Case study

Transforming tourism with AI: scalable design in action

Case study

Transforming tourism with AI: scalable design in action

TL’DR

This AI-driven program for the tourism industry spanned 5 products and a team of 25+ developers. One key project, Review Spotter, used sentiment analysis to help tour operators. I helped define requirements, facilitate workshops, and ensure a seamless experience across products. I also created a scalable component library, which continues to support the products even after my departure in 2021.

This AI-driven program for the tourism industry spanned 5 products and a team of 25+ developers. One key project, Review Spotter, used sentiment analysis to help tour operators. I helped define requirements, facilitate workshops, and ensure a seamless experience across products. I also created a scalable component library, which continues to support the products even after my departure in 2021.

Areas of responsibility

Product design

UX/UI

Product Strategy

UX/UI

This program was centered around innovation, with a focus on creating industry-firsts for the tourism sector. Driven by the CEO's passion, an R&D team was created to develop AI-powered solutions tailored for tour operators.

While I can’t cover all five products here, let’s dive into one where we leveraged machine learning and NLP for sentiment analysis, specifically analyzing reviews for tour operator products, “Review Spotter”.

We developed a two-part solution:

  1. Admin Platform

  2. Review Insights

This program was centered around innovation, with a focus on creating industry-firsts for the tourism sector. Driven by the CEO's passion, an R&D team was created to develop AI-powered solutions tailored for tour operators.

While I can’t cover all five products here, let’s dive into one where we leveraged machine learning and NLP for sentiment analysis, specifically analyzing reviews for tour operator products, “Review Spotter”.

We developed a two-part solution:

  1. Admin Platform

  2. Review Insights

Admin platform

When I was introduced to this idea, I was curious to understand how everything would come together, especially since I had only a basic understanding of the technology behind it. Fortunately, my development background helped, and the team was more than willing to bring me up to speed.

After multiple discussions, I facilitated a workshop with the Product Owner, tech leads, and business analysts to strike the right balance between user experience, technological feasibility, and business goals. Finally, we moved forward with the outcome, shaping the Admin Platform into a well-designed system.

Review insights

When we finalized the Admin Platform, a second requirement emerged—one we had anticipated but weren’t quite sure how to execute effectively. That led us to a second round of workshops with stakeholders.

Ideally we had all these analytics in the dashboard of the Admin platform, yet we wanted to bring this to public view. The logical approach was to repurpose what we had while ensuring the Admin Platform retained control over what was shared.

I had a great time bridging the gaps between the platforms and to ensure the user experience is seamless across both.

When we finalized the Admin Platform, a second requirement emerged—one we had anticipated but weren’t quite sure how to execute effectively. That led us to a second round of workshops with stakeholders.

Ideally we had all these analytics in the dashboard of the Admin platform, yet we wanted to bring this to public view. The logical approach was to repurpose what we had while ensuring the Admin Platform retained control over what was shared.

I had a great time bridging the gaps between the platforms and to ensure the user experience is seamless across both.

The design process

This project kicked off back in 2019, before Figma introduced ‘Tokens’ and ‘Components.’ Fortunately, my UI development background came in handy, allowing me to componentize similar to using Sass variables. This approach turned out to be the perfect solution, seamlessly scaling across all five products I mentioned earlier.

I developed a plan to establish a single source of truth for components, ensuring consistency and efficiency. Over time, this evolved into a full-fledged component library that seamlessly scaled as the products expanded and required more components.

This project kicked off back in 2019, before Figma introduced ‘Tokens’ and ‘Components.’ Fortunately, my UI development background came in handy, allowing me to componentize similar to using Sass variables. This approach turned out to be the perfect solution, seamlessly scaling across all five products I mentioned earlier.

I developed a plan to establish a single source of truth for components, ensuring consistency and efficiency. Over time, this evolved into a full-fledged component library that seamlessly scaled as the products expanded and required more components.

The outcome

Since I documented all my processes, designs, and developments thoroughly, handoff was seamless. With a strong foundation in place, even after I moved on from the project in 2021, the products and designs I contributed to are still in use today, proving the long-term value of well-structured systems.

Since I documented all my processes, designs, and developments thoroughly, handoff was seamless. With a strong foundation in place, even after I moved on from the project in 2021, the products and designs I contributed to are still in use today, proving the long-term value of well-structured systems.