Restaurant revenue management: a systematic literature review and future challenges Open Access

The restaurant industry is pivotal in the global economy, accounting for approximately 3–4% of the global Gross Domestic Product (GDP) in 2023, being one of the sectors that has achieved the fastest recovery and with very promising projections for the coming years, as reported by the United Nations World Tourism Organisation UNWTO, 2024 However, the COVID-19 pandemic significantly disrupted this sector, compelling businesses to reassess and transform their operations. The persistence of innovations introduced during COVID-19, such as digital menus and delivery services, underscores the sector's transition towards digitalisation. Beyond digitalisation, the health crisis generated pent-up demand from prolonged restrictions and closures, leading to a rapid recovery in restaurant activity once restrictions were lifted (Pascual-Fraile et al., 2024). This phenomenon, combined with the “carpe diem” effect, resulted in increased spending on dining experiences, driven by a mindset of immediate enjoyment (Hostelería de España, 2023). These shifts in consumer behaviour have influenced the evolution of research on demand management and Revenue Management (RM) strategies in restaurants.

RM, a well-established management philosophy in service industries, focuses on optimising profits and customer value through the effective analysis of information and management of price, capacity, and sales (Talón-Ballestero et al., 2023). It involves the strategic allocation of capacity and distribution channels, ensuring the right prices and timings for targeted consumers (Kimes, 2011a, b). This approach is particularly beneficial in industries characterised by fixed capacities, perishable inventory, variable demand, segmented markets, advanced reservations, high fixed costs, and low variable costs (Kimes et al., 1999). Despite its proven efficacy in sectors such as airlines and hotels, the adoption of RM strategies in the restaurant industry has remained relatively limited (Tyagi and Bolia, 2021).

Restaurants face unique operational challenges stemming from their relatively fixed capacities and the unpredictable duration of service (Desiraju and Shugan, 1999; Heo and Lee, 2010). These characteristics necessitate the development of innovative RM strategies. Research in Restaurant Revenue Management (RRM) has underscored a growing reliance on data analytics, driven by the availability of big data and the emergence of digital tools (Cohen, 2018). Moreover, advancements in forecasting methods and machine learning techniques have enabled the optimisation of key processes (Mišić and Perakis, 2020). The widespread adoption of digital menus, initially accelerated by COVID-19, has further transformed restaurant operations, facilitating the implementation of dynamic pricing, menu engineering, and behaviour-based pricing strategies.

Despite the advancements in RRM, the literature remains underdeveloped, with Tyagi and Bolia’s (2021) review representing, to the authors’ knowledge, the only comprehensive synthesis to date. However, their study only includes articles published up to 2020. Therefore, it does not analyse the changes in RRM practices that have emerged during the post-COVID-19 period. The current study addresses this gap by employing a systematic literature review (SLR) methodology, noted for its unbiased, auditable, and repeatable nature (Fink, 2005; Petticrew and Roberts, 2005; Kitchenham, 2004). By compiling, evaluating, and synthesising existing RRM research, this study identifies critical gaps and offers a robust foundation for decision-making. Consistent with the traditional structure of RRM and incorporating the categorisation proposed by Talón-Ballestero et al. (2023), RRM literature is examined across five strategic processes, providing a comprehensive perspective on the opportunities and challenges within this evolving field.

Traditional RRM is centred on three primary strategic levers: capacity, meal duration, and price management (Kimes, 1999). However, Talón-Ballestero et al. (2023) introduced two additional levers (information and sales) that are now indispensable in contemporary RRM. These five strategic levers form a robust framework, blending traditional practices with innovative approaches to optimise resources effectively (see Figure 1).

The foundational levers of capacity, meal duration, and price management continue to play a pivotal role in RRM (Kimes, 1999; Tyagi and Bolia, 2021). At the same time, the increasing reliance on digitalisation and data analytics underscores the importance of the information lever (Choi et al., 2022; Talón-Ballestero et al., 2023). Advances in big data and machine learning have transformed data management into a critical component of profitability, enabling precise forecasting and data-driven decision-making (Gómez-Talal et al., 2024). The sales lever, on the other hand, emphasises the impact of customer perception on dish selection and revenue generation (Shoemaker, 2003).

“Capacity management” is vital when demand exceeds a restaurant's limited capacity, requiring space optimisation. Lai et al. (2020) note that capacity also depends on kitchen and room service efficiency, highlighting the need to streamline processes and reduce waiting times. Addressing cancellations and no-shows is crucial, with overbooking policies (common in other RM industries) offering a promising yet underused strategy in the restaurant sector (Tyagi and Bolia, 2021).

The strategic management of “meal duration management”, identified by Tyagi and Bolia (2021) as a core RRM lever, builds on Kimes’ (1999) framework. Unlike other sectors where service durations are fixed, restaurant service times are heavily influenced by customer behaviour unless structured interventions, such as designated shifts, are implemented. Controlling service duration is therefore a vital component of operational efficiency.

“Price management”, another traditional RRM lever, encompasses demand-driven pricing strategies, price fences, and menu engineering. These strategies consider variables such as cost, profit margins, demand elasticity, competition, and reputation (Tyagi and Bolia, 2021). While menu engineering has historically been grouped under price management, this study categorises it as part of the sales lever, focusing on its role in influencing customer choices.

