The digital product development landscape has always been clear-cut: Designers envision the interface and engineers make it a reality. This was particularly true in the travel industry, with its complex booking flows and search engines. However, the ascent of AI is rapidly rewriting this narrative, leading to an evolutionary shift in the designer-engineer dynamic.
BEYOND CHATBOTS: THE ADVENT OF VISUAL AI
Envision the scenario of a traveler organizing their Southern California getaway, scouring a hospitality brand’s booking site for accommodations that grant easy access to both serene hiking paths and the thrill of the surf. If this scenario were proposed in H1 2023, the likely solution would be an AI chatbot. Designers would craft the chatbot’s interface, and engineers would construct the system that enables dialogue with a generative AI model, providing a text-heavy selection of choices to the user.
Just as we’ve progressed beyond the basic interfaces of the past, we’re experiencing a similar leap in AI experience design. The cutting-edge vision for AI in travel tech features an interface that adapts as fluidly as the information it presents.
This new paradigm eschews linear conversation for a dynamic, visual space where the traveler’s input and the AI’s feedback are not just communicated but explored. The traveler’s inputs and the AI’s responses are visualized, letting guests navigate across multiple paths and allowing them not only to branch out from any point, but also to climb up and down the hierarchy of abstraction.
THE AI-DRIVEN UX
In this envisioned AI-driven UI, users would seamlessly transition from specific search results to a higher-level overview that abstracts away unnecessary details to emphasize the traveler’s unique vacation criteria. It would dynamically present contextual options, versus simply showing a set of predefined filters and markers.
As the traveler narrows their search, the AI could shift to a comparative view, allowing them to easily juxtapose their options. Whether it’s a side-by-side comparison of a few properties or a tabular overview of many, the AI could present information in an accessible format.
Imagine a plot where each lodging option is marked by user ratings and proximity to activities, painted in a visual language that speaks directly to the user’s criteria. From this abstracted view, the traveler could effortlessly pinpoint and book their ideal stay.
DESIGNER-ENGINEER SYNERGY
This pivot from a linear to a dynamic user-centric experience is more than a technological shift—it signifies a deeper fusion of design and development roles. The creative and technical aspects are no longer siloed but are converging into a unified process of discovery and implementation.
The most significant change in the workflow is that designers and engineers now initiate their collaborative process with the generative AI model, not the user interface. They delve into the solution space together, with the AI providing suggestions that are jointly assessed for their practicality and value to the user.
There’s a shared commitment to steering the AI’s learning process to enhance the user experience. This involves deep discussions about data, the implications of prompting and fine-tuning the model, and how it dovetails with the broader data ecosystem of the brand.
The collaboration involves cycles of testing and refining their interactions with the AI model. Engineers provide results from prompt engineering experiments, which inform design decisions. Designers interpret these results to envision interfaces that can guide the AI toward more accurate and helpful responses. This iterative process accelerates beyond traditional development cycles, with daily enhancements and evaluations focusing on improving AI outputs.
When adopting a shared mindset toward solving the complex problems that arise with AI such as inconsistent outputs or hallucinations, designers and engineers would need to collaboratively understand the impact of each design or development choice on the model’s behavior, and patiently work through to overcome the challenges.
INTEGRATING INTELLIGENCE WITH FLUIDITY
In this generative user experience, static user flows are replaced with agile UI elements capable of morphing to accommodate the fluid nature of AI-generated content.
Engineers organize data and craft algorithms for presenting information at varying levels of abstraction—for example, autonomously curating summaries tailored to the guest’s intent, identifying guest-relevant insights within imagery, and extracting critical data points such as targeted reviews.
Designers ensure that the information presented is lucid and actionable, maintains a seamless flow between data complexities, and weaves these AI-derived insights in a manner that is both meaningful in context and visually compelling.
Each level of data abstraction is crafted to support user decisions, like making a booking without causing disorientation or the need to switch views. This necessitates precise choices about the data presented or omitted at each stage. Achieving a seamless flow between varying data granularities is essential, calling for a finely tuned balance between functionality and performance.
EMBEDDING ETHICS
Designers and engineers should also collectively detect and navigate AI complexities such as inconsistent outputs or biases. It’s imperative to understand how each design and development decision influences the model’s bias and take corrective action to ensure the AI’s reasoning is transparent and free of it.
An example is ensuring property listings are not biased towards certain locations, which could skew the user’s options based on pre-existing biases. Collaboratively embedding ethical considerations into the design and development sprints is crucial.
EMBRACING UNITY: THE KEY TO SUCCESS
Gone are the days of designers crafting a vision and simply passing the baton to engineers. Such an approach, in the realm of generative AI, is a recipe for inefficiency and subpar products.
For travel and hospitality brands and enterprises, this evolution is a call to action. The traditional model of compartmentalized teams or outsourcing segments of a project to different vendors won’t just lead to wasted resources—it could spell the difference between a revolutionary product and a mediocre one.
In this new era, the successful development of generative experiences will depend heavily on teams that are united—not just in their skills, but in their aspirations and commitment to the creation of a future where the traveler’s digital experience is as enriching as the physical journey they are planning.
Ravi Evani is the CTO for Travel & Hospitality and leads a global, large-scale engineering organization at Publicis Sapient.