research proved that showing users everything the AI can do leads to paralysis. Curating the feedback into a single focus area was more effective than displaying a comprehensive analysis. A good UX filters the noise to highlight the next step.
My research proved that showing users everything the AI can do leads to paralysis. Curating the feedback into a single focus area was more effective than displaying a comprehensive analysis. A good UX filters the noise to highlight the next step.
Perceived Progress Drives Habit
Perceived Progress Drives Habit
It wasn't enough for the feedback to be accurate; users needed to feel they were winning. Introducing the "Future Score" graph (+12%) shifted the mindset from "fixing errors" to "chasing growth," creating the dopamine loop necessary for retention.
It wasn't enough for the feedback to be accurate; users needed to feel they were winning. Introducing the "Future Score" graph (+12%) shifted the mindset from "fixing errors" to "chasing growth," creating the dopamine loop necessary for retention.
design.dubova@gmail.com
+44-7471-541-313
ED