A New Competency of the Modern Manager

Prompt Engineering for Artificial Intelligence in Hospitality: A New Competency of the Modern Manager ()

1. Introduction: Prompt Engineering as a New Managerial Competency

Prompt engineering, defined as the structured formulation of precise and contextualised instructions for artificial intelligence systems [1], is increasingly recognised as a core competency of contemporary hotel management [2]. Unlike traditional approaches that treated AI primarily as a technical tool, this skill emphasises the human factor: managers translate business objectives into machine-readable commands by integrating analytical thinking, contextual awareness, and digital literacy [3] [4]. Grounded in cognitive linguistics and human-computer interaction (HCI), this competency enables managers to leverage AI for service personalisation, data analysis, and operational optimisation, while also requiring continuous education [5].

According to Pan and Fu (2024) [3], European hotel managers utilise AI for tasks such as review analysis and dynamic pricing, where the quality of the prompts is crucial to success. Empirical findings suggest that AI integration increases guest satisfaction by 23% – 35% [6], yet limited empirical validation, primarily derived from secondary sources, underscores the need for further research.

This competency does not replace the human factor but rather reinforces it, positioning the manager as a strategic intermediary. For example, a generic prompt such as “Write a response to a complaint” results in a bland and impersonal output. In contrast, a targeted prompt like “Create a response to noise complaints at the pool after 10 PM in three paragraphs: apology, compensation (dinner at the restaurant), action plan. Tone: empathetic” produces a specific and actionable solution [4].

This article contributes to the literature by framing prompt engineering as a key managerial competency with a focus on the human factor. In contrast to earlier studies, it emphasises the role of prompt engineering in strategic management.

Research question:

How does prompt engineering, as a new managerial competency, affect operational efficiency, service quality, and strategic agility in hotel operations?

Structure of the paper (roadmap):

The article consists of six main sections: 1) theoretical foundations and definition of prompt engineering, 2) presentation and analysis of the five-stage cycle, 3) review of empirical benefits and limitations, 4) critical examination of risks, 5) education and competency development, and 6) conclusions with recommendations for future research.

2. What Is Prompt Engineering and Why Is It Important?

Prompt engineering is a systematic and interdisciplinary discipline that enables hotel managers to translate business goals, challenges, and operational needs into precise, structured, and contextually relevant instructions for artificial intelligence systems [1]. This skill goes beyond generic digitalisation and requires a deep understanding of the hospitality context, analytical reasoning, and clear digital communication [3]. Rooted in the principles of cognitive linguistics and human–computer interaction (HCI), the discipline emphasises linguistic precision, reducing algorithmic bias and ensuring ethical, transparent, and fair application of AI systems [4] [7]. Unlike vague or overly broad instructions, a carefully designed and well-structured prompt allows hotel managers to control the content, tone, and effect of AI outputs. This competency directly enhances communication personalisation, automates key processes, and enables strategic use of AI tools in daily operations. While earlier studies were limited to secondary sources, recent empirical work confirms tangible benefits in hospitality contexts, though longitudinal primary research is still needed to accurately map long-term organisational effects [6] [8]-[10]. A particularly relevant recent study [11] involving a sample of 212 hotel managers in Spain confirms that digital innovation, including advanced AI techniques and prompt engineering, positively correlates with sustainable competitive advantage (β = 0.767), through strategic differentiation, flexibility, and agility.

2.1. Application Examples in Hotel Management

1) Automated Guest Sentiment Analysis:

AI tools analyse hundreds of reviews and detect key themes and emotional tones significantly faster and more reliably than manual processing [12].

Example prompt: “Analyse reviews for Hotel X from last month and list the three most frequently mentioned strengths and problems.”

