How to redesign a 100M+ MAU service, to unlock +7.38% in revenue
For over 20 years, the service had been over-optimized for revenue. Any minor pixel shift "bled" money metrics. But working autonomously alongside the Lead PM, We’ve rolled out a modern UI that not only survived the A/B tests but significantly grew the bottom line.
What I’ve Done
100% "Turnkey" delivery. I handled the entire design lifecycle—from the first concept to production. A lead designer wasn't engaged in management.
Learn More
Result
+7.38%* increase in desktop revenue.
Measurable growth across all key user satisfaction metrics.
*Due to the multi-billion ruble baseline of the service, even a fractional percentage increase represents millions in absolute revenue growth.
"I just found devs for the redesign. We got 3 days to get approved mocks ready for dev."
—Lead Product Manager, Yandex Images
Business Goal: money, money, fucking money
A redesign is never the goal. I quickly aligned with the PM on the real objective: make more money. It’s always about money. Yes, usability. Yes, user happiness. Yes, “such a nice UI!”. But in the end, survive only the ideas that grow the bottom line. Period.
Playing on "Insane" Difficulty: 100M+ users, 3 days, and 20+ Stakeholders
- The Revenue Trap: The viewer was so heavily monetized that past redesign attempts failed. Any changes historically dropped revenue, making stakeholders extremely risk-averse.
- The "We Tried This" Barrier: Every early draft faced the classic corporate pushback: "We already tried this, it didn't work." It took high-level creativity and hard data to squeeze more profit out of a 20-year-old flow.
- Heavy Stakeholder Management: Search is massive. We had to navigate and align 20+ Decision Makers and Influencers using drive and bulletproof UX arguments to push scale changes through.
- Universal Edge-Cases: With 20M+ DAU searching for everything from postcards to extreme NSFW content, the UI solution had to be bulletproof across all possible scenarios.
What I’ve Done: Design from 0 to Production
- Defined the objectives, metrics, and tech bandwidth with the PM to avoid burning resources on low-ROI ideas.
- Conducted ×32+ interviews to reveal problems and opportunities.
- Delivered ×7 major iterations within 72 hours to grab a limited spot to approve solutions with C-level reps.
- Ran ×6 UX tests to kill bad ideas.
- Defended mocks with PM to get approval, push mocks to devs ASAP, and to unblock development.
- Worked out all edge-case states to avoid wasting dev’s time.
- Built the Dev Hand-off and Design Review pipeline to ensure the solutions deliver the intended value, and we are not ashamed of production.
- Designed hotfixes based on A/B test data to rollout solutions in production.
Enlarged the "Related images" section to reduce friction in the search scenario
"Related images" enlargement
During research, I noticed people used the “Related images” section heavily to find the right image. But I've seen how badly "Related images" served that need—the narrow right-hand column forced too much scrolling and made the images feel tiny and easy to miss. So I pushed for a simple fix: move them below the main content.
Now, people can check a full screen of large images with just a single scroll. And for us, it opened up a profitable surface for placing ads and driving revenue.
Improved UI accents to eliminate user overload and drive CTR
UI accents Improvement
User testing kept surfacing the same issue: the interface felt overloaded. It raised the same questions internally—why are there three different components showing images? Why is the top-right so noisy? I stripped the noise, reduced competing elements, and CTAs.
"Visual accents are a powerful tool to drive users’ attention towards product’s value. Poor hierarchy is among the top reasons why products lose money."
Removed a low ROI feature so the business doesn't lose money to maintain it
Dead Feature Removal
Another problem I’ve discovered on UX's is that people didn’t understand that the gallery on the left navigates by images from the previous step. So the feature took the precious real estate above the fold just to distract. Moreover, devs wasted time ensure it works properly during each change.
We removed the gallery. Now, when users want to get back to the previous images, they just move to the step before. There they see familiar UI with big images and… ads (wink!).
"Dead features destroy your product. Business throws away resources to maintain them. Alternative costs are also involved—same resources could be invested in the feature that would scale the product’s revenue."
Dragged all ads to the right column to kill friction the user flow, while scaling profit
Ads relocation
The "find image" scenario still had friction due to ads being mixed in with related images. We pushed all ads to an anchored right column. We’ve unblocked the user scenario and scaled profit, despite reducing the total ad volume.