The “information management” lever represents the growing importance of data collection, analysis, and decision-making in RRM. Despite technological barriers and reliance on intuition (Tyagi and Bolia, 2021), recent advancements in machine learning offer promising opportunities for forecasting accuracy (Choi et al., 2022). Effective information management requires continuous evaluation of RRM strategies, incorporating historical data, current trends, competitive analysis, customer segmentation and distribution channels to optimise outcomes (Talón-Ballestero et al., 2023).

Incorporating “sales management” as a strategic lever emphasises selecting the most profitable distribution channels and managing costs effectively (Talón-Ballestero et al., 2023). Shoemaker (2003) highlights the role of customer perception in sales, with the menu as a key tool. Customers’ perceived value of a dish strongly influences their purchasing decisions, and digital menus now allow real-time adjustments to enhance the appeal of menu items. These menus integrate audiovisual elements and facilitate dynamic updates reflecting product availability and demand trends. Techniques such as menu engineering and behavioural pricing enable restaurant managers to optimise how dishes are presented and priced to maximise revenue (Kasavana and Smith, 1982; Lai et al., 2020). While traditionally linked to price management (Tyagi and Bolia, 2021), menu engineering focuses on analysing item popularity and profitability to optimise design, pricing, and strategies (Kasavana and Smith, 1982). By shifting menu engineering under sales optimisation, this study emphasises its role in influencing customer choices and boosting revenue. Menu engineering ultimately enhances sales and financial performance. The sales lever also addresses the complexities of managing new distribution channels, reinforcing the need for strategic oversight to maximise profitability (Talón-Ballestero et al., 2023). This dual focus on optimising capacity, pricing, meal duration and to enhance sales by increasing the perceived value of specific products and managing distribution channels ensures a comprehensive approach to increasing overall revenue. Recent studies (Lai et al., 2020) advocate for qualitative data and revenue metrics for menu analysis, positioning it as a vital RRM component.

Continuous evaluation of RRM strategies is essential to achieve set objectives, focusing on metrics measuring revenue performance and optimising sales (Talón-Ballestero et al., 2023; Lai et al., 2020). Ensuring customer satisfaction and fairness in pricing practices is crucial for successfully adopting RRM techniques (Tyagi and Bolia, 2021).

This study analyses the literature on RRM from 1997 until 2023 and applies SLR for two reasons. Firstly, because it is systematic, explicit and reproducible, making it suitable to identify, evaluate, and interpret academic literature (Hohenstein et al., 2014). And secondly, it is a valid method to generate knowledge by synthesising the most relevant research (Cooper, 2010; Vada et al., 2020). SLRs aim to objectively assess and understand all relevant studies on a particular research question or topic area using a reliable, rigorous, and verifiable methodology (Kitchenham, 2004). As emphasised by Tranfield et al. (2003), SLRs are particularly important for addressing the shortcomings of traditional narrative reviews by adopting a structured and transparent process to mitigate researcher bias and ensure reproducibility. The approach focuses on systematic data collection, clear criteria for inclusion/exclusion, and comprehensive reporting, which enhances both the methodological rigour and the practical relevance of findings for academic and practitioner communities.

In this paper we utilise a quantitative empirical method to conduct an SLR to gather the main works on RRM. Following the process proposed by Kitchenham and Charters (2007a, b), the five steps employed are described below.

The formulation of the research questions (RQs) was grounded in the identification of specific gaps in the existing literature on RRM and guided by established methodological frameworks for question development in SLRs (Denyer et al., 2008; Tarhan et al., 2016). This process identified key gaps in the existing research, such as the temporal evolution of the field (RQ1), the predominant methodologies and their limitations (RQ2), and the issues addressed within each of the strategic RRM levers (RQ3). These gaps informed the formulation of questions aimed at advancing both theoretical understanding and practical applications of RRM.

The main research questions are:

The formulation of these questions addresses three key dimensions: the historical development of the field (RQ1), a critical evaluation of its methodological approaches (RQ2), and a thematic and prospective view of the strategic areas (RQ3). Together, these questions not only address gaps in the literature but also provide a framework that connects theory to practice, facilitating the development of more effective RRM strategies.

A key differentiating aspect of this study compared to Tyagi and Bolia (2021) is considering how the COVID-19 pandemic has influenced the evolution of RRM research. As mentioned, the pandemic led to significant changes in restaurant operations, digitalisation, and consumer behaviour, which have likely shaped new areas of academic inquiry. Understanding these shifts is essential for identifying emerging research trends and assessing how they have redefined RRM.

As recommended by Kitchenham (2004, 2013), a pre-defined protocol was followed in our study to address researcher and language bias, aiming for greater transparency and reproducibility in the systematic review process. This protocol includes clearly defined inclusion and exclusion criteria, peer-reviewed search strategies, and an iterative keyword refinement process validated by a panel of experts. The experts on our panel comprised seasoned researchers in RM and RRM with extensive experience in conducting systematic reviews and developing RM strategies in hospitality contexts. This expert validation process helped ensure that our search terms were comprehensive and minimised the risk of missing relevant studies, which is essential to maintaining the study's quality and validity (Kitchenham and Brereton, 2013). Previous studies validated the use of the selected search terms (Ammirato et al., 2020). The strategy for identifying search terms was to use the Boolean ORs search to include substitute spellings and synonyms and the ANDs search to link application domains such as restaurant or food and beverage (Martín-Navarro et al., 2023). Additionally, beyond quality assessment questions, we implemented iterative reviews of search results and rigorous selection criteria to further enhance study validity and reduce subjectivity in study inclusion.