2) Generating Personalised Offers and Campaigns:

Based on booking history and preferences, AI creates highly personalised marketing content. Empirical data confirm that the implementation of artificial intelligence in hotel processes and dynamic pricing can lead to an increase in guest satisfaction in the range of 23% to 35% [6], as well as greater guest willingness to pay for personalized AI services [9] [10]. Market and industry reports indicate that hotels using advanced AI-based revenue management systems see increases in total revenue and RevPAR, but such rates (e.g., 7% – 17% for revenue and up to 10% for RevPAR) are based on industry analyses and have not been confirmed by formal scientific studies [13].

Example prompt: “Create a promotional email for guests from Germany, Berlin region, who previously used wellness services. Highlight seasonal discounts and include a call to action.”

3) Creating SOPs:

AI accelerates the creation of Standard Operating Procedures (SOPs), increases consistency, and reduces human error [14].

Example prompt: “Generate an SOP for the kitchen department with clear instructions for handling allergens.”

4) Complaint Resolution with Personalisation:

A structured prompt enables clear, branded responses with minimal editing, enhancing professionalism and response speed [1].

Example prompt: “Write a response to a complaint about pool noise after 10 PM. Begin with an apology, offer dinner for two as compensation, and announce a preventive measure. Tone: empathetic.”

5) Reputation Revitalisation and Efficiency:

Recent industry reports [15] indicate a global increase in positive guest review indices and accelerated response times in hotels investing in AI communication technologies, though precise impact figures attributable solely to prompt engineering are not yet available in scientific literature.

Prompt engineering has become a strategic competency for modern hotel managers, enabling nuanced, ethical, and data-driven intelligence in all aspects of service delivery [1] [3] [6] [8] [10] [11]. It delivers gains in efficiency, personalisation, and strategic capability, with its greatest innovation lying in the human-centred design of AI-driven hospitality. Nevertheless, for specific performance claims, only peer-reviewed and verifiable data should be credited; industry benchmarks must be clearly distinguished from academic research in any scientific discussion.

2.2. Output Parameter Control

One of the greatest advantages of prompt engineering is the ability to fully control AI output parameters, including tone, format, and content structure, which is also confirmed by recent research in hotel management. Controlling prompt parameters is directly crucial for reducing errors and increasing communication efficiency, enabling managers to align AI responses precisely with brand standards and operational guidelines [1] [9]. This achieves consistency in the customer experience and reduces operational costs related to editing automated texts.

Based on the most relevant scientific and industry sources from 2023-2025, the following Table 1 summarises the key benefits of applying prompt engineering in the hotel sector, with clear attribution of each piece of data:

Table 1. Key benefits of prompt engineering in the hospitality industry.

Dimension	Applied impact	Empirical evidence
Operational agility	Reduction of task duration (review analysis, SOP creation) from days to minutes	[16]
Personalisation and satisfaction	Increase in guest satisfaction by 23% – 35% after the introduction of AI solutions; higher willingness to pay for personalised services	[6] [9] [10]
Reputation and loyalty	Global benchmarking indicates a gradual improvement of GRI with digitalisation and faster responses; no direct quantification for prompt engineering	[15]
Data-driven decision-making	Optimisation of pricing, staff schedules, and CRM through AI recommendations	[14] [17]
Competitive advantage	Positive correlation of innovation and sustainable advantage (β = 0.767) through offer differentiation	[11]

Source: Author’s analysis based on scientific and industry literature (2023-2025).

Prompt engineering in modern hotel management is recognised as a strategic competency that enables a shift from passive automation to proactive design of the digital guest experience.

Carefully structured prompts increase operational efficiency, the level of personalisation, and the quality of data-driven management. Although the benefits of this skill are receiving increasing empirical support in the scientific literature [6] [8] [11], further research is still needed, especially in the area of risk management, guest loyalty development, and integration of new digital competencies into existing organisational cultures [2] [9] [10]. To fully operationalise this competence, it is essential to understand its development process through a structured prompt design model, which is the subject of the next chapter.