Built the process to ensure implementation meets high-quality standards
Before me, hand-offs were a mess: devs wasted hours parsing outdated mockups. On one occasion, they developed outdated design due to versioning ambiguity.
I’ve built hand-off process
- Prominently marked the final mockups as “Ready to Dev”.
- Removed all unnecessary, outdated layouts from the file.
- Marked them carefully—all states, corner cases, anticipated what might break during development, and provided the solutions how we would fix them.
We started to get implemented flows faster and with higher quality. Moreover, we improved relationships within the team.
Hotfixed issues during the experiments to prevent metrics dip
In a 100M+ MAU ecosystem, the ultimate challenge is safe deployment. Historically, a 2013 redesign of Yandex Search wiped out tenth of revenue. To avoid repeating that catastrophe, we adopted an incremental shipping strategy. I intentionally restored some of familiar UI elements to avoid disrupting established user mental models and habits, ensuring growth without the risk of a collapse.
We successfully shipped the redesign and scaled profit.
This was made possible by a tight strategic partnership: the PM and I worked late into the night to draft hotfixes for the experiment, followed by 2 AM brainstorms for the next iterations.
Finally, my ability to draft mockups in minutes transformed months of work into days of high-speed discovery.
Get a Design Partner
Get design fast to test riskiest assumptions and avoid sales troubles
Free
No micro-managment
Build trust, cut down CPA and raise LTV
A design partner, not a freelancer
I stay until we see the results.
LTV
Premium craft UI
Build trust, cut down CPA and raise LTV
Book a 15-min Call
Check Available Slots
only ×6 slots / week
Влад Ви
I work only with the people with the highest standards.
Direct Inquiry
Medium
Product Hunt
© Vlad Vi, 2026
How to redesign a 100M+ MAU service, to unlock +7.38% in revenue
For over 20 years, the service had been over-optimized for revenue. Any minor pixel shift "bled" money metrics. But working autonomously alongside the Lead PM, We’ve rolled out a modern UI that not only survived the A/B tests but significantly grew the bottom line.
What I’ve Done
100% "Turnkey" delivery. I handled the entire design lifecycle—from the first concept to production. A lead designer wasn't engaged in management.
Learn More
Results
+7.38%* increase in desktop revenue.
Measurable growth across all key user satisfaction metrics.
*Due to the multi-billion ruble baseline of the service, even a fractional percentage increase represents millions in absolute revenue growth.
"I just found devs for the redesign. We got 3 days to get approved mocks ready for dev."
—Lead Product Manager, Yandex Images
Business Goal: money, money, fucking money
A redesign is never the goal. I quickly aligned with the PM on the real objective: make more money. It’s always about money. Yes, usability. Yes, user happiness. Yes, “such a nice UI!”. But in the end, survive only the ideas that grow the bottom line. Period.
Playing on "Insane" Difficulty: 100M+ users, 3 days, and 20+ Stakeholders
- The Revenue Trap: The viewer was so heavily monetized that past redesign attempts failed. Any changes historically dropped revenue, making stakeholders extremely risk-averse.
- The "We Tried This" Barrier: Every early draft faced the classic corporate pushback: "We already tried this, it didn't work." It took high-level creativity and hard data to squeeze more profit out of a 20-year-old flow.
- Heavy Stakeholder Management: Search is massive. We had to navigate and align 20+ Decision Makers and Influencers using drive and bulletproof UX arguments to push scale changes through.
- Universal Edge-Cases: With 20M+ DAU searching for everything from postcards to extreme NSFW content, the UI solution had to be bulletproof across all possible scenarios.
What I’ve Done: Design from 0 to Production
- Defined the objectives, metrics, and tech bandwidth with the PM to avoid burning resources on low-ROI ideas.
- Conducted ×32+ interviews to reveal problems and opportunities.
- Delivered ×7 major iterations within 72 hours to grab a limited spot to approve solutions with C-level reps.
- Ran ×6 UX tests to kill bad ideas.
- Defended mocks with PM to get approval, push mocks to devs ASAP, and to unblock development.
- Worked out all edge-case states to avoid wasting dev’s time.
- Built the Dev Hand-off and Design Review pipeline to ensure the solutions deliver the intended value, and we are not ashamed of production.
- Designed hotfixes based on A/B test data to rollout solutions in production.