The indexing systems we used were Scopus and Web of Science, a recognised citation databases of over 50 million scientific articles (Ammirato et al., 2020; Vieira and Gomes, 2009; Mishra et al., 2018; Mariani and Baggio, 2022). Other four well-known digital libraries: ProQuest (Tsai et al., 2011), ScienceDirect (Tsang and Hsu, 2011), Leisure Tourism (a reference database specialised in tourism accommodation, leisure and the sports industry), and Emerald (a multidisciplinary Social Sciences platform) were employed.

The Scopus database enables user friendly structure of complex searches. The study’s search was therefore based on TITLE, ABSTRACT and KEYWORDS (TITLE-ABS-KEY (“revenue management”) OR TITLE-ABS-KEY (“yield management”) OR TITLE-ABS-KEY (“dynamic pricing”) AND TITLE-ABS-KEY (“restaurant”) OR TITLE-ABS-KEY (“food and beverage”).

In WOS (Web of Science) the search was based on TOPICS ((TS=(“revenue management” and “restaurant”)) OR TS=(“revenue management” and “food and beverage”)) OR TS=(“yield management” and “restaurant”)) OR TS=(“yield management” and “food and beverage”)) OR TS=(“dynamic pricing” and “restaurant”)) OR TS=(“dynamic pricing” and “food and beverage”).

In the various digital libraries used, the search was based on TITLE, ABSTRACT and KEYWORDS, specific searches were carried out for each topical area, adding up the results. The following searches were conducted.

1. “Revenue management” AND “restaurant”

2. “Revenue management” AND “food and beverage”

3. “Yield management” AND “restaurant”

4. “Yield management” AND “food and beverage”

5. “Dynamic pricing” AND “restaurant”

6. “Dynamic pricing” AND “food and beverage”

To ensure the scientific quality of the selected papers, the search was restricted to full papers published in English in academic journals until December 2023. Conference papers, book chapters and conference reviews were excluded as they did not align with the objectives of this paper (Law et al., 2012; Tsang and Hsu, 2011; Nag and Mishra, 2023).

As previously done by authors such as Kitchenham and Brereton (2013) and Martín-Navarro et al. (2018), once the keywords were entered and the initial set of articles that formed the basis for this study was obtained, the next step involved refining the selection of studies by applying inclusion and exclusion criteria (Table 1). Specifically, following the approaches of Echeverri and Cruz (2014) and Ramírez Correa and García Cruz (2005), only scientific articles published in English were retrieved.

To ensure the quality of the literature included, the search was limited to articles published in scientific journals, as recommended by David and Han (2004). Similarly, based on the works of Houy et al. (2010), Guinea et al. (2016), and Martín-Navarro et al. (2018), the primary exclusion criteria were established. These led to the rejection of all articles that: (1) were duplicates, (2) did not address the research topic, (3) did not have full text available online, and (4) were oriented toward unrelated strategies.

After integrating the results of the different searches and eliminating duplicates, we got a single list of 248 articles. These articles were carefully reviewed to confirm compliance with the exclusion and inclusion criteria, The selection of articles was carried out manually by reading the abstracts and main conclusions to discard those studies that met the exclusion criteria. After completing this process and applying both the inclusion and exclusion criteria, a total of 109 articles remained.

To ensure the quality and reliability of the information presented in the final systematic review, these 109 articles were subsequently subjected to a comprehensive quality assessment. The evaluation followed established practices in the literature to guarantee methodological rigour, credibility, and relevance (Dybå and Dingsøyr, 2008; Kitchenham and Brereton, 2013; Unterkalmsteiner et al., 2012). Specifically, six questions were employed:

• Q1: Is this research empirical?

• Q2: Was the analysis rigorously conducted?

• Q3: Was the research methodology suitable for addressing the objectives of the study?

• Q4: Are the objectives of the study clearly defined?

• Q5: Are the limitations of the study explicitly addressed?

• Q6: Does the study include theoretical and practical implications?

Each question could be answered with “yes” (1 point), “partially” (0.5 points), or “no” (0 points). Only studies scoring over 4 (i.e. more than 50% of the perfect score) were retained for subsequent data extraction and analysis.

The formulation of these questions was grounded in prior research and has been widely employed in systematic reviews to evaluate empirical rigour, methodological suitability, and the theoretical and practical contributions of research studies. These questions provided a structured framework to ensure consistency in the evaluation process.

Additionally, the entire process was supervised and validated by a panel of experts comprising seasoned researchers in RM and RRM. This expert panel reviewed all articles and corroborated the consistency and validity of the evaluations conducted. Their contributions helped minimise potential biases and ensured the robustness of the assessment.

As seen in Supplementary material, a total of 108 papers were ultimately selected. Figure 2 illustrates the search process.