3. Prompt Engineering as a New Managerial Competency

Prompt engineering is recognised as a key managerial competency in hospitality, as it enables effective communication with advanced AI systems. This skill is not isolated but a dynamic and cyclical process that connects analytical thinking, hospitality experience, and digital literacy, grounded in theories of cognitive linguistics, human-computer interaction (HCI), and dynamic capabilities [3] [4] [11]. Empirical findings show that an iterative approach improves guest satisfaction by up to 35% through personalised communication [6], but limited primary research calls for further studies [11].

To better understand this process, we highlight five interrelated phases that form an iterative cycle, from defining context to integration into business workflows (See Table 2).

Table 2. Five-phase iterative cycle of prompt engineering.

Phase	Description	Practical example in hospitality	Required skills	Empirical evidence
Contextualization	Defining the business framework for precise prompt formulation.	A hotel in Dubrovnik analyzes pool noise complaints: “As the manager of Hotel ‘Jug’, a 4-star hotel in central Dubrovnik, analyze reviews from June 1 to July 1, 2024, focusing on pool noise between 10 PM and 6 AM. Identify causes (e.g., lack of supervision) and propose solutions.”	Analytical thinking, context understanding [1].	Without context, AI provides templated responses, reducing guest trust [3].
Goal definition	Precisely specifying output, format, and parameters.	Generating a complaint response: “Create a response in 3 paragraphs: apology, compensation (free lunch), action plan. Tone: empathetic.”	Strategic planning, clear communication [3].	Prevention over reaction—AI identifies root causes, reducing complaints by 40% [19].
Iterative refinement	Refining the prompt through testing and modifications.	From generic “Respond to a complaint” to detailed, with timelines and icons (see example in text).	Critical thinking, iterative approach [7].	Each iteration reduces ambiguity, increasing precision [7].
Evaluation	Checking output quality quantitatively and qualitatively.	Checking cultural adaptation: the use of a formal tone for Japanese guests.	Evaluation skills, ethical awareness [4].	Combined methods ensure a balance of precision and emotional intelligence [6].
Integration	Embedding AI outputs into business processes (CRM, PMS).	Automatic sending of complaint responses via CRM, reducing resolution time to hours [14].	Integration skills, change management [6].	Integration reduces complaint resolution time from days to hours [14].

Source: Author’s analysis based on secondary sources (2023-2025).

3.1. Contextualization: Setting the Scenario

Everything begins with clearly defining the business framework.

Example: A hotel in Dubrovnik notices an increase in complaints about pool noise at night.

Instead of a generic prompt “Write a response to a complaint,” the manager specifies: “As the manager of Hotel ‘Jug’, a 4-star hotel in central Dubrovnik, analyse reviews from June 1 to July 1, 2024, focusing on complaints about pool noise between 10 PM and 6 AM. Identify causes (e.g., lack of supervision, guest groups) and propose solutions.”

Why is it important? Without context, AI provides templated responses (e.g., “We will try to resolve the issue”), lacking operational plans or timelines, which reduces guest trust. Managers who apply a contextualization approach to AI strategically not only solve problems but also anticipate and prevent them.

3.2. Goal Definition: From Generic to Targeted

After establishing context, the manager precisely defines the output format and parameters. Example prompt: “Generate a response in 3 paragraphs:

• Apology: Express regret for the noise and acknowledge the impact on guests.

• Compensation: Offer a free lunch for two at our award-winning restaurant.

• Action Plan: Announce the introduction of a poolside security guard from 10 PM.

• Tone: professional but empathetic. Use icons to emphasise key points.”

Resulting AI-generated guest response (clear, targeted, and personalised):

“Dear Guest,

We sincerely apologize for the noise that disrupted your peace during the night hours… As a sign of our commitment to your comfortable stay, we are offering a free lunch for two at our award-winning restaurant… Starting tomorrow:

Poolside security guard from 10 PM to 6 AM

Thank you for your trust and patience.”

This approach connects context and goals:

• Apology: Directly addresses complaints identified in the review analysis.

