Ninetailed AI
Ninetailed (acquired by Contentful in 2024) · 2024
Overview
Ninetailed (acquired by Contentful in 2024) is the leading personalization and experimentation technology that empowers brands to understand their audience and create, deliver, and test tailored experiences.
In 2024, I led the design of the AI Copilot initiative, aimed at redefining how users create and optimize personalized experiences through intelligent suggestions and automation.
Overview
Ninetailed (acquired by Contentful in 2024) is the leading personalization and experimentation technology that empowers brands to understand their audience and create, deliver, and test tailored experiences.
In 2024, I led the design of the AI Copilot initiative, aimed at redefining how users create and optimize personalized experiences through intelligent suggestions and automation.
Overview
Ninetailed (acquired by Contentful in 2024) is the leading personalization and experimentation technology that empowers brands to understand their audience and create, deliver, and test tailored experiences.
In 2024, I led the design of the AI Copilot initiative, aimed at redefining how users create and optimize personalized experiences through intelligent suggestions and automation.
Impact
The AI Copilot feature led to a 30% increase in product adoption and a 17% rise in upsell opportunities within one quarter.
Its success played a pivotal role in Ninetailed’s acquisition by Contentful, where it evolved into Contentful’s AI Suggestions, now used by enterprise clients worldwide.
Impact
The AI Copilot feature led to a 30% increase in product adoption and a 17% rise in upsell opportunities within one quarter.
Its success played a pivotal role in Ninetailed’s acquisition by Contentful, where it evolved into Contentful’s AI Suggestions, now used by enterprise clients worldwide.
Impact
The AI Copilot feature led to a 30% increase in product adoption and a 17% rise in upsell opportunities within one quarter.
Its success played a pivotal role in Ninetailed’s acquisition by Contentful, where it evolved into Contentful’s AI Suggestions, now used by enterprise clients worldwide.
Role
Lead Product Designer
Role
Lead Product Designer
Role
Lead Product Designer
Responsibilities
Product Strategy
UX/UI
Design System
Prototyping
Responsibilities
Product Strategy
UX/UI
Design System
Prototyping
Responsibilities
Product Strategy
UX/UI
Design System
Prototyping
Business need
At the start of 2024, during a strategic planning session, our leadership team outlined the need to maintain the rapid growth trajectory established in 2023. After reviewing several options it was decided to focus on the following goals:
Ninetailed market leadership
Opportunities for upselling additional services and features
Business need
At the start of 2024, during a strategic planning session, our leadership team outlined the need to maintain the rapid growth trajectory established in 2023. After reviewing several options it was decided to focus on the following goals:
Ninetailed market leadership
Opportunities for upselling additional services and features
Business need
At the start of 2024, during a strategic planning session, our leadership team outlined the need to maintain the rapid growth trajectory established in 2023. After reviewing several options it was decided to focus on the following goals:
Ninetailed market leadership
Opportunities for upselling additional services and features
Product statement
Apart from strategic value, product party saw a massive opportunity in embedding AI in the process of experiences creation in a way of AI Copilot. For Ninetailed to become a leader in personalization, we needed to develop a virtual companion to help users create unique and tailored user experiences. We had to ensure it would create actionable suggestions to design, modify and optimise experiences with ease and precision.
Product statement
Apart from strategic value, product party saw a massive opportunity in embedding AI in the process of experiences creation in a way of AI Copilot. For Ninetailed to become a leader in personalization, we needed to develop a virtual companion to help users create unique and tailored user experiences. We had to ensure it would create actionable suggestions to design, modify and optimise experiences with ease and precision.
Product statement
Apart from strategic value, product party saw a massive opportunity in embedding AI in the process of experiences creation in a way of AI Copilot. For Ninetailed to become a leader in personalization, we needed to develop a virtual companion to help users create unique and tailored user experiences. We had to ensure it would create actionable suggestions to design, modify and optimise experiences with ease and precision.
High-level metrics
By understanding data from Insights, user behaviour as well as the context users are modifying and publishing content for, we could integrate AI Copilot across all of Ninetailed to increase engagement and product adoption. All while focusing on experimenting and personalizing in a risk-free way.