Our content analysis of the selected papers, conducted chronologically from 1997 to 2023, enables to find trends in the evolution of RRM literature (Gabbott, 2004; Mustak et al., 2013). All papers were analysed in depth according to the research questions. An excel document was manually created to sort and classify the items. Finally, considering information management and sales management as important RM levers (Talón-Ballestero et al., 2023), all papers were divided into five RRM strategic levers.

We have highlighted the results obtained in the studies conducted after COVID-19 in order to examine the theoretical and practical implications derived from the new digital environment.

Our analysis of the 108 selected studies revealed the following results.

• RQ1: What is the temporal evolution of RRM studies?

Kimes and Chase (1998) were pioneers in identifying pivotal strategic elements for enhancing restaurant revenues. According to our findings, research in this area has witnessed three peaks in 2008, focusing on capacity and waiting time management, and pricing strategies; in 2018, highlighting the implementation of rate fences and broader acceptance of RM practices; and in 2023, driven by digitalisation, evolving customer behaviour, and increased availability to big data. Enhanced data accessibility and transparency have provided deeper insights into optimising restaurant operations, with the Internet and online intermediaries playing key roles. Figure 3 illustrates the chronological progression of these publications.

• RQ2: Which type of research, study design, and methodologies are most often considered within this field? What are their most common limitations?

Table 2 findings indicate that RRM research is predominantly empirical (79.6%), consistent with prior studies on RM in hospitality and tourism (Guillet and Mohhamed, 2015). As RRM research progresses, a balance between empirical and conceptual studies is warranted. Quantitative methods are most prevalent, with 49 studies (47.1%), followed by 34 qualitative studies (32.7%) and 21 employing mixed methodology (20.2%). Following COVID-19, improvements in data management have significantly increased the relevance of the mixed-methods approach, as the enhanced ability to process and analyse larger datasets has reinforced its importance in research.

Guillet and Mohhamed (2015) noted that due to the predominance of quantitative studies in RRM, the primary methodology employed typically involves regression models, heuristic methods, and variance and covariance analysis. Regarding the evaluation of technology, frameworks and literature reviews are commonly utilised methods, while observational and simulation-based studies are prevalent in the context of capacity management. Investigations into meal duration encompass a range of surveys, questionnaires, and simulation-based analyses of restaurant performance. Studies on pricing practices have adopted a balanced approach between qualitative and quantitative methods, such as survey-based enquiries, interviews, linear regression, multiple regression, and statistical analyses involving means, standard deviations, and bivariate correlations.

The limitations of the reviewed studies relate to the diverse situations analysed, such as the nature of demand, dining purpose (Noone et al., 2009; Seo and Hwang, 2014), time period (Chan and Chan, 2008), weekdays considered (Thompson and Sohn, 2009), and seasonal effects (Ng et al., 2018). Random data collection design may also be a limitation. Regarding duration management, meal durations from the point of sale (POS) data can be inaccurate due to delays caused by customers or servers (Kimes and Robson, 2004) or different payment methods. As Legg et al. (2019) suggested, simulated data and assumptions may not fully represent all outcomes. Our review indicates that sample selection, particularly sample size (Bloom et al., 2012), sample nature and scenario bias (Heo et al., 2013), may directly affect results. Customer perception outcomes also vary based on familiarity with RM practices (Wirtz and Kimes, 2007). Perceptions of restaurant owners reveal differences in RRM application depending on management type (independent or chain) (Thompson, 2003; Rowson et al., 2016).

• RQ3: What issues are addressed in the literature on each RRM lever, and how has the crisis influenced future lines of research?

The main issues detected in each RRM strategic lever are shown in Table 3. The most relevant papers are analysed below.

Effective implementation of a RRM strategy begins with robust information management and the inability to access information about the customer and their behaviour (Webb et al., 2022) complicates the development of forecasts. Scholars have extensively explored the role of technology in enabling access to the metrics required for performance measurement and forecasting. As a foundational step, managers should collect daily data on arrivals, meal durations, and revenues (Kimes, 1999). Advances in smart technologies and enhanced data accessibility now allow restaurateurs to gain deeper insights into customer behaviour while expediting service delivery, reducing labour costs, and attracting additional business (Kimes, 2008). The integration of diverse internal and external data sources has been emphasised by Kimes and Beard (2013) and Noone and Maier (2015) as critical for supporting both strategic and tactical RRM decisions.

Technology’s role in enabling more effective RM practices has been well-documented. It enhances profitability (Kimes, 2011a, b; Gregorash, 2016), competitiveness, service quality, human resource management, and cost optimisation, factors that are particularly vital during periods of market turbulence (Zemlina et al., 2023). Properly implemented, technology can also enhance customers’ perception of control and comfort, improving satisfaction and fostering loyalty through repeat patronage (Roy et al., 2022). Despite the apparent benefits, the adoption of RM systems in restaurants remains relatively recent, with no studies comprehensively analysing their characteristics or impacts. Similarly, widely used technologies such as POS systems, which record sales and payments have been highlighted in studies by Bertsimas and Shioda (2003) and Kimes and Robson (2004). Regarding customer relationship management (CRM) systems, Zemlina et al. (2023) underscore potential benefits, including improved operational efficiency, higher customer spending, better customer experiences, and enhanced loyalty, even though the direct relationship between CRM systems and restaurants is yet to be explored. CRM systems, by capturing and disseminating customer data, play a crucial role in tailoring services to individual needs.

