Boost sales conversion with Conversational AI Product Recommendation

Currently, travel companies convert online up to 50 times less than in physical agencies or call centers.

Online conversion rates range from 0.2% to 4% compared to an average of 22% in physical agencies

The primary issue is that the products being sold are overly complex, with numerous variations and options. Customers visiting the website not only need to have a clear idea of what they want, but they are also faced with making an overwhelming number of decisions.

This complexity often drives customers to contact the call center or, worse, abandon their search or project entirely because they cannot identify or understand the right product for them. This results in an opportunity cost of tens of millions of euros in potential business for an average travel agency.

Inefficient search systems

Marketing and sales teams face numerous challenges:

1. Many users leave the website without finding what they’re looking for, even though the product is available in the catalog. The main culprit is the search engine, which is limited to keywords and imposes rigid entry points. For example, if the user does not specify a destination or travel date, they are unable to start their search.

‍

‍

2. A significant portion of users seek human assistance for specific questions, such as whether a hotel room or cruise cabin offers a double bed. This issue is exacerbated when these inquiries, often made on weekends or outside of business hours, go unanswered.

On average, about 80% of shopping carts are abandoned in the travel industry

3. Teams spend hours labeling each product, yet they can never be truly exhaustive. With thousands of listings to update and 15 minutes required for each one, the workload from this ‘criteria-setting’ effort quickly becomes unmanageable. Furthermore, the criteria are often insufficient and become outdated as trends change, making it difficult for consumers to find the products they are looking for.

Generative AI failed to deliver; technology has evolved in the meantime

Customer engagement through conversational interfaces is not a new issue: who hasn't encountered a chatbot unable to understand their intentions? Following the success of ChatGPT in November 2022, generative AI was touted as the solution that would finally provide human-like conversation quality and significantly enhance the experience for potential customers.

Once again, this turned out to be a disappointment.

‍

Two years later, the situation is quite different. Technology has advanced significantly, providing better control and enhanced features. Customers have adapted to this new mode of communication and appreciate its benefits. As a result, companies in the travel sector have shifted from a “build-and-fail” mentality to a “buy-and-profit” approach, focusing on creating economic value.

Today, there is a genuine opportunity for innovative companies to differentiate themselves in the rapidly growing travel market, where customer expectations for unique experiences are increasingly high.

Autodm AI offers personalized product recommendations

To meet this demand, we offer a conversational solution specialized in sales. This solution provides a unique shopping experience, particularly through its recommendation module.

Autodm AI’s product recommendation module

This module enables users to:

1. Describe their search in natural language, such as: “I want to see mountains, visit temples, and relax on the beach for under €5,000.”
   • Natural language search allows for identifying the nuances of the prospect's preferences and optimally leveraging the catalog.
2. Request specific details about offers, such as: “Does the hotel on day 3 have a children-friendly pool with a waterslide?”
   • Traditionally, reservations requiring significant human engagement, like cruises, which are finalized over the phone in 70% of cases, can now be completed online with clear, precise, and instant responses.
3. Filter various offers by interest themes, making exploration simple and intuitive, with dynamically generated filter buttons like: “Family,” “Romantic,” “Itinerant,” “Beach,” and “Mountains.”
   • Users can quickly discover the most relevant offers based on their selected themes, without having to sift through hundreds of results.

4. Find the most relevant offers by personalizing search results based on the user and the business objectives of the travel company.
   • The dynamic ranking of search results is adjusted to prioritize offers that align with business rules (e.g., maximizing high-margin sales) and personalization (e.g., prioritizing offers similar to the user's purchase history).

These features make it easier to access products and enhance the user experience by making the search process smoother and more intuitive.

The technology enabling this new type of discovery and recommendation experience for complex products relies on an approach that combines AI Agents with Knowledge Graphs. This method dynamically connects offers through a graph structure built from product information (price, location, size, etc.) and descriptions (facilitating semantic searches), and it can be progressively enriched with additional data (customer profiles and preferences, bestsellers, trending destinations, etc.).

If you share our vision of providing a memorable shopping experience for your customers through conversational AI in travel, or if you have any questions or comments, please don't hesitate to reach out to us!