Enlarged the "Related images" section to reduce friction in the search scenario
"Related images" enlargement
During research, I noticed people used the “Related images” section heavily to find the right image. But I've seen how badly "Related images" served that need—the narrow right-hand column forced too much scrolling and made the images feel tiny and easy to miss. So I pushed for a simple fix: move them below the main content.
Now, people can check a full screen of large images with just a single scroll. And for us, it opened up a profitable surface for placing ads and driving revenue.
Improved UI accents to eliminate user overload and drive CTR
UI accents Improvement
User testing kept surfacing the same issue: the interface felt overloaded. It raised the same questions internally—why are there three different components showing images? Why is the top-right so noisy? I stripped the noise, reduced competing elements, and CTAs.
"Visual accents are a powerful tool to drive users’ attention towards product’s value. Poor hierarchy is among the top reasons why products lose money."
Removed a low ROI feature so the business doesn't lose money to maintain it
Dead Feature Removal
Another problem I’ve discovered on UX's is that people didn’t understand that the gallery on the left navigates by images from the previous step. So the feature took the precious real estate above the fold just to distract. Moreover, devs wasted time ensure it works properly during each change.
We removed the gallery. Now, when users want to get back to the previous images, they just move to the step before. There they see familiar UI with big images and… ads (wink!).
"Dead features destroy your product. Business throws away resources to maintain them. Alternative costs are also involved—same resources could be invested in the feature that would scale the product’s revenue."
Dragged all ads to the right column to kill friction the user flow, while scaling profit
Ads relocation
The "find image" scenario still had friction due to ads being mixed in with related images. We pushed all ads to an anchored right column. We’ve unblocked the user scenario and scaled profit, despite reducing the total ad volume.
Built the process to ensure implementation meets high-quality standards
Before me, hand-offs were a mess: devs wasted hours parsing outdated mockups. On one occasion, they developed outdated design due to versioning ambiguity.
I’ve built hand-off process
- Prominently marked the final mockups as “Ready to Dev”.
- Removed all unnecessary, outdated layouts from the file.
- Marked them carefully—all states, corner cases, anticipated what might break during development, and provided the solutions how we would fix them.
We started to get implemented flows faster and with higher quality. Moreover, we improved relationships within the team.
Hotfixed issues during the experiments to prevent metrics dip
In a 100M+ MAU ecosystem, the ultimate challenge is safe deployment. Historically, a 2013 redesign of Yandex Search wiped out tenth of revenue. To avoid repeating that catastrophe, we adopted an incremental shipping strategy. I intentionally restored some of familiar UI elements to avoid disrupting established user mental models and habits, ensuring growth without the risk of a collapse.
We successfully shipped the redesign and scaled profit.
This was made possible by a tight strategic partnership: the PM and I worked late into the night to draft hotfixes for the experiment, followed by 2 AM brainstorms for the next iterations.
Finally, my ability to draft mockups in minutes transformed months of work into days of high-speed discovery.
Get a Design Partner
Dominate your niche with premium interfaces that earn user trust and drive revenue.
Free
No micro-managment
I take full ownership of the design process.
A design partner, not a freelancer
I stay until we see the results.
LTV
Premium craft UI
Build trust, cut down CPA and raise LTV
Book a 15-min Intro Call
Check Available Slots
only ×6 slots / week
Влад
Ви
I work only with the people with the highest standards.
Direct Inquiry
Medium
Product Hunt
© Vlad Vi, 2026
How to redesign a 100M+ MAU service, to unlock +7.38% in revenue
For over 20 years, the service had been over-optimized for revenue. Any minor pixel shift "bled" money metrics. But working autonomously alongside the Lead PM, We’ve rolled out a modern UI that not only survived the A/B tests but significantly grew the bottom line.
What I’ve Done
100% "Turnkey" delivery. I handled the entire design lifecycle—from the first concept to production. A lead designer wasn't engaged in management.
Learn More
Results
+7.38%* increase in desktop revenue.
Measurable growth across all key user satisfaction metrics.
*Due to the multi-billion ruble baseline of the service, even a fractional percentage increase represents millions in absolute revenue growth.
"I just found devs for the redesign. We got 3 days to get approved mocks ready for dev."