Key performance indicators (KPIs) are pivotal in assessing RRM practices. Kimes et al. (1999) introduced revenue per available seat hour (RevPASH) as a measure of revenue performance by integrating occupancy data to optimise capacity. Muller (1999) cautioned that RevPASH assessments must consider restaurant size, proposing the restaurant capacity ratio (RCR) to evaluate seating utilisation and customer counts. Thompson (2003) introduced contribution margin per available seat hour (CMPASH) as a profit metric, while Kimes and Robson (2004) developed the average spending per person per minute (SPM), which accounts for spending, party size, and meal duration. More recently, Heo (2017) proposed profit per available seat hour (ProPASH) and profit per available square metre (ProPASM) to measure profitability based on seat and space utilisation, respectively. Lai et al. (2020) added a KPI for menu profitability, focusing on revenue per seat per hour per meal period. Furthermore, Kalan (2023) suggested revising the RevPASH formula to incorporate the seat utilisation ratio (SUR), highlighting its importance in evaluating demand behaviour and operational efficiency.

Maximising capacity requires addressing capacity constraints, a critical aspect for restaurants due to the perishability of their inventory. Kimes et al. (1999) stressed the importance of accurate forecasting in this context. Early forecasting efforts were basic, focusing primarily on daily customer counts (Kimes, 2005). Pattern processing models have since been developed to aid managers in creating effective RRM strategies (Lasek et al., 2016). However, Choi et al. (2022) note a significant research gap in forecasting, with most studies concentrating on model development and accuracy improvements rather than identifying relevant predictive variables. A broader trend is evident as forecasting methods transition from traditional approaches, such as multiple regression and exponential smoothing, to advanced machine learning techniques, including Bayesian linear regression and boosted decision tree models.

The digital transformation accelerated by the post-COVID-19 era has prompted research on efficient forecasting methods for small restaurants with limited resources (Roy et al., 2022). Choi et al. (2022) highlight the advantages of advanced tools, such as machine learning, over traditional methods, notably in managing large datasets, improving accuracy, and reducing errors across various restaurant types.

Roy et al. (2022) further explore the use of customised dashboards for managing operational and strategic data, improving decision-making and efficiency. By integrating demographic, socioeconomic, and social media data with traditional metrics, restaurants can better understand customer behaviour, optimise inventory, and uncover new opportunities. This data-driven approach extends to logistics and service optimisation, enhancing both in-house and delivery operations.

In this context, Zemlina et al. (2023) stress the need for restaurateurs to build stronger customer communication channels, not only for data collection but also to reinforce brand loyalty. They confirm the role of digital tools in enhancing customer engagement and operational efficiency through automation and personalised services. Their study on CRM technologies reveals significant improvements in customer interaction strategies, sales, and marketing, demonstrating their value, particularly during crises.

Benchmarking, while a cornerstone of RM, remains underexplored in RRM. To date, no studies have addressed the integration of competitor analysis in RRM practices. This gap is likely due to the absence of industry-specific platforms like STR or Hostass in the hotel sector, which facilitate competitor price benchmarking.

Restaurants can enhance their operational efficiency and optimise capacity utilisation through strategic table configurations, combinations, and assignments based on demand (Kimes, 2004; Kimes and Thompson, 2004, 2005; Kimes and Robson, 2004; Thompson, 2002, 2003, 2011; Karmarkar and Dutta, 2011; Guerriero et al., 2014; Hwang and Yoon, 2009; Miao et al., 2018). Hwang and Yoon (2009) investigated the relationship between table preferences and willingness to pay, demonstrating that seating arrangements significantly influence customer experiences and satisfaction levels. They noted that seating groups of varying sizes at identical tables often results in income disparities due to differing preparation times, food costs, and service durations. Although each group occupies equal capacity in terms of table usage, they demand unequal capacity in terms of labour and service costs. Thompson (2011) emphasised the importance of tailoring seating policies to group size, while Wang et al. (2017) argued that a spend-per-minute strategy is more effective than policies focused on party size or arrival order. They recommended cost-efficient practices aimed at increasing spending per minute rather than relying solely on group size as a determinant.

Research has also identified significant insights into customer group dynamics and profitability. Seo and Hwang (2014) revealed that gender-balanced groups tend to spend less time in restaurants while achieving higher expenditures, making them more profitable than homogeneous groups. Gregorash (2016) highlighted that reservation customers typically spend more than walk-in patrons, underscoring the importance of managing reservation systems to maximise revenue.

In the context of reservation management, bookings provide dual benefits by reducing uncertainty for both customers and managers. However, they may also lead to revenue losses from no-shows, delays, or the rejection of transient customers (Thompson and Kwortnik, 2008). Effective use of reservation systems enables managers to improve forecast accuracy and optimise capacity allocation (Seo and Hwang, 2014). Bertsimas and Shioda (2003) proposed mathematical programming models to determine optimal advance booking limits and employed approximate dynamic scheduling methods to develop seating policies during service, achieving revenue increases of 3.5–7.3%. Additionally, enhancing table turnover and minimising waiting times are key strategies for maximising capacity and increasing revenue (Muller, 1999; Hwang, 2008; Hwang and Lambert, 2008).

