Redesigning how people organize insight

Collabwriting is a knowledge management tool helping teams capture, structure, and rediscover insights.

Role

UX/UI design lead

Timeline

February 2025 -

September 2025

Tools used

Results

30%

increased user

user activation rate

40%

increase in the

user retention rate

20%

increased developer

shipping time

100%

increased speed

in user research flow

Context

Collabwriting was growing, but users weren’t activating or returning. The issue wasn’t a lack of features. It was fragmentation: people saved insights, but couldn’t organize, use, or retrieve them efficiently.

The big win

I transformed Collabwriting from a “note dumping tool” into a structured knowledge management system users trust daily.

Trusted by marketing teams, researchers, and legal professionals at established companies.

Problem

Collabwriting didn’t help people bridge Capture → Structure → Retrieval.

Collabwriting didn’t help people bridge Capture → Structure → Retrieval

Signals from data + feedback

Low activation (13%) showed early friction & unclear value

Retention dips tied to inability to “find what I saved”

Support logs full of “Where did this go?”, “I can’t find my topics”

Teams using PDFs, bookmarks, e-mails, posts, needed structure

18%

18%

22%

22%

16%

16%

14%

14%

18%

18%

17%

17%

25%

25%

21%

21%

16%

16%

16%

16%

17%

17%

10%

10%

User activation rates from February of 2025 to May of 2025

User Feedback

Via the Sleekplan app, I had access to suggestions from users about bugs, and features.



Other community members could vote and leave comments. This way, I would know which features to prioritize.

The most important feedback came from team (enterprise) plan users. I worked with them on qualitative research: 1 on 1 calls and A/B testing. Those sessions helped me create 3 common user personas and their pain-points

The most important feedback came from team (enterprise) plan users. I worked with them on qualitative research: 1 on 1 calls and A/B testing. Those sessions helped me create 3 common user personas and their pain-points

Real user feedback from collabwriting.sleekplan.app

Real user feedback from collabwriting.sleekplan.app

Personas

& Pain Points

Together, these personas reveal the system-level problem: users struggled to transform collected content into actionable knowledge.

Ana

Marketing specialist

Short bio

Collects insights from YouTube, blogs, LinkedIn. A million open tabs, messy research, no structure.

Pain-point

I don’t know where things go. I want one place where my research is actually usable.

Damian

Lawyer

Short bio

Reads long, complex laws and amendments which he collects as snippets inside of his dashboard.

Pain-point

I lose track of what I highlighted. My knowledge becomes scattered across PDFs, notes, and tabs.

Paul

Student

Short bio

Pulls insights from papers, lectures, forums. Lot's of knowledge nuggets but forgets where he saves them.

Pain-point

I never know where I saw something. Half of my research time is hunting for the other half.

Product strategy


Product strategy

To solve the chaos, I created a framework guiding all decisions:

The Retrieval Framework

Capture

Make it effortless to bring content into Collabwriting

(shortcut, quick-save, inbox)

Structure

Give users intuitive, visual ways to organize information

(PDF hub, tag buckets, icons, clusters, topic UI)

Retrieve

Make insights instantly findable and re-usable

(Inbox, dashboard redesign, AI chat, improved navigation)

Every feature feeds into this loop. This turns your “many features” into one strategic story.

UX Research

By mapping user behavior across marketers, students, and legal professionals, clear patterns emerged around organization and retrieval pain points. These findings shaped the core product direction and guided every major design decision that followed.

Insight 1

Users expected visual cues where they lacked. Especially Ana and students accustomed to Notion-like mental models.

Insight 2

Complex PDFs needed a dedicated workflow. Legal and academic users demanded better reading, highlighting, and referencing.

Insight 3

Inconsistent UI added friction. The research process needs to be as seamless as possible so the users focus solely on the research.

The CJM helped me map out the exact points where customers lost retention. Which helped me know exactly which features I should prioritize.

Before the overhaul

Patterns emerged quickly. Researchers and marketers asked for a faster, less cluttered workspace, while legal professionals and students consistently depended on their PDFs.

I rebuilt the interface around clarity, hierarchy, and predictable patterns users could trust. The new design system aligned the entire team, improving both usability and development speed.

Issues to solve

  • And old looking dashboard with a lacking navigation

  • A buggy PDF reader that was not existent on mobile

  • No research customization features for better organizing

A dark and heavy top bar with no navigation, rigid angles and cluttered snippets with too much information.