—Lead Product Manager, Yandex Images
Business Goal: money, money, fucking money
A redesign is never the goal. I quickly aligned with the PM on the real objective: make more money. It’s always about money. Yes, usability. Yes, user happiness. Yes, “such a nice UI!”. But in the end, survive only the ideas that grow the bottom line. Period.
Playing on "Insane" Difficulty: 100M+ users, 3 days, and 20+ Stakeholders
- The Revenue Trap: The viewer was so heavily monetized that past redesign attempts failed. Any changes historically dropped revenue, making stakeholders extremely risk-averse.
- The "We Tried This" Barrier: Every early draft faced the classic corporate pushback: "We already tried this, it didn't work." It took high-level creativity and hard data to squeeze more profit out of a 20-year-old flow.
- Heavy Stakeholder Management: Search is massive. We had to navigate and align 20+ Decision Makers and Influencers using drive and bulletproof UX arguments to push scale changes through.
- Universal Edge-Cases: With 20M+ DAU searching for everything from postcards to extreme NSFW content, the UI solution had to be bulletproof across all possible scenarios.
What I’ve Done: Design from 0 to Production
- Defined the objectives, metrics, and tech bandwidth with the PM to avoid burning resources on low-ROI ideas.
- Conducted ×32+ interviews to reveal problems and opportunities.
- Delivered ×7 major iterations within 72 hours to grab a limited spot to approve solutions with C-level reps.
- Ran ×6 UX tests to kill bad ideas.
- Defended mocks with PM to get approval, push mocks to devs ASAP, and to unblock development.
- Worked out all edge-case states to avoid wasting dev’s time.
- Built the Dev Hand-off and Design Review pipeline to ensure the solutions deliver the intended value, and we are not ashamed of production.
- Designed hotfixes based on A/B test data to rollout solutions in production.
Enlarged the "Related images" section to reduce friction in the search scenario
"Related images" enlargement
During research, I noticed people used the “Related images” section heavily to find the right image. But I've seen how badly "Related images" served that need—the narrow right-hand column forced too much scrolling and made the images feel tiny and easy to miss. So I pushed for a simple fix: move them below the main content.
Now, people can check a full screen of large images with just a single scroll. And for us, it opened up a profitable surface for placing ads and driving revenue.
Improved UI accents to eliminate user overload and drive CTR
UI accents Improvement
User testing kept surfacing the same issue: the interface felt overloaded. It raised the same questions internally—why are there three different components showing images? Why is the top-right so noisy? I stripped the noise, reduced competing elements, and CTAs.
"Visual accents are a powerful tool to drive users’ attention towards product’s value. Poor hierarchy is among the top reasons why products lose money."
Removed a low ROI feature so the business doesn't lose money to maintain it
Dead Feature Removal
Another problem I’ve discovered on UX's is that people didn’t understand that the gallery on the left navigates by images from the previous step. So the feature took the precious real estate above the fold just to distract. Moreover, devs wasted time ensure it works properly during each change.
We removed the gallery. Now, when users want to get back to the previous images, they just move to the step before. There they see familiar UI with big images and… ads (wink!).
"Dead features destroy your product. Business throws away resources to maintain them. Alternative costs are also involved—same resources could be invested in the feature that would scale the product’s revenue."
Dragged all ads to the right column to kill friction the user flow, while scaling profit
Ads relocation
The "find image" scenario still had friction due to ads being mixed in with related images. We pushed all ads to an anchored right column. We’ve unblocked the user scenario and scaled profit, despite reducing the total ad volume.
Built the process to ensure implementation meets high-quality standards
Before me, hand-offs were a mess: devs wasted hours parsing outdated mockups. On one occasion, they developed outdated design due to versioning ambiguity.
I’ve built hand-off process
- Prominently marked the final mockups as “Ready to Dev”.
- Removed all unnecessary, outdated layouts from the file.
- Marked them carefully—all states, corner cases, anticipated what might break during development, and provided the solutions how we would fix them.
We started to get implemented flows faster and with higher quality. Moreover, we improved relationships within the team.
Hotfixed issues during the experiments to prevent metrics dip
In a 100M+ MAU ecosystem, the ultimate challenge is safe deployment. Historically, a 2013 redesign of Yandex Search wiped out tenth of revenue. To avoid repeating that catastrophe, we adopted an incremental shipping strategy. I intentionally restored some of familiar UI elements to avoid disrupting established user mental models and habits, ensuring growth without the risk of a collapse.