High-level metrics
By understanding data from Insights, user behaviour as well as the context users are modifying and publishing content for, we could integrate AI Copilot across all of Ninetailed to increase engagement and product adoption. All while focusing on experimenting and personalizing in a risk-free way.
High-level metrics
By understanding data from Insights, user behaviour as well as the context users are modifying and publishing content for, we could integrate AI Copilot across all of Ninetailed to increase engagement and product adoption. All while focusing on experimenting and personalizing in a risk-free way.
Hypotheses
Based on an accumulated user knowledge and insights shared by customer success team I have listed down hypotheses that I wanted to check for validity with the following interviews round.
Hypothesis 1: The current setup for creating personalized experiences and experiments isn’t scalable. Automating these processes will allow users to scale effectively.
Hypothesis 2: Content creation takes too much effort. AI-driven content generation will improve quality while saving time.
Hypothesis 3: Users struggle to identify where and how to begin personalizing. AI recognition of existing audiences and components will guide users in making impactful personalization decisions.
Hypotheses
Based on an accumulated user knowledge and insights shared by customer success team I have listed down hypotheses that I wanted to check for validity with the following interviews round.
Hypothesis 1: The current setup for creating personalized experiences and experiments isn’t scalable. Automating these processes will allow users to scale effectively.
Hypothesis 2: Content creation takes too much effort. AI-driven content generation will improve quality while saving time.
Hypothesis 3: Users struggle to identify where and how to begin personalizing. AI recognition of existing audiences and components will guide users in making impactful personalization decisions.
Hypotheses
Based on an accumulated user knowledge and insights shared by customer success team I have listed down hypotheses that I wanted to check for validity with the following interviews round.
Hypothesis 1: The current setup for creating personalized experiences and experiments isn’t scalable. Automating these processes will allow users to scale effectively.
Hypothesis 2: Content creation takes too much effort. AI-driven content generation will improve quality while saving time.
Hypothesis 3: Users struggle to identify where and how to begin personalizing. AI recognition of existing audiences and components will guide users in making impactful personalization decisions.
Interviews
I ran a round of interviews with a selection of our current users. Considering Ninetailed had less than 30 customers at that time and not all of them were meeting technical readiness for this AI Copilot feature exploration, it was a challenging process to hire respondents at first place. In collaboration with the marketing team, I led the development of an Early Access Program (EAP), which granted participants access to beta version of AI Copilot in exchange for feedback. This way I was able to hire about 15 respondents initially which would eventually result in 7 regular users sharing their feedback 2-3 times over the course of 3 months.
Interviews
I ran a round of interviews with a selection of our current users. Considering Ninetailed had less than 30 customers at that time and not all of them were meeting technical readiness for this AI Copilot feature exploration, it was a challenging process to hire respondents at first place. In collaboration with the marketing team, I led the development of an Early Access Program (EAP), which granted participants access to beta version of AI Copilot in exchange for feedback. This way I was able to hire about 15 respondents initially which would eventually result in 7 regular users sharing their feedback 2-3 times over the course of 3 months.
Interviews
I ran a round of interviews with a selection of our current users. Considering Ninetailed had less than 30 customers at that time and not all of them were meeting technical readiness for this AI Copilot feature exploration, it was a challenging process to hire respondents at first place. In collaboration with the marketing team, I led the development of an Early Access Program (EAP), which granted participants access to beta version of AI Copilot in exchange for feedback. This way I was able to hire about 15 respondents initially which would eventually result in 7 regular users sharing their feedback 2-3 times over the course of 3 months.
Developing Iteration 1
Features
Experience Suggestions
Experience Suggestions are designed in a way to merge all key fields of a personalization or experiment page into one card. By utilizing advanced AI algorithms, Experience Suggestions analyzes an array of data points, including user behavior, demographics, and first-party data to provide the best possible experience for the specific audience segments.
We debated the optimal placement of the Experience Suggestions card within the user journey. Despite user preferences leaning toward starting at the component page, technical constraints led us to introduce the card on the audience page for the initial iteration, with plans to gather feedback for future enhancements.
Experience Suggestions
Experience Suggestions are designed in a way to merge all key fields of a personalization or experiment page into one card. By utilizing advanced AI algorithms, Experience Suggestions analyzes an array of data points, including user behavior, demographics, and first-party data to provide the best possible experience for the specific audience segments.