Design System

I rebuilt the design system around clarity, hierarchy, and predictable patterns users could trust. The new design system aligned the entire team, improving both usability and development speed.

I rebuilt the interface around clarity, hierarchy, and predictable patterns users could trust. The new design system aligned the entire team, improving both usability and development speed.

Goals

Unify visual language

Improve speed of execution for engineers

Reduce cognitive load for users

Establish scalable patterns for upcoming features

Impact

20% faster feature development

Reduced UX inconsistencies → more trust + clarity

Enabled rapid iteration for AI Chat, Inbox, and Tag Buckets

Key feature 1

Tag buckets

(Litter Box)

Charles Darwin always had the theory of evolution inside his notes. Though he didn't have a way to structure them together through tag buckets. Luckily we do now.

Problem

Users needed a simple way to structure scattered research without building complex taxonomies.

Why it matters

People don’t want to “manage tags”. They want meaning groups.

Solution

A visual grouping system allowing users to group snippets by tags, inside of a "non-existing" topic which they can share.

Outcome

Higher organization success rate

Increased activation among marketing + student personas

Key feature 2

Inbox

The feeling when you hear a song that is too good, but you still didn't find the right playlist on YouTube for it, so you save it for later? The Inbox is just that, but for research on the go.

Problem

Users wanted a faster way of collecting snippets. Sometimes when you're in the zone, you don't have time to search for the right "playlist".

Solution

A separate "Inbox" cluster that would be already active when you use the new highlight saving shortcut ALT+SHIFT+S.

Why it works

Users return when they know the product remembers what they need.

Outcome

The highest contributing feature for the user activation rate.

Key feature 3

PDF Reader revamp

Students use kindle, lawyers use… law .pdfs.

They together consist about 40% of the active user base. With a non-functional PDF feature, you lose 40% of your userbase.

Students use kindle, lawyers use… law .pdfs. They together consist about 40% of the active user base. With a non-functional PDF feature, you lose 40% of your userbase.

Problem

The users could not view their PDFs on the phone. While on desktop, they had no search bar, zoom options, nor different highlight colors.

Solution

A custom PDF reading + highlighting tool built directly inside Collabwriting, both on desktop and mobile.

Benefits

During the research process, a lot of external tools costed a fortune. Having a native, straight-forward PDF feature was the solution.

Outcome

Major adoption in legal and research-heavy personas

Key feature 4

Dashboard redesign

According by many users through the intercom platform, the dashboard felt rigid. Too much information clutter in the left side navigation, losing track of which cluster they are in. The design was not up to par with modern apps.

What Changed

The dark navigation at the top was changed for a seamless breadcrumbs navigation, which allows users to expand the dashboard for a research focused workflow. A much needed home button was added, as well as a clear navigation for users to know in which cluster they currently are.

Why it matters

This is the core unit of Collabwriting.
Improving clarity = improving the whole product.

Alongside the new overhauled dashboard, the snippets gave

Supporting features

Alongside the key features, the supporting features gave more life and aesthetic appeal to the whole product, differentiating it from a vast sea of lifeless and minimalistic products.

Our office CHO Jerry became the face of the app and people loved it!

Features list

  • Avatar uploading

  • Icons for Topics & Clusters

  • Empty state & loading animations

Avatar uploading workflow

The icons workflow

Empty states. Most of these are Lottie animations, but If I were to put them all as gifs, you would get a headache…

Outcomes

Business Metrics

  • Activation: 13% → 33% in 6 months

  • Retention: 40% in ~18 months

  • Dev speed: +20% faster due to design system

  • Feature adoption: PDF reader, Tag Buckets, Inbox among highest-utilized new features

User Outcomes

  • Anas find and organize insights 2x faster

  • Damians process legal PDFs without switching tools

  • Pauls reduce research chaos and retain context

Company Outcomes

  • More coherent product story

  • Scalable foundation for new AI and collaboration features

21%

21%

18%

18%

24%

24%

23%

23%

27%

27%

28%

28%

27%

27%

30%

30%

23%

23%

27%

27%

29%

29%

33%

33%

User activation rates from May of 2025 to August of 2025

What I'd do next

Focus on product evolution, not UI tweaks.