We successfully shipped the redesign and scaled profit.
This was made possible by a tight strategic partnership: the PM and I worked late into the night to draft hotfixes for the experiment, followed by 2 AM brainstorms for the next iterations.
Finally, my ability to draft mockups in minutes transformed months of work into days of high-speed discovery.
Get a Design Partner
Dominate your niche with premium interfaces that earn user trust and drive revenue.
Free
No micro-managment
I take full ownership of the design process.
A design partner, not a freelancer
I stay until we see the results.
LTV
Premium craft UI
Build trust, cut down CPA and raise LTV
Book a 15-min Intro Call
Check Available Slots
only ×6 slots / week
Влад
Ви
I work only with the people with the highest standards.
Direct Inquiry
Medium
Product Hunt
© Vlad Vi, 2026
How to redesign a 100M+ MAU service, to unlock +7.38% in revenue
For over 20 years, the service had been over-optimized for revenue. Any minor pixel shift "bled" money metrics. But working autonomously alongside the Lead PM, We’ve rolled out a modern UI that not only survived the A/B tests but significantly grew the bottom line.
What I’ve Done
100% "Turnkey" delivery. I handled the entire design lifecycle—from the first concept to production. A lead designer wasn't engaged in management.
Learn More
Results
+7.38%* increase in desktop revenue.
Measurable growth across all key user satisfaction metrics.
*Due to the multi-billion ruble baseline of the service, even a fractional percentage increase represents millions in absolute revenue growth.
"I just found devs for the redesign. We got 3 days to get approved mocks ready for dev."
—Lead Product Manager, Yandex Images
Business Goal: money, money, fucking money
A redesign is never the goal. I quickly aligned with the PM on the real objective: make more money. It’s always about money. Yes, usability. Yes, user happiness. Yes, “such a nice UI!”. But in the end, survive only the ideas that grow the bottom line. Period.
Playing on "Insane" Difficulty: 100M+ users, 3 days, and 20+ Stakeholders
- The Revenue Trap: The viewer was so heavily monetized that past redesign attempts failed. Any changes historically dropped revenue, making stakeholders extremely risk-averse.
- The "We Tried This" Barrier: Every early draft faced the classic corporate pushback: "We already tried this, it didn't work." It took high-level creativity and hard data to squeeze more profit out of a 20-year-old flow.
- Heavy Stakeholder Management: Search is massive. We had to navigate and align 20+ Decision Makers and Influencers using drive and bulletproof UX arguments to push scale changes through.
- Universal Edge-Cases: With 20M+ DAU searching for everything from postcards to extreme NSFW content, the UI solution had to be bulletproof across all possible scenarios.
What I’ve Done: Design from 0 to Production
- Defined the objectives, metrics, and tech bandwidth with the PM to avoid burning resources on low-ROI ideas.
- Conducted ×32+ interviews to reveal problems and opportunities.
- Delivered ×7 major iterations within 72 hours to grab a limited spot to approve solutions with C-level reps.
- Ran ×6 UX tests to kill bad ideas.
- Defended mocks with PM to get approval, push mocks to devs ASAP, and to unblock development.
- Worked out all edge-case states to avoid wasting dev’s time.
- Built the Dev Hand-off and Design Review pipeline to ensure the solutions deliver the intended value, and we are not ashamed of production.
- Designed hotfixes based on A/B test data to rollout solutions in production.
Enlarged the "Related images" section to reduce friction in the search scenario
"Related images" enlargement
During research, I noticed people used the “Related images” section heavily to find the right image. But I've seen how badly "Related images" served that need—the narrow right-hand column forced too much scrolling and made the images feel tiny and easy to miss. So I pushed for a simple fix: move them below the main content.
Now, people can check a full screen of large images with just a single scroll. And for us, it opened up a profitable surface for placing ads and driving revenue.
Improved UI accents to eliminate user overload and drive CTR
UI accents Improvement
User testing kept surfacing the same issue: the interface felt overloaded. It raised the same questions internally—why are there three different components showing images? Why is the top-right so noisy? I stripped the noise, reduced competing elements, and CTAs.
"Visual accents are a powerful tool to drive users’ attention towards product’s value. Poor hierarchy is among the top reasons why products lose money."