• Compensation: (free lunch) is aligned with the hotel’s facilities, enhancing perceived value.

• Action Plan: Resolves specific causes (e.g., lack of supervision).

Why does this approach work?

1) Prevention over reaction: AI not only responds to complaints but identifies root causes and proposes systemic solutions (e.g., adding a guard). According to HospitalityNet [18], hotels implementing preventive measures reduce complaints by 40%.

2) Personalisation and brand consistency: A prompt that includes hotel-specific details (e.g., “award-winning restaurant”) increases authenticity.

3) Operational efficiency: A structured format (3 paragraphs + icons) simplifies CRM integration and accelerates guest communication.

3.3. Iterative Refinement: From Idea to Optimisation

Iterative refinement is the process of shaping a prompt through successive cycles of evaluation and modification to achieve the highest output quality, as no prompt should be perfect on the first attempt. This phase requires continuous testing, analysis, and prompt revision to ensure AI responses align with specific hotel needs. The goal is to transform generic outputs into precise, result-oriented solutions.

Practical example: Reducing poolside noise at Hotel “Jug”

1) First iteration

• Prompt: “Generate a response to a complaint about pool noise.”

• AI response: “Dear Guest, we are sorry for the inconvenience. We will try to resolve the issue.”

• Problem: The response is too generic, lacking concrete actions, timelines, or compensation.

2) Prompt revision

• New version: “Revise the response focusing on technical measures: pool supervision after 10 PM, installing a barrier, in-app notifications.”

• AI response: “We are introducing a poolside security guard from 10 PM and a noise-reducing barrier.”

• Problem: Lacks guest compensation and detailed implementation timelines.

3) Final version

a) Apology: Express regret for the noise and acknowledge guest impact.

b) Compensation: Offer a complimentary lunch for two in the restaurant.

c) Technical measures: Announce pool supervision from 10 PM, barrier installation, and in-app notifications.

4) Add implementation deadlines and use icons to emphasize key actions.”

“Dear Guest, we apologize for the noise… As compensation, we offer a complimentary lunch for two in our award-winning restaurant…

🔹 Starting tomorrow: Security guard by the pool after 10 PM

🔹 By Friday: Installation of a sound barrier

🔹 Notifications: Push alerts in our app about quiet hours”

Why does the iterative approach work?

• Greater precision: Each iteration reduces ambiguity (e.g., adding deadlines like “by Friday” avoids vague promises).

• Increased accuracy: According to Bhandarkar et al. [7], iterative refinement improves response accuracy by removing vagueness and adding context that enhances relevance and tone alignment.

• Clarity through specificity (“supervision” “security guard by the pool after 10 PM”).

• Contextual enhancement (e.g., “sound barrier”).

• Stronger guest trust: Detailed plans and transparency lead to higher guest satisfaction.

Iterative refinement is not just a technical adjustment; it is a continuous improvement strategy that requires critical thinking and hospitality expertise. Hotel Managers who master this skill transform generic AI responses into tools for building trust and achieving operational excellence.

3.4. Evaluation of Results: Measuring Quality

Evaluating generated AI responses is a critical phase to ensure outputs meet hotel standards and guest expectations. Quantitative and qualitative methods are used for objective and subjective analysis. By combining approaches, managers ensure AI becomes an increasingly precise tool for personalisation and efficiency.

Example: Cultural adaptation as part of evaluation

Check whether the response is culturally appropriate for the guest’s background.

For instance:

• Croatian guests: Prefer a direct, problem-solving approach (“We are adding a guard tomorrow”).

• Japanese guests: Expect a more indirect, highly formal tone (“We believe that increased supervision may help”).

Example adaptation for Japanese guests:

“Dear Guest, we sincerely apologise for not meeting your expectations. As a measure of improvement, we will assign an additional supervisor to the pool area starting tomorrow.”