However, last-minute cancellations and no-shows present persistent challenges. No-show rates range from 3–15%, as reported by Bertsimas and Shioda (2003), and 11–16%, according to Chiang (2023). Alexandrov and Lariviere (2012) recommended imposing no-show penalties as a deterrent. Despite the criticality of this issue, Chen (2016) observed that most RM research overlooks the impact of cancellations and no-shows, rendering such models less realistic. In response, Tse and Poon (2017) developed a comprehensive model that accounts for business-specific scenarios, no-shows, cancellations, and walk-ins to determine optimal booking limits. Strategies to mitigate no-show losses include offering unclaimed tables to walk-ins, waitlisted customers, or overbooked patrons, alongside charging reservation deposits below the service price. Given the fixed capacities and uncertain turnover rates of restaurants, it is imperative to develop tailored cancellation models to address these challenges effectively (Chiang, 2023).

The changes in practices emerging in the post-COVID-19 period have led to a convergence between RRM and that of industries such as hotels and airlines, which determine overbooking ratios to avoid empty seats caused by last-minute cancellations and no-shows. This shift has sparked increased research interest, with scholars like Chiang (2023) proposing a model to calculate the overbooking ratio. However, Chiang notes that this model is specifically applicable to fine-dining establishments, as the unpredictability of cancellations is more pronounced in smaller, independent restaurants, where such practices may not be as effective or reliable.

Time dependency in RRM is a critical focus due to the inherent uncertainty associated with meal duration. Several scholars have explored how dining duration impacts revenue outcomes (Kimes et al., 1999). Kimes (1999, 2004) argued that reducing meal durations directly correlates with increased revenue, while Kimes (2008) highlighted that minimising changeover times between seatings enhances both capacity and revenue. However, Noone et al. (2007) emphasised that initiatives to increase table turnover by accelerating service must carefully balance revenue optimisation with customer satisfaction, particularly in fine dining contexts. While shorter durations may yield financial advantages, they can adversely influence the perceived quality of service and customer satisfaction.

Noone et al. (2009) found that customers evaluate the perceived pace of service throughout their dining experience, with a greater tolerance for speedier service as the meal progresses. Thompson and Sohn (2009) identified multiple factors influencing dining duration, including restaurant size, average party size, peak demand intensity, the duration of peak demand windows, and customers' willingness to wait. Consequently, restaurateurs must skilfully manage meal durations to optimise revenue without compromising customer satisfaction (Kimes, 2011a, b). Effective measurement and control of meal duration are essential, as they significantly affect operational outcomes.

Legg et al. (2019) proposed several methodologies for predicting meal durations, recommending linear regression models as a practical option for operators new to RRM systems and survival models for experienced operators seeking to refine turn-time predictions further. Accurate forecasting of meal duration is pivotal for effective capacity management. Re-designed menus and streamlined processes, combined with improved customer arrival forecasts, can positively influence meal duration, thereby supporting enhanced operational efficiency and RM.

Following the pandemic, Roy et al. (2022) observe that emerging RRM practices are increasingly integrating all data collected by POS systems. Beyond menu selection, these now encompass variables such as arrival time and dining duration, enhancing decision-making processes to optimise capacity management. Moreover, the data collected from one restaurant can be effectively applied to another with similar characteristics within the same organisation, further improving operational efficiency.

Restaurants employ diverse price fences to optimise capacity by considering factors such as menu offerings, the time of day or year, and seating preferences (Kimes and Wirtz, 2003). Price fences are instrumental in enhancing customer perceptions of fairness regarding demand-based pricing strategies. Kimes and Wirtz (2003) further highlighted that socio-demographic differences and customer behaviour significantly influence perceptions of restaurant pricing, underscoring the importance of incorporating customer segmentation into the design of price fences.

Given the constraints of limited capacity and fluctuating demand, demand-based pricing strategies are essential for maximising high-demand periods and redistributing demand to off-peak times. Muller (1999) and Susskind et al. (2004) demonstrated the utility of incentives in encouraging customers to dine during off-peak hours, thereby alleviating peak load pressures. Tang et al. (2019) identified opportunities for restaurants to raise prices during special events or busy weekdays, using peak-load discrimination to deliver added value to diners. Additionally, Miao et al. (2018) proposed bid pricing, whereby table prices are dynamically adjusted based on demand fluctuations to maximize utilisation.

Price discrimination is a technique allowing restaurants to select customers based on their expenditure and duration (Kimes and Robson, 2004). However, segmentation must be approached with caution. Thompson (2011) observed that larger parties tend to spend less per person and occupy tables for longer durations, presenting challenges for effective pricing. Managers must remain vigilant, as excessive price increases could undermine customer satisfaction and loyalty (Herrera and Young, 2023).

