Describe how Tripadvisor uses AI both internally and in your customer facing proposition?
AI is at the heart of our core features and products that we build for our travelers and partners. We have structured three pillars of AI across our organisation: product, productivity and partnerships. Across both strategy and impact, we have a thoughtful, pragmatic and innovative approach. As it relates to product features, we have been already leveraging AI to drive our customer-focused experiences like personalisation, recommendations, sort and pricing. Recently, we have been pushing into GenAI intelligence based features like our AI trip builder.
Our AI trip builder is a free tool which gives you a personalised itinerary for your destination of choice, based on who you’re travelling with, and your interests. The itinerary will include recommendations on hotels, restaurants and activities in your chosen location, informed by over one billion reviews and contributions from real travellers. So far, our AI trip builder has generated over 2.9 million personalised travel recommendations. We also offer an AI assistant, where travelers interact with an AI chatbot to provide personalised answers to travel related questions
In addition, we are using AI to automate workflows to facilitate efficiencies.
Go deeper with GlobalData
Describe Tripadvisor’s AI adoption milestones?
We are the place travellers go when they are looking for personalisation, relevance and guidance, and the wealth of our data natural resources cannot be overstated. AI adoption milestones include that we’ve been using ML for sorting and recommendations for over a decade starting around 2013. We continued to refine those systems to best serve travelers, until this recent wave of AI.
Over the course of years, we launched many personalisation, pricing, marketing optimisation features and enhancements using AI – so that has been tremendously impactful. Today, we’re well into our journey to push heavily into GenAI capabilities.
Early in 2023, we piloted and then launched many features like AI trip builder and AI review summaries. You can now see AI across our site, including our newly launched AI assistant. We have AI adoption for internal applications, as well, for solving particular business use cases. As the data and AI stack matures, we are always evolving to drive even bigger impact with AI capabilities.
As a brand, we have achieved tremendous scale and more importantly depth and breadth of data. This is evident in our 300 million monthly active users (MAU), over a billion reviews and contributions, 11 million businesses and 300,000 plus attractions contributions. We see our AI journey similar to consumer technology companies, like Amazon, Uber, etc. Bringing the best of AI, data, and technology to serve our travelers, team, and partners has always been and will continue to be a top priority for Tripadvisor.
How well do you really know your competitors?
Access the most comprehensive Company Profiles on the market, powered by GlobalData. Save hours of research. Gain competitive edge.
Thank you!
Your download email will arrive shortly
Not ready to buy yet? Download a free sample
We are confident about the unique quality of our Company Profiles. However, we want you to make the most beneficial decision for your business, so we offer a free sample that you can download by submitting the below form
By GlobalDataWhen did you leverage Snowflake’s AI Data Cloud for Travel and Hospitality?
We recognise that to do AI right, you need solid data foundations. For us, that happens internally and with the support of our partners. Snowflake’s AI Data Cloud is one important component of our strategy, we went fully on AWS and Snowflake two years ago, and we’re modernising our entire data stack and using best of breed technologies. Data and AI is an executive level priority with all the leaders in the company deeply committed and very well versed in what high quality data requires and the value it offers, especially in AI and machine learning.
Our scale of data is very unique in the travel industry – we have one billion plus reviews and contributions, 10 petabytes of structured data – so a great combination of structured, semi-structured and unstructured data. Plus, we operate across 43 markets in 22 languages which makes it very complex from a data availability, uptime and privacy standpoint.
All of this scale, cost and performance plus our upcoming AI first robust roadmap were factors in our decision to choose the best data storage, compute and data cloud solution provider for us. We have always been an innovator, focused on being a data-first company that drives growth. As such, we chose Snowflake’s platform as it offers us a single, integrated platform, secure data sharing and near-zero maintenance, with near unlimited scale, concurrency, and performance.
What was the selection criteria for choosing a platform and why Snowflake?
We needed a cloud native AI platform that streamlined operations, had a future-forward product strategy driven by AI, and was more focused on delivering value than marketing. We needed an architecture that separates functions and offers us flexibility so we can build and move at the pace of our business (for example separating storage and compute, click of a button administration as opposed to countless hours spent fine-tuning).
We also needed a platform that provides ever-evolving horizontal features that drive AI ease-of-use across multiple and varied use cases across the enterprise, like Snowpark, Cortex, and Streamlit. We needed a company and a platform that evolves quickly at the pace of our business, enhancing our platform underneath us so we can focus on what we do best.
We originally selected Snowflake to establish a beachhead for our critical datasets requiring redundancy and protection against accidental data deletion
Given the success, we onboarded more and more datasets and use cases, and given the features Snowflake offers the process was easier to manage with minimal complications.
The act of onboarding our data processing workloads to Snowflake was easy. The challenge was determining the sequencing of what to move and when given various dependencies. The full migration took two years end-to-end, but the primary datasets were migrated in well under a year. We started execution in Jan 2022 and completed execution in Jan 2024. In addition, Snowflake also helped us easily integrate with modern SaaS tooling and bootstrap our data compliance, governance and data clean room program
What security and reputational challenges does AI present for your business and how are you addressing them?
AI features and capabilities are very powerful and create several user benefits. However, we need to be aware and cautious of certain risks like trust, data/AI privacy, ethics and bias in AI. We take this as a core design principle and build our products and features with that in mind.
We protect our data, enjoy high trust and are committed to data privacy. Our travelers and partners know this about us: people come to us for authentic travel guidance which comes from safeguarding our data and our users.
The Human-in-the-Loop (HitL) is an essential part of any successful and healthful AI product which is why we’re committed to HitL processes. Another core operating principle for us is AI governance and safety. Safe AI starts with the data, and our commitment to trust and data privacy is evident in how embedded best-in-class governance features are across our data ecosystem.
Tripadvisor was one of the first review platforms to prohibit reviews that we identified as AI generated from our platform. As a result, we’ve removed over 300,000 reviews we’ve detected to be AI-generated. However, as GenAI is adopted more widely, we can envision scenarios where the technology could support users in their engagement with our platform (for example, people with disabilities who might struggle to type a review). As with all emerging technologies and trends, Tripadvisor will continue to evaluate the impact on traveler trust and how to adapt our policy in the future.
How will Tripadvisor use AI over the next five years, ten years?
We see AI democratising our business in incredible ways that will help our travelers and partners. With latest features in LLMs, and overall capabilities like multi-modal inputs, reasoning and AI agentic workflows, sky is the limit of what can be done.
From turbocharging how fast our teams can learn new skills and master new aspects of our business, to how we personalise our communications and guidance, we’re excited about the potential.
However, as a travel brand, we know nothing replaces the joy of going in person to a place and experiencing it with your own eyes. We hope that it simply accelerates planning the best trip for each individual, and makes us more efficient along the way.