We debated the optimal placement of the Experience Suggestions card within the user journey. Despite user preferences leaning toward starting at the component page, technical constraints led us to introduce the card on the audience page for the initial iteration, with plans to gather feedback for future enhancements.
Experience Suggestions
Experience Suggestions are designed in a way to merge all key fields of a personalization or experiment page into one card. By utilizing advanced AI algorithms, Experience Suggestions analyzes an array of data points, including user behavior, demographics, and first-party data to provide the best possible experience for the specific audience segments.
We debated the optimal placement of the Experience Suggestions card within the user journey. Despite user preferences leaning toward starting at the component page, technical constraints led us to introduce the card on the audience page for the initial iteration, with plans to gather feedback for future enhancements.
AI Variant Generation
Because generation of variants was reported to be the most time and effort consuming task, the complexity of which would grow exponentially with the number of variants, I had designed a One-Click variant generation button. By analyzing users’ data, content, experience and audience insights, and content types, it achieved higher accuracy.
AI Variant Generation
Because generation of variants was reported to be the most time and effort consuming task, the complexity of which would grow exponentially with the number of variants, I had designed a One-Click variant generation button. By analyzing users’ data, content, experience and audience insights, and content types, it achieved higher accuracy.
AI Variant Generation
Because generation of variants was reported to be the most time and effort consuming task, the complexity of which would grow exponentially with the number of variants, I had designed a One-Click variant generation button. By analyzing users’ data, content, experience and audience insights, and content types, it achieved higher accuracy.
Usability Testing
To test usability of designs and quality of algorithm outputs we had to iterate in an extremely fast pace so we cut out any excessive functionality from iteration 1. The main goals of usability test #1 were to improve the usability of Experience Suggestions and Variant generation as well as explore if there were other features that would add value to the current users’ workflow.
Usability Testing
To test usability of designs and quality of algorithm outputs we had to iterate in an extremely fast pace so we cut out any excessive functionality from iteration 1. The main goals of usability test #1 were to improve the usability of Experience Suggestions and Variant generation as well as explore if there were other features that would add value to the current users’ workflow.
Usability Testing
To test usability of designs and quality of algorithm outputs we had to iterate in an extremely fast pace so we cut out any excessive functionality from iteration 1. The main goals of usability test #1 were to improve the usability of Experience Suggestions and Variant generation as well as explore if there were other features that would add value to the current users’ workflow.
Developing Iteration 2
Features
Experience Suggestions It.2
After gathering feedback from our users, I began iterating on the Experience suggestion card. I prototyped and tweaked specifics like clickable audience and component elements, fold/unfold card animations, and predictive analytics to help users make a decision easier.
My hypothesis regarding users starting their experience creation journey at the component page was proved during the interviews and so I have placed 2nd iteration of Experience Suggestion card on both pages: Audience and Component to align with a natural users’ workflow.
Experience Suggestions It.2
After gathering feedback from our users, I began iterating on the Experience suggestion card. I prototyped and tweaked specifics like clickable audience and component elements, fold/unfold card animations, and predictive analytics to help users make a decision easier.
My hypothesis regarding users starting their experience creation journey at the component page was proved during the interviews and so I have placed 2nd iteration of Experience Suggestion card on both pages: Audience and Component to align with a natural users’ workflow.
Experience Suggestions It.2
After gathering feedback from our users, I began iterating on the Experience suggestion card. I prototyped and tweaked specifics like clickable audience and component elements, fold/unfold card animations, and predictive analytics to help users make a decision easier.
My hypothesis regarding users starting their experience creation journey at the component page was proved during the interviews and so I have placed 2nd iteration of Experience Suggestion card on both pages: Audience and Component to align with a natural users’ workflow.
Audience Suggestions
While the pain points concerning scalability of personalization creation and time spent to create variants were solved with new features, Ninetailed users still seemed to be frustrated at the beginning of the process. Our analytics shown that 30% of our users had created only one Audience - All Visitors. Without experience in recognizing their own website visitor segments, it was simply hard to know where to start.