Evaluation is not just a formality: it is the bridge between generated content and real impact on hotel operations. Through combined quantitative and qualitative assessments and by applying cultural and contextual testing, managers ensure a balance between technical accuracy and emotional intelligence, key to building long-term guest trust [6].

3.5. Integration into Business Processes: From Idea to
Implementation

Once a prompt is crafted, tested, and evaluated, the key step is integration into the hotel’s daily operations. Only then do AI outputs become a true support system for hotel management and staff, turning artificial intelligence from theory into practice.

Examples of integration:

• CRM system: When AI generates a personalised response to a guest complaint, it is automatically sent through the CRM system, ensuring fast, consistent, and personalised communication.

Example: A guest who complained about noise receives a response including an apology, compensation, and explanation of actions taken.

• PMS system: Operational outputs, such as pool supervision plans, are automatically integrated into staff schedules and task lists.

Example: Once AI suggests adding a poolside guard after 10 PM, the PMS assigns staff and sends reminders.

Example: A new SOP for managing nighttime noise becomes immediately available to reception and security teams.

According to Acropolium [14], such integration reduces complaint resolution time from an average of 3 days to just 3 hours, while also minimising human error and increasing communication consistency.

The five phases of prompt engineering are not just technical steps. They are a comprehensive managerial tool for transforming hotel operations. Through cyclical and iterative application, managers build a bridge between human expertise and algorithmic capabilities, ensuring competitiveness, agility, and excellence in the era of digital transformation. This level of integration distinguishes hotels that use AI superficially from those that leverage it as a strategic driver of growth and innovation [6].

4. A Critical Review of the Limitations and Risks of AI Implementation and Prompt Engineering in Hospitality

Although artificial intelligence and prompt engineering offer evident benefits to hotel operations with automation, faster data processing, personalised communication, and optimisation, they also present clear limitations and risks, as highlighted in academic and industry literature and editorial guidelines [3] [8] [19].

4.1. Loss of Personal Contact and Emotional Distance

One of the main challenges is the reduction of direct human interaction in delivering guest services. Automated chatbots and standardised AI responses may result in a perception of mechanical, “cold” communication and a lack of empathy, especially in situations that require emotional intelligence. Empirical studies and industry reports confirm that guests still value warmth and authenticity in human communication [8] [20].

4.2. Overreliance on Technology and Lack of Flexibility

AI tools are limited by programmed rules and struggle to respond to complex, unexpected, or creative guest requests. In negotiations, emergencies, or when developing innovative solutions, the human factor remains indispensable [3] [21].

4.3. Privacy and Data Security Risks

AI-based systems require the collection and processing of large amounts of sensitive guest data, which raises questions about privacy and data security. Any breach or misuse of data directly impacts the hotel’s reputation, customer trust and can have legal consequences [9] [22]. While protection standards exist, practice in the sector is highly variable.

4.4. Technical and Integration Difficulties

AI implementation requires significant financial investment, advanced technical knowledge, and compatibility with all existing hotel systems (PMS, CRM). Poorly integrated AI tools lead to data fragmentation, hinder performance monitoring, and reduce process efficiency [14] [23].

4.5. Risk of Bias and Discrimination

AI learns from historical data, which may contain biases or discriminatory patterns. Without ongoing ethical oversight and algorithmic evaluation, such inequalities can be transferred and amplified through automatic decision-making [4] [8].

4.6. Financial and Operational Challenges

Initial investment, employee training, and continued AI system maintenance can be especially challenging for independent and smaller hotels. Additional hurdles include the risk of “lock-in” to one supplier [14] [23].

4.7. Guest Trust and Acceptance

Some guests remain distrustful of AI, especially regarding privacy, personalisation, and the authenticity of communication. Transparent communication about how and why AI is being used, s well as clear boundaries between automated and human actions, are crucial for building and maintaining trust [21] [24].