Removed a low ROI feature so the business doesn't lose money to maintain it
Dead Feature Removal
Another problem I’ve discovered on UX's is that people didn’t understand that the gallery on the left navigates by images from the previous step. So the feature took the precious real estate above the fold just to distract. Moreover, devs wasted time ensure it works properly during each change.
We removed the gallery. Now, when users want to get back to the previous images, they just move to the step before. There they see familiar UI with big images and… ads (wink!).
"Dead features destroy your product. Business throws away resources to maintain them. Alternative costs are also involved—same resources could be invested in the feature that would scale the product’s revenue."
Dragged all ads to the right column to kill friction the user flow, while scaling profit
Ads relocation
The "find image" scenario still had friction due to ads being mixed in with related images. We pushed all ads to an anchored right column. We’ve unblocked the user scenario and scaled profit, despite reducing the total ad volume.
Built the process to ensure implementation meets high-quality standards
Before me, hand-offs were a mess: devs wasted hours parsing outdated mockups. On one occasion, they developed outdated design due to versioning ambiguity.
I’ve built hand-off process
- Prominently marked the final mockups as “Ready to Dev”.
- Removed all unnecessary, outdated layouts from the file.
- Marked them carefully—all states, corner cases, anticipated what might break during development, and provided the solutions how we would fix them.
We started to get implemented flows faster and with higher quality. Moreover, we improved relationships within the team.
Hotfixed issues during the experiments to prevent metrics dip
In a 100M+ MAU ecosystem, the ultimate challenge is safe deployment. Historically, a 2013 redesign of Yandex Search wiped out tenth of revenue. To avoid repeating that catastrophe, we adopted an incremental shipping strategy. I intentionally restored some of familiar UI elements to avoid disrupting established user mental models and habits, ensuring growth without the risk of a collapse.
We successfully shipped the redesign and scaled profit.
This was made possible by a tight strategic partnership: the PM and I worked late into the night to draft hotfixes for the experiment, followed by 2 AM brainstorms for the next iterations.
Finally, my ability to draft mockups in minutes transformed months of work into days of high-speed discovery.
Get a Design Partner
Dominate your niche with premium interfaces that earn user trust and drive revenue.
Free
No micro-managment
I take full ownership of the design process.
A design partner, not a freelancer
I stay until we see the results.
LTV
Premium craft UI
Build trust, cut down CPA and raise LTV
Book a 15-min Intro Call
Check Available Slots
only ×6 slots / week
Влад
Ви
I work only with the people with the highest standards.
Direct Inquiry
Medium
Product Hunt
© Vlad Vi, 2026
How to redesign a 100M+ MAU service, to unlock +7.38% in revenue
For over 20 years, the service had been over-optimized for revenue. Any minor pixel shift "bled" money metrics. But working autonomously alongside the Lead PM, We’ve rolled out a modern UI that not only survived the A/B tests but significantly grew the bottom line.
What I’ve Done
100% "Turnkey" delivery. I handled the entire design lifecycle—from the first concept to production. A lead designer wasn't engaged in management.
Learn More
Results
+7.38%* increase in desktop revenue.
Measurable growth across all key user satisfaction metrics.
*Due to the multi-billion ruble baseline of the service, even a fractional percentage increase represents millions in absolute revenue growth.
"I just found devs for the redesign. We got 3 days to get approved mocks ready for dev."
—Lead Product Manager, Yandex Images
Business Goal: money, money, fucking money
A redesign is never the goal. I quickly aligned with the PM on the real objective: make more money. It’s always about money. Yes, usability. Yes, user happiness. Yes, “such a nice UI!”. But in the end, survive only the ideas that grow the bottom line. Period.
Playing on "Insane" Difficulty: 100M+ users, 3 days, and 20+ Stakeholders
- The Revenue Trap: The viewer was so heavily monetized that past redesign attempts failed. Any changes historically dropped revenue, making stakeholders extremely risk-averse.
- The "We Tried This" Barrier: Every early draft faced the classic corporate pushback: "We already tried this, it didn't work." It took high-level creativity and hard data to squeeze more profit out of a 20-year-old flow.
- Heavy Stakeholder Management: Search is massive. We had to navigate and align 20+ Decision Makers and Influencers using drive and bulletproof UX arguments to push scale changes through.