We introduced Audience Suggestions, powered by AI-driven analysis of behavior, demographics, and first-party data, to guide users in identifying optimal audience segments for personalizations and experiments.
Audience Suggestions
While the pain points concerning scalability of personalization creation and time spent to create variants were solved with new features, Ninetailed users still seemed to be frustrated at the beginning of the process. Our analytics shown that 30% of our users had created only one Audience - All Visitors. Without experience in recognizing their own website visitor segments, it was simply hard to know where to start.
We introduced Audience Suggestions, powered by AI-driven analysis of behavior, demographics, and first-party data, to guide users in identifying optimal audience segments for personalizations and experiments.
Audience Suggestions
While the pain points concerning scalability of personalization creation and time spent to create variants were solved with new features, Ninetailed users still seemed to be frustrated at the beginning of the process. Our analytics shown that 30% of our users had created only one Audience - All Visitors. Without experience in recognizing their own website visitor segments, it was simply hard to know where to start.
We introduced Audience Suggestions, powered by AI-driven analysis of behavior, demographics, and first-party data, to guide users in identifying optimal audience segments for personalizations and experiments.
Results
We observed a 30% increase in product adoption rate and a growth in user engagement after launching the AI Copilot feature.
This opened new upselling opportunities that contributed to a 17% revenue increase over the following quarter.
The success of this initiative became a key factor in Ninetailed’s acquisition by Contentful, a global leader in composable content management.
Today, the product lives on as AI Suggestions - a flagship feature publicly featured on Contentful’s website and used by enterprise customers worldwide.
Results
We observed a 30% increase in product adoption rate and a growth in user engagement after launching the AI Copilot feature.
This opened new upselling opportunities that contributed to a 17% revenue increase over the following quarter.
The success of this initiative became a key factor in Ninetailed’s acquisition by Contentful, a global leader in composable content management.
Today, the product lives on as AI Suggestions - a flagship feature publicly featured on Contentful’s website and used by enterprise customers worldwide.
Results
We observed a 30% increase in product adoption rate and a growth in user engagement after launching the AI Copilot feature.
This opened new upselling opportunities that contributed to a 17% revenue increase over the following quarter.
The success of this initiative became a key factor in Ninetailed’s acquisition by Contentful, a global leader in composable content management.
Today, the product lives on as AI Suggestions - a flagship feature publicly featured on Contentful’s website and used by enterprise customers worldwide.
Summary
By listening to users and iterating on real feedback, we built AI-driven suggestions that made personalization effortless and intuitive, turning complex experimentation into an accessible, everyday workflow.
This project not only improved core product metrics but also shaped Ninetailed’s long-term vision and market value, culminating in its acquisition by Contentful.
A few key learnings I took away:
There is always value in listening to users. Their insights reveal nuances that data alone can’t.
Small, iterative releases create momentum. Testing early and often allowed us to validate ideas quickly and refine AI outputs with confidence.
Designing AI features means designing for trust. Transparency, predictability, and user control are essential for adoption and long-term engagement.
Summary
By listening to users and iterating on real feedback, we built AI-driven suggestions that made personalization effortless and intuitive, turning complex experimentation into an accessible, everyday workflow.
This project not only improved core product metrics but also shaped Ninetailed’s long-term vision and market value, culminating in its acquisition by Contentful.
A few key learnings I took away:
There is always value in listening to users. Their insights reveal nuances that data alone can’t.
Small, iterative releases create momentum. Testing early and often allowed us to validate ideas quickly and refine AI outputs with confidence.
Designing AI features means designing for trust. Transparency, predictability, and user control are essential for adoption and long-term engagement.
Summary
By listening to users and iterating on real feedback, we built AI-driven suggestions that made personalization effortless and intuitive, turning complex experimentation into an accessible, everyday workflow.
This project not only improved core product metrics but also shaped Ninetailed’s long-term vision and market value, culminating in its acquisition by Contentful.
A few key learnings I took away:
There is always value in listening to users. Their insights reveal nuances that data alone can’t.
Small, iterative releases create momentum. Testing early and often allowed us to validate ideas quickly and refine AI outputs with confidence.
Designing AI features means designing for trust. Transparency, predictability, and user control are essential for adoption and long-term engagement.