Discounts and special offers are effective in stimulating demand during low periods, though they must be carefully evaluated for potential cannibalisation (sales that would have occurred without discounts) (Thompson, 2015). Norvell and Horky (2017) pointed out the difficulty of assessing promotions' true impact, as restaurants often lack clarity on whether such strategies generate incremental transactions or merely discount transactions from existing customers. Quick-service restaurants are more inclined to implement price-based promotions, owing to the sector's inherent inseparability and perishability challenges.

The COVID-19 pandemic has significantly shifted the focus of research in RRM, particularly about the new opportunities' digitalisation offers. One of the most extensively studied areas is consumer behaviour in response to price changes. Kalyanaram and Winer (2022) explore price discrimination, or fencing, as an effective strategy when consumers are familiar with pricing structures, suggesting that it can be implemented without triggering feelings of unfairness. In the context of variable pricing, Webb et al. (2022) demonstrate positive outcomes from the use of the Priority Mixed Bundle strategy, where customers can opt for a fixed menu reservation, thereby increasing revenue while maintaining customer fairness perceptions, similar to the traditional à la carte options. This shift reflects the growing importance of flexible pricing strategies in the post-COVID-19 restaurant landscape.

Establishing online sales channels is crucial for modern restaurants, though caution is advised in their implementation. Guo and Zheng (2017) observed that offering discounts through third-party platforms may inadvertently erode customer loyalty cultivated via direct online interactions.

Group-buying platforms, which gained prominence around 2010, are particularly effective in attracting new customers and increasing traffic by offering discount coupons. Restaurateurs often deploy such coupons during periods of low demand (Heo, 2016). Heo (2016) emphasised that these coupons are prepaid and time-limited, providing a distinct advantage for restaurant managers. Additionally, Zhang et al. (2020) recommended integrating regular and coupon-based customers using predictive forecasts. Despite the benefits of group-buying for managing low-demand periods, limited research has examined how restaurants can optimise this channel to its full potential.

This study also examines menu engineering as a sales management strategy. The design and presentation of menu items influence customer perceptions and choices through behavioural pricing and neuromarketing techniques (Noone and Cachia, 2020). Menu engineering, initially conceptualised by Kasavana and Smith (1982), involves analysing menu item performance in terms of popularity, profitability, preparation time, and consumption time. This analysis aims to optimise menu design and drive revenue growth. The process unfolds in two phases: first, categorising and grouping dishes on the menu, and second, implementing strategic decisions to draw consumer attention to specific items.

Price presentation is another critical aspect of menu engineering. Parsa and Njite (2004) demonstrated that pricing significantly affects consumer preferences and perceptions of quality and value, varying by restaurant segment (e.g. fine dining, casual dining, or quick service). Their follow-up studies revealed that price acts more as a selective factor than a qualifying criterion in decision-making. Supporting these findings, Kwon and Jang (2011) confirmed that perceived acquisition value is a primary determinant in consumer choices, with bundled menus often encouraging the purchase of lower-quality products. Kalyanaram and Winer (2022) also noted the influence of reference prices, which shape choice probabilities and demand.

Additionally, upselling and cross-selling strategies are widely adopted in restaurants to increase average customer expenditure. These practices are generally well-received by consumers and represent an effective means of enhancing revenue (Mhlanga, 2018a, b).

The COVID-19 pandemic catalysed significant changes in demand patterns, particularly the rise in delivery services. Ma et al. (2021) highlighted that introducing delivery services enhances sales. However, the adoption of delivery services necessitates careful evaluation of potential cannibalisation effects and shifting demand dynamics across intermediaries. Tyagi and Bolia (2021) cautioned that restaurants should remain vigilant to the challenges posed by third-party platforms and avoid over-reliance on e-commerce providers, who may dominate market conditions. Liang et al. (2023) further noted that these platforms possess vast datasets, enabling them to develop sophisticated sales strategies, potentially to the detriment of restaurant partners.

Understanding customers’ perceptions of fairness is essential for the successful implementation of RRM strategies. McGuire and Kimes (2006) posited that restaurateurs can leverage RRM techniques to enhance revenue without compromising customer satisfaction. Rate fences are widely considered fair pricing mechanisms (Kimes and Wirtz, 2002, 2003; Hwang and Yoon, 2009). Examples of these include promotional offers such as “two for the price of one”, time-of-day pricing, and distinct lunch/dinner pricing strategies (Etemad-Sajadi, 2018). Peak-load pricing, which involves offering discounts during off-peak periods rather than surcharges during peak times, is particularly effective in improving perceptions of fairness. However, weekday/weekend pricing tends to be perceived as neutral or slightly unfair. Tang et al. (2019) identified happy-hour pricing as the most widely accepted RRM practice.

Customer perceptions of fairness also influence the acceptability of waitlist management policies. Seating arrangements based on party size are generally deemed fair, whereas prioritising seating by status often causes discomfort. Allocating tables to waiting customers based on group size is another policy that aligns with fairness perceptions (McGuire and Kimes, 2006; Etemad-Sajadi, 2018). However, table location pricing tends to elicit negative reactions, as customers often perceive this practice as inequitable. McGuire and Kimes (2006) reported mixed opinions regarding prioritising reservations for large parties, while seating VIPs ahead of other patrons was largely considered unfair. Similarly, policies such as charging a non-refundable reservation fee (Tang et al., 2019) and restrictions related to the duration of table occupancy are less accepted by customers.