- Universal Edge-Cases: With 20M+ DAU searching for everything from postcards to extreme NSFW content, the UI solution had to be bulletproof across all possible scenarios.
What I’ve Done: Design from 0 to Production
- Defined the objectives, metrics, and tech bandwidth with the PM to avoid burning resources on low-ROI ideas.
- Conducted ×32+ interviews to reveal problems and opportunities.
- Delivered ×7 major iterations within 72 hours to grab a limited spot to approve solutions with C-level reps.
- Ran ×6 UX tests to kill bad ideas.
- Defended mocks with PM to get approval, push mocks to devs ASAP, and to unblock development.
- Worked out all edge-case states to avoid wasting dev’s time.
- Built the Dev Hand-off and Design Review pipeline to ensure the solutions deliver the intended value, and we are not ashamed of production.
- Designed hotfixes based on A/B test data to rollout solutions in production.
Enlarged the "Related images" section to reduce friction in the search scenario
"Related images" enlargement
During research, I noticed people used the “Related images” section heavily to find the right image. But I've seen how badly "Related images" served that need—the narrow right-hand column forced too much scrolling and made the images feel tiny and easy to miss. So I pushed for a simple fix: move them below the main content.
Now, people can check a full screen of large images with just a single scroll. And for us, it opened up a profitable surface for placing ads and driving revenue.
Improved UI accents to eliminate user overload and drive CTR
UI accents Improvement
User testing kept surfacing the same issue: the interface felt overloaded. It raised the same questions internally—why are there three different components showing images? Why is the top-right so noisy? I stripped the noise, reduced competing elements, and CTAs.
"Visual accents are a powerful tool to drive users’ attention towards product’s value. Poor hierarchy is among the top reasons why products lose money."
Removed a low ROI feature so the business doesn't lose money to maintain it
Dead Feature Removal
Another problem I’ve discovered on UX's is that people didn’t understand that the gallery on the left navigates by images from the previous step. So the feature took the precious real estate above the fold just to distract. Moreover, devs wasted time ensure it works properly during each change.
We removed the gallery. Now, when users want to get back to the previous images, they just move to the step before. There they see familiar UI with big images and… ads (wink!).
"Dead features destroy your product. Business throws away resources to maintain them. Alternative costs are also involved—same resources could be invested in the feature that would scale the product’s revenue."
Dragged all ads to the right column to kill friction the user flow, while scaling profit
Ads relocation
The "find image" scenario still had friction due to ads being mixed in with related images. We pushed all ads to an anchored right column. We’ve unblocked the user scenario and scaled profit, despite reducing the total ad volume.
Built the process to ensure implementation meets high-quality standards
Before me, hand-offs were a mess: devs wasted hours parsing outdated mockups. On one occasion, they developed outdated design due to versioning ambiguity.
I’ve built hand-off process
- Prominently marked the final mockups as “Ready to Dev”.
- Removed all unnecessary, outdated layouts from the file.
- Marked them carefully—all states, corner cases, anticipated what might break during development, and provided the solutions how we would fix them.
We started to get implemented flows faster and with higher quality. Moreover, we improved relationships within the team.
Hotfixed issues during the experiments to prevent metrics dip
In a 100M+ MAU ecosystem, the ultimate challenge is safe deployment. Historically, a 2013 redesign of Yandex Search wiped out tenth of revenue. To avoid repeating that catastrophe, we adopted an incremental shipping strategy. I intentionally restored some of familiar UI elements to avoid disrupting established user mental models and habits, ensuring growth without the risk of a collapse.
We successfully shipped the redesign and scaled profit.
This was made possible by a tight strategic partnership: the PM and I worked late into the night to draft hotfixes for the experiment, followed by 2 AM brainstorms for the next iterations.
Finally, my ability to draft mockups in minutes transformed months of work into days of high-speed discovery.
Get a Design Partner
Dominate your niche with premium interfaces that earn user trust and drive revenue.
Free
No micro-managment
I take full ownership of the design process.
A design partner, not a freelancer
I stay until we see the results.
LTV
Premium craft UI
Build trust, cut down CPA and raise LTV
Book a 15-min Intro Call
Check Available Slots
only ×6 slots / week
Влад
Ви
I work only with the people with the highest standards.
Direct Inquiry
Medium
Product Hunt
© Vlad Vi, 2026