Final Analysis

The limitations and risks of AI technology and prompt engineering in hospitality must be addressed through:

• systematic employee education [19],

• regular ethical oversight of algorithms and collected data [4] [8],

• flexible implementation strategies combining AI and human interaction for specific guest needs,

• proactive, open communication with guests regarding the use of artificial intelligence.

Only through this approach can AI become a sustainable and responsible tool for enhancing hotel performance, while minimising risks to guest satisfaction and brand reputation, in alignment with the strictest editorial and scholarly guidelines.

5. Developing a New Competency and Education in the Daily Work of a Hotel Manager

The integration of digital tools and the development of new competencies have become an essential part of the daily responsibilities of hotel managers, critical for maintaining competitiveness and excellence in today’s hospitality landscape [25]. Prompt engineering and working with AI are transforming the traditional managerial role: from operational oversight to that of a strategic innovator who shapes guest experience, optimises internal processes, and makes data-driven decisions.

To successfully navigate this transition, managers must continuously acquire new knowledge and remain open to change. This enables a more creative approach to service personalisation and faster adaptation to evolving guest needs [26]. Managers who consistently develop their digital skills and understand AI capabilities are not merely following industry trends but actively shaping them [27].

Key tools for development include:

• Specialised training and workshops: Managers can attend courses, micro-qualifications, and practical workshops focused on prompt engineering, often through online platforms or professional consultants [25]

• Mentorship and benchmarking: Exchanging experiences and comparing best practices across the industry accelerates learning and application of new skills [28].

• Practical frameworks (COAST, TRACE): Structured formats like COAST (Context, Objective, Actions, Scenario, Task) and TRACE (Task, Request, Actions, Context, Example) help managers formulate clear and effective prompts [26] [29].

• Creating an internal “prompt bank”: Sharing successful prompt examples within the team facilitates implementation and the continuous improvement of AI communication skills [27].

Developing these competencies is not a one-time task but requires ongoing education, evaluation, and adaptation to emerging market needs. Only through this synergy between human and digital competencies can hotel operations unlock the full potential of AI and view prompt engineering not as a passing trend, but as a foundation for sustainable competitiveness and excellence.

6. Conclusions

This paper conceptualises prompt engineering as a new strategic competency in hotel management, uniquely connecting analytical thinking, operational experience, and algorithmic precision. By shifting the focus from mere technical support to a managerial function, prompt engineering enables hotel managers to translate complex business objectives and contextual knowledge into effective AI-driven processes. The five-phase iterative model: contextualisation, goal specification, refinement, evaluation, and integration, demonstrates how AI can be systematically embedded into key service areas, including guest communication and operational standardisation.

It is important to emphasise that the effectiveness of prompt engineering does not diminish the need for deep professional knowledge, practical experience, or familiarity with the specific characteristics of one’s property and market environment. The success of this skill depends on the manager’s ability to analyse guest expectations, local dynamics, and business specificities, as well as on a continuous openness to learning and professional development. The critical evaluation of AI outputs, their alignment with the brand and guest profile, and the ongoing process of evaluation and improvement remain within the domain of the manager’s experience and expertise.

Although preliminary results indicate an increase in guest satisfaction, greater efficiency, and enhanced strategic differentiation, these benefits are not automatic. They require a combination of digital skills and interpersonal capabilities, an ethical approach, and a commitment to lifelong learning. Risks such as privacy violations, loss of emotional authenticity, or algorithmic bias further underline the need for strong managerial oversight, transparent communication, and a shared adaptation culture among staff.

Prompt engineering should be viewed as a catalyst for transforming the managerial role, managers are not passive users of AI technology, but become architects of digital processes within the hotel. Those who develop both practical sector knowledge and digital literacy can use prompt engineering to enhance the guest experience in a sustainable, context-sensitive, and responsible way, ensuring that technological progress serves the core values of hospitality rather than undermining them.

Conflicts of Interest

The author declares no conflicts of interest.

Originally Appeared Here