  • AI-assisted auto-tagging and auto-clustering

  • Topic-level summarization

  • Organization templates

  • Knowledge graphs

  • Gamification

  • Smart reminders for forgotten snippets

  • Offline mode for researchers on the go

Closing words

Collabwriting grew from a fragmented tool into a structured knowledge system. My work helped transform how users save, organize, and rediscover insights. This wasn’t just UI improvement. It was a product shift: from “save things” to “understand things”.

Redesigning how people organize insight

Collabwriting is a knowledge management tool helping teams capture, structure, and rediscover insights.

Role

UX/UI design lead

Timeline

February 2025 -

September 2025

Tools used

Redesigning how people organize insight

Collabwriting is a knowledge management tool helping teams capture, structure, and rediscover insights.

Role

UX/UI design lead

Timeline

February 2025 -

September 2025

Tools used

Results

30%

increased user

user activation rate

20%

increased developer

shipping time

40%

increase in the

user retention rate

100%

increased speed

in user research flow

Context

Collabwriting was growing, but users weren’t activating or returning. The issue wasn’t a lack of features. It was fragmentation: people saved insights, but couldn’t organize, use, or retrieve them efficiently.

The big win

I transformed Collabwriting from a “note dumping tool” into a structured knowledge management system users trust daily.

Problem

Collabwriting didn’t help people bridge Capture → Structure → Retrieval.

Signals from data + feedback

Low activation (13%) showed early friction & unclear value

Retention dips tied to inability to “find what I saved”

Support logs full of “Where did this go?”, “I can’t find my topics”

Teams using PDFs, bookmarks, e-mails, posts, needed structure

18%

22%

16%

14%

18%

17%

25%

21%

16%

16%

17%

10%

User activation rates from February of 2025 to May of 2025

Personas

& Pain Points

Ana

Marketing specialist

Short bio

Collects insights from YouTube, blogs, LinkedIn. A million open tabs, messy research, no structure.

Pain-point

I don’t know where things go. I want one place where my research is actually usable.

Together, these personas reveal the system-level problem: users struggled to transform collected content into actionable knowledge.

Damian

Lawyer

Short bio

Reads long, complex laws and amendments which he collects as snippets inside of his dashboard.

Pain-point

I lose track of what I highlighted. My knowledge becomes scattered across PDFs, notes, and tabs.

Paul

Student

Short bio

Pulls insights from papers, lectures, forums. Lot's of knowledge nuggets but forgets where he saves them.

Pain-point

I never know where I saw something. Half of my research time is hunting for the other half.

Product strategy


To solve the chaos, I created a framework guiding all decisions:

The Retrieval Framework

Capture

Make it effortless to bring content into Collabwriting

(shortcut, quick-save, inbox)

Structure

Give users intuitive, visual ways to organize information

(PDF hub, tag buckets, icons, clusters, topic UI)

Retrieve

Make insights instantly findable and re-usable

(Inbox, snippet redesign, AI chat, improved navigation)

Every feature feeds into this loop. This turns your “many features” into one strategic story.

Key feature 2

Inbox

The feeling when you hear a song that is too good, but you still didn't find the right playlist on YouTube for it, so you save it for later? The Inbox is just that, but for research on the go.

Problem

Users wanted a faster way of collecting snippets. Sometimes when you're in the zone, you don't have time to search for the right "playlist".

Solution

A separate "Inbox" cluster that would be already active when you use the new highlight saving shortcut ALT+SHIFT+S.

Why it works

Users return when they know the product remembers what they need.

Outcome

The highest contributing feature for the user activation rate.

Key feature 3

PDF Reader revamp

Students use kindle, lawyers use… law .pdfs.

They together consist about 40% of the active user base. With a non-functional PDF feature, you lose 40% of your userbase.

Problem

The users could not view their PDFs on the phone. While on desktop, they had no search bar, zoom options, nor different highlight colors.

Solution

A custom PDF reading + highlighting tool built directly inside Collabwriting, both on desktop and mobile.

Benefits

During the research process, a lot of external tools costed a fortune. Having a native, straight-forward PDF feature was the solution.

Outcome

Major adoption in legal and research-heavy personas

What I'd do next

Focus on product evolution, not UI tweaks.