Dynamic pricing, while an effective tool for stimulating revenue by appealing to diverse customer segments with varying price sensitivities, must be carefully managed. Palmer and McMahon-Beattie (2008) noted that excessive price options can lead to choice overload, diminishing customer satisfaction. Etemad-Sajadi (2018) further observed that booking policies influence customer intention to patronise a restaurant, whereas table management and duration control, even when perceived as unfair, do not significantly deter patronage. Song et al. (2019) demonstrated that marketers could alleviate decision-making difficulties by implementing rate fences that differentiate price options, thereby reducing customer confusion.

Familiarity with revenue management practices plays a moderating role in customer perceptions of dynamic pricing. Herrera and Young (2023) highlighted that customers who are accustomed to variable pricing in industries such as aviation or hospitality are more likely to accept such practices in restaurants. To foster acceptance of dynamic pricing and other RRM practices, restaurant managers should proactively educate their customers. Clear communication of the benefits associated with these strategies can mitigate resistance and enhance customer understanding.

The COVID-19 pandemic marked a turning point in restaurant management, accelerating digitalisation and fostering new RM strategies (Talón-Ballestero et al., 2023). As discussed in previous sections, this transformation has reshaped multiple aspects of RM, including demand forecasting, pricing strategies, capacity optimisation, and menu engineering. In this section, we synthesise the key changes that have emerged in RM literature post-COVID-19, focusing on how these transformations have influenced research directions and industry practices. This synthesis provides a structured overview of the evolution of RM strategies in the restaurant sector, aligning with the challenges and opportunities of the pandemic.

1. Digitalisation and the evolution of data-driven decision-making

The widespread adoption of digital menus has enabled restaurants to collect and analyse large volumes of customer behaviour data, facilitating more precise studies on operational efficiency, demand forecasting, dynamic pricing, and menu engineering through advanced machine learning techniques (Choi et al., 2022; Roy et al., 2022).

The growing use of CRM systems has also fostered research on personalised offers, customer loyalty strategies, and engagement, allowing for more effective segmentation and increased profitability (Zemlina et al., 2023).

Recent literature has emphasised the role of dashboards and analytical tools in restaurant decision-making, optimising table occupancy, queue management, and customer rotation (Roy et al., 2022; Zemlina et al., 2023). Furthermore, real-time data tracking has facilitated research on dynamic pricing strategies, exploring consumer reactions to price variability and concepts such as loss aversion and perceived fairness (Kalyanaram and Winer, 2022; Herrera and Young, 2023).

1. The rise of online reservations and its impact on capacity management

Digitalisation has also enhanced the development of web platforms and mobile applications for online reservations, a trend that has solidified in the post-pandemic era. Literature has identified that this shift has resulted in earlier booking patterns, driven by pent-up demand and the carpe diem effect, fostering research into capacity optimisation strategies such as overbooking (Chiang, 2023).

Chiang (2023) expanded the newsvendor model to the restaurant context, proposing a stretched capacity approach, which allows restaurants to reduce losses from no-shows and maximise occupancy rates. This research responds to the increasing demand uncertainty post-COVID and highlights the need for more advanced strategies to optimise revenue per available seat. Additionally, the increased relevance of service shifts as an operational strategy has led to reforming traditional RM metrics, such as RevPASH, incorporating seat utilisation ratio for a more precise evaluation of operational efficiency (Kalan, 2023).

1. Dynamic pricing and behavioural pricing in restaurants

The pandemic forced restaurants to reassess their pricing strategies, driving more significant interest in dynamic pricing models to adjust rates based on fluctuating demand (Webb et al., 2023).

Among the most notable advancements is the Priority Mixed Bundle Strategy (PMB), which enables restaurants to segment pricing by offering fixed-price menus to customers who book in advance while allowing walk-ins to order à la carte (Webb et al., 2023). This approach has been shown to improve price fairness perception and increase revenue.

Additionally, implementing digital menus has facilitated research in Behavioural Pricing, leading to studies on perceived value and customers’ willingness to pay. Key concepts such as loss aversion and perceived price fairness have gained prominence in dynamic pricing and peak-load pricing research (Herrera and Young, 2023; Kalyanaram and Winer, 2022).

1. The Consolidation of Delivery Services and Their Impact on Revenue Management

The rise of delivery services during the pandemic has persisted as a key business model in the post-pandemic era (Circana, 2024), driving research into route optimisation and demand forecasting for delivery platforms.

Recent studies have applied Poisson-based distribution models to predict real-time demand for food delivery, enabling dynamic pricing adjustments, dispatch strategies, and fleet management optimisation (Liang et al., 2023). Moreover, the interconnection between delivery platforms and restaurant management systems has enhanced order allocation accuracy and enabled real-time adjustments to the menu and pricing based on demand fluctuations.

1. Advances in Menu Engineering and Price Personalisation

Digitalisation has also transformed menu engineering research, allowing for more dynamic menu optimisation and encouraging the application of Behavioural Pricing techniques. Lai and Karim (2023) have developed models enabling small and medium-sized restaurants to implement advanced menu design strategies, balancing profitability and customer satisfaction.

Digital menus have facilitