Artificial Intelligence & Hospitality: A Good Partnership
By Robert O'Halloran
The hospitality industry seeks to embrace AI for its support and enhancement of guest experiences, operational efficiency, planning and positive impact on the bottom line. That is a tall order, and operators need to identify the optimal mix of personal service with technological efficiency.
I grew up in an industry where personal service was a leading trait and value for the best operators. Now, given the supply and demand for employees, labor appears more complex problem to solve for operating frameworks. For me personally, I don’t need any more apps on my phone than I already have. I would argue that not everything is better delivered by technology. I admit that I am part of a mature target market (baby boomers) that may want to interact with people when engaging in travel, lodging, and dining in the greater hospitality and tourism industry. Be that as it may, AI and advanced technologies are here to stay, and we need to embrace this.
AI is defined as a family of technologies that can recognize, analyze, act, learn and demonstrate advanced features of human intelligence in the process of problem-solving (McCartney & McCartney, Citation2020 ). I am not against technology and appreciate its value and or the value it can help us create. My caution is to remember that AI like other technology applications are tools for people to use to enhance performance and productivity. Understanding the balance between automated assistance and the human touch is crucial in hospitality. The performance enhancement leveraged by AI may be to bring data and new ideas to the forefront for consideration when planning and delivering hospitality products and services.
AI and Hospitality Management
For the hospitality industry, AI can help to transform lodging operations, revenue management, marketing, sales and MICE, guest experience, human resources and labor, and optimally enhance these functional areas for large chains and small independents. I concur with the theory that hoteliers who do not engage with AI and its applications will be left behind. It is imperative for brands and operators to utilize AI and its tools and connections to existing software to optimize business results. The following discussion highlights some of these opportunities.
The Hotel Tech Report reported in their State of Hotel Guest Technology Report 2024 that 70% of guests find chatbots helpful for simple inquiries but prefer human interaction for more complex requests. AI can automate repetitive tasks theoretically, allowing for better service. Additionally, AI’s use of predictive analytics is optimizing pricing and revenue management through data analysis and predictive modeling. AI can also be supportive of sustainability efforts. For example, housekeeping room assignments and maintenance can be supported by using sensors and cameras to monitor and control temperature, lighting, and the security of guest rooms, and the improvement of energy efficiency by using smart building systems.
The integration of AI into the hospitality sector marks a significant shift in how hotels deliver customer service. AI transforms customer processes and services by engaging smart and predictive customer care and by employing predictive and augmented product and service design. AI-powered chatbots and virtual assistants can also be helpful with human resource tasks and be helpful in interacting with job applicants. To reiterate, some people want to talk with a person and not a machine if they are seeking employment. However, it is noted that more and more virtual hiring companies are available.
For example, Indeed, Ziprecruiter, and FlexJob s advertise heavily. I have worked with a hotel operator who has shifted his focus from operating hotels to developing hotels. His reasoning was that it was so difficult to find and keep housekeeping employees, his cost/benefit analysis led him to focus on a more defined aspect of the lodging industry. AI is not just helpful with recruiting but also with scheduling and forecasting demand. Think of AI as a restaurant staffing guide tool; think of the analysis of a staffing guide process.
Estimating what the forecasted demand for a restaurant is to be, an operator wants to schedule an optimal number of employees. AI can quickly analyze hours in a shift, expected covers, average checks and the level of service to be delivered. That is upscale or at least boutique restaurants will likely provide a more relaxed and longer experience than a coffee shop and or quick service restaurant. The trick for the operator is to have enough staff to provide the right or optimal level of service for guests without an out-of-line labor cost. The efficiency and effectiveness of employees can also be measured and fed into an operator’s data.
AI Benchmarks
Additionally, some lodging chains have developed their own AI applied systems for guest service and more. For example, Hilton’s Connie, powered by IBM Watson’s AI, acts as a concierge, assisting guests with information about hotel amenities, dining recommendations, and local attractions. AccorHotels uses AI to analyze guest preferences and booking history to send personalized offers and recommendations, leading to increased guest engagement and loyalty. Also, Marriott International has experimented with AI-powered assistants in rooms that allow guests to control room settings, including lighting, temperature, and entertainment systems, through voice commands. This not only adds convenience but also provides a tailored experience to each guest based on their preferences.
Another channel to identify AI benchmarks and best practices in the lodging industry can be to utilize AI itself as the identifier of innovation and application of AI in hotels. As is noted by many experts, AI responses are a function of its data content, and the questions asked. These factors can limit AI effectiveness. It is suggested that direct and logical questions are the best route to gain useful information. An introduction to ChatGPT maybe helpful. Your leadership team may be a great place to start to brainstorm what are the questions a lodging operation needs answers to? Your staff may also be a great place to get input for questions and you benefit from a different perspective.
In response to a question asking how lodging chains are leveraging AI to enhance their businesses, ChatGPT provided the following: Top lodging chains are leveraging AI in several innovative ways to enhance guest experiences, optimize operations, and improve revenue management. Here are some key applications:
- Personalized Guest Experiences: AI analyzes guest data to offer tailored recommendations for services, activities, and room preferences. (Notes: I reiterate that a balance is needed to provide optimal guest service. Technological efficiency and labor savings are great but not at the expense of guest service. A deep dive into target market segments and their demographics and preferences is in order.)
- Dynamic Pricing: AI algorithms help in adjusting room rates in real-time based on demand, competitor pricing, and market trends. (Notes: clearly AI can perform this task more quickly, but a reminder that the decisions on pricing are made by people. The machines are providing tools for decision-making).
- Chatbots and Virtual Assistants: These bots can handle inquiries, make reservations, and provide information 24/7, improving efficiency and guest satisfaction. (Notes: Again, I refer to market segments and knowing your guests. I hate talking to machines.)
- Operational Efficiency: AI is used for predictive maintenance, helping to identify when equipment is likely to fail, thus reducing downtime. (Notes: This ties AI to sustainability and efficiency of operation. AI is perhaps a tool to help sustainability become truly mainstream in decision-making processes.)
- Staffing Optimization: AI tools analyze booking patterns to optimize staffing levels, ensuring the right number of staff are available during peak times without overstaffing during quieter periods. (Notes: As previously discussed, AI analysis of booking patterns can optimize staffing levels to provide the optimal level of service needed for a hotel.)
- Fraud Detection: AI systems monitor transactions for unusual patterns, helping to identify and prevent fraudulent activities. (Notes: AI can provide additional support for fraud detection and prevention).
- Sentiment Analysis: AI analyzes online reviews and social media feedback to gauge guest sentiment. (Notes: this is a great tool, but managers and owners will need to input the needed parameters to help interpret guest feedback. Maybe the guest is not always right, but an appreciation of their impressions of receiving good value for the price charged is great information.)
- Enhanced Booking Platforms: AI-powered booking engines can provide a smoother and more intuitive user experience. (Notes: I can only imagine that booking engines will get better and make the experience easier and smoother than ever. The booking online platforms on a computer or using mobile apps are not always easy to navigate.)
The ChatGPT response is fairly comprehensive and is responsive to the question asked. I think key to the responses are the actions and operationalized terms noted in the response. These terms include things like analyzes, adjusting, dynamic, efficiency, predictive, booking patterns and more. As a follow up, my next question was What are AI benchmark “best practices” at the top hotel brands? The ChatGPT response was: Top hotel brands implement several AI benchmark best practices to enhance their operations and guest experiences. Here are some key practices, highlighted and in italics for presumed actions and results expected.
- Data-Driven Decision Making: Establish robust data collection and management systems to ensure accurate, comprehensive data is available for AI algorithms.
- Personalization Strategies: Utilize AI to analyze guest data and tailor experiences, from personalized marketing to customized room settings.
- Integrating AI with Existing Systems: Ensure that AI tools seamlessly integrate with existing property management systems (PMS), customer relationship management (CRM) systems, and other operational technologies for smooth functionality.
- Real-Time Analytics: Implement AI solutions that provide real-time data insights, allowing hotel managers to respond quickly to changing market conditions, guest needs, and operational challenges.
- Cross-Department Collaboration: Foster collaboration between departments (marketing, operations, customer service) to share insights gained from AI, ensuring that all teams are aligned in their approach to enhancing guest experiences.
- Continuous Learning and Adaptation: Regularly update AI models based on new data and changing market dynamics. Result: This ensures that AI systems remain effective and relevant.
- Customer Feedback Integration: Use AI to analyze guest feedback and reviews continuously, identifying areas for improvement and adapting services accordingly.
- Training and Development: Invest in training staff to work effectively with AI tools, ensuring they understand how to leverage AI insights in their daily operations and interactions with guests.
- Ethical AI Practices: Prioritize transparency and ethical considerations in AI deployment, particularly regarding data privacy and security, to build trust with guests.
- Scalability: Choose AI solutions that can scale with the organization’s growth, ensuring they can handle increased data and complexity as the brand expands.
The response to the second question has similar but not exact responses and, in some cases, provides more detail for consideration. AI responses (ChatGPT) will differ based on how the questions are posed. What you will find is that the responses may indicate the type of result a practice may provide. For example, under Ethical AI Practices, the result is indicated as “to build trust with guests.” I recommend hotels and their owners and management teams focus on education for use of AI and better understanding of use and application and how it interacts with guests.
Education and Training for AI Optimization
AI and Education are like services in some ways. Do guests and or students want a machine to decide on a grade in a class and or provide interaction with a guest for a possible purchase? One way to examine AI preparedness from an educational perspective may be to view it through the lens of Bloom’s Taxonomy of Educational Objectives inclusive of knowledge, comprehension, application, analysis, synthesis and evaluation.