  • AI-assisted auto-tagging and auto-clustering

  • Topic-level summarization

  • Organization templates

  • Knowledge graphs

  • Gamification

  • Smart reminders for forgotten snippets

  • Offline mode for researchers on the go

Closing words

Collabwriting grew from a fragmented tool into a structured knowledge system. My work helped transform how users save, organize, and rediscover insights. This wasn’t just UI improvement. It was a product shift: from “save things” to “understand things”.

User

Feedback

Via the Sleekplan app, I had access to suggestions from users about bugs, and features.



Other community members could vote and leave comments. This way, I would know which features to prioritize.

The most important feedback came from team (enterprise) plan users. I worked with them on qualitative research: 1 on 1 calls and A/B testing. Those sessions helped me create 3 common user personas and their pain-points

Real user feedback from collabwriting.sleekplan.app

Before the overhaul

I rebuilt the interface around clarity, hierarchy, and predictable patterns users could trust. The new design system aligned the entire team, improving both usability and development speed.

Issues to solve

  • And old looking dashboard with a lacking navigation

  • A buggy PDF reader that was not existent on mobile

  • No research customization features for better organizing

A dark and heavy top bar with no navigation, rigid angles and cluttered snippets with too much information.

Key feature 1

Tag buckets

(Litter Box)

Charles Darwin always had the theory of evolution inside his notes. Though he didn't have a way to structure them together through tag buckets. Luckily we do now.

Problem

Users needed a simple way to structure scattered research without building complex taxonomies.

Why it matters

People don’t want to “manage tags”. They want meaning groups.

Solution

A visual grouping system allowing users to group snippets by tags, inside of a "non-existing" topic which they can share.

Outcome

Higher organization success rate

Increased activation among marketing + student personas

Key feature 4

Dashboard redesign

According by many users through the intercom platform, the dashboard felt rigid. Too much information clutter in the left side navigation, losing track of which cluster they are in. The design was not up to par with modern apps.

What Changed

The dark navigation at the top was changed for a seamless breadcrumbs navigation, which allows users to expand the dashboard for a research focused workflow. A much needed home button was added, as well as a clear navigation for users to know in which cluster they currently are.

Why it matters

This is the core unit of Collabwriting.
Improving clarity = improving the whole product.

Alongside the new overhauled dashboard, the snippets gave

The CJM helped me map out the exact points where customers lost retention. Which helped me know exactly which features I should prioritize.

UX Research

By mapping user behavior across marketers, students, and legal professionals, clear patterns emerged around organization and retrieval pain points. These findings shaped the core product direction and guided every major design decision that followed.

Insight 1

Users expected visual cues where they lacked. Especially Ana and students accustomed to Notion-like mental models.

Insight 2

Complex PDFs needed a dedicated workflow. Legal and academic users demanded better reading, highlighting, and referencing.

Insight 3

Inconsistent UI added friction. The research process needs to be as seamless as possible so the users focus solely on the research.

Design

System

I rebuilt the interface around clarity, hierarchy, and predictable patterns users could trust. The new design system aligned the entire team, improving both usability and development speed.

Goals

Unify visual language

Improve speed of execution for engineers

Reduce cognitive load for users

Establish scalable patterns for upcoming features

Impact

20% faster feature development

Reduced UX inconsistencies → more trust + clarity

Enabled rapid iteration for AI Chat, Inbox, and Tag Buckets

Supporting features

Alongside the key features, the supporting features gave more life and aesthetic appeal to the whole product, differentiating it from a vast sea of lifeless and minimalistic products.

Our office CHO Jerry became the face of the app and people loved it!

Features list

  • Avatar uploading

  • Icons for Topics & Clusters

  • Empty state & loading animations

Avatar uploading workflow

The icons workflow

Empty states. Most of these are Lottie animations, but If I were to put them all as gifs, you would get a headache…

Outcomes

Business Metrics

  • Activation: 13% → 33% in 6 months

  • Retention: 40% in ~18 months

  • Dev speed: +20% faster due to design system

  • Feature adoption: PDF reader, Tag Buckets, Inbox among highest-utilized new features

User Outcomes

  • Anas find and organize insights 2x faster

  • Damians process legal PDFs without switching tools

  • Pauls reduce research chaos and retain context

Company Outcomes

  • More coherent product story

  • Scalable foundation for new AI and collaboration features

21%

18%

24%

23%

27%

28%

27%

30%

23%

27%

29%

33%

User activation rates from May of 2025 to August of 2025