These original objectives are an early (1956) examination of what education and or training want to get accomplished for students and or employees and, in this case, for education and training in AI. A lot has happened since 1956, but this framework can support decisions on content, methods of delivery, and use of technology. For AI and, more specifically, ChatGPT, the following notes may be helpful.
- Knowledge: What is AI? How can it be used in a classroom and in a lodging operation? A student and or employee should be able to recall or define AI from at least their perspective.
- Comprehension: Moving on from definitions, can an employee and or student explain AI to someone? Note, I have used ChatGPT in my courses and asked students to use AI to help create an explanation that they could share in a training setting.
- Application: Now that a student can define it and explain it, can they apply it? Or simply get on the site and ask it the optimal questions. As a faculty member, the objective of the assignment is to operationalize the use of ChatGPT for a specific topic in a class or training session.
- Analysis: In this stage of learning, maybe the best example would be to use a ChatGPT-generated response for a question a faculty member created and ask the students or employees to analyze the responses. In this stage it is a compare and contrast stage where the employees are looking for key responses, and assessing applicability etc.
- Synthesis: At this stage, the employees and students need to “create” an AI-generated response. This could involve a series of testing questions and assessment of those responses for veracity and accuracy. Thus, creating a final product for the task assigned using AI tools.
- Evaluation: Finally, a judgment about the results is in order. In a classroom, a faculty member may need to assign a grade for the work completed. Or the final products submitted could be evaluated by other students, and the submitter would need to justify their results.
These objectives can help layout the framework for AI training and education. As noted previously, AI, whether predictive or generative, will be successful based on how it is used, data input and the questions asked. AI can be a valuable tool, or as we are hearing in the news, in some cases tactic or weapon to create “false” facts. For example, AI-generated videos have been cited in multiple news and political campaigns. On the education side of the issue, most universities have created their own AI standards and policies. A review of these is available at the preceding link. The standard operating procedure on AI is shared campus-wide at our institution, Office of Faculty Excellence on Generative AI, and also includes some other resources provided at the University level regarding AI.
Going Forward
In the future, AI is expected to be integrated with virtual reality and augmented reality. The result will be guests getting immersive experiences. Tours of location, resorts, hotel rooms and facilities such as virtual tours, virtual room selection, and virtual events. It does feel a little like an episode of The Jetsons; I loved that show as a kid!
Additional AI offerings are and could be : predictive room service. AI-powered sustainability initiatives, enhanced security measures, predictive personalization, AI-driven inventory management, automated staff scheduling, predictive maintenance, green technology integration and guest security and safety. Also, given the diversity of our industry and its labor force, AI can also prove a useful tool for employees, which is speech AI. AI support includes streamlined internal communication, automating administrative tasks, real-time support and assistance, and multilingual support.
Many brands and larger hotel operations are already immersed in AI to support and enhance their products. However, there are some smaller operations and independents that need to get started. The choices in the market include both free and paid solutions. Most importantly lodging properties should have an AI strategy. As noted earlier in this discussion, those that do not engage in AI may be left behind. For example:
- Assess where you as a hotel are in terms of technology, innovation, and potential AI application.
- What are your needs? What efficiencies are you looking to get accomplished?
- Are you assessing labor costs? Is AI application expected to help cut labor costs?
- Are your guest satisfaction scores good or in need of help?
- Know your market segments and match AI actions and technologies with guest needs.
AI is not a panacea. It has to be tailored made to be really beneficial. It has to be empowered with the right knowledge, and ultimately, it must be smart enough to garner valuable data throughout the guest journey. The AI has to be fed good knowledge. It has to be monitored. It could do more harm than good if not well-executed. It takes resources to enact. There is a cost. Hospitality has suffered through the pandemic and afterward in workforce shortages. This tool can alleviate that problem, but it does not replace a human.
I will make one note concerning AI and labor. There continues to be a lot of talk about the lack of labor available and or that people do not want to work. I disagree with that premise. COVID-19 was the catalyst for a new way of thinking about jobs and work and a more focused analysis of the jobs people had. During the pandemic, many hotels and restaurants closed; they did not adapt nor figure out new ways to make a business work. Due to hospitality business closures, other non-hospitality service businesses recruited our employees. They valued their skills, knowledge and experiences and also found that many of them adapted extremely well to new work environments. Our alumni that were furloughed and or let go, have found life after traditional hospitality positions.
My recommendation is that hospitality needs to adapt, but it also needs to make the jobs better than they were. Enhanced work environments, pay scale, opportunities for advancement and more. The employees are out there, and people do want to work, but they have many more options than they used to have. I have been around a long time, and I remember that not all hospitality jobs were good ones. We can fix that, enhance our jobs, create new positions for a new era of innovation and technology that values our employees. AI can help us do that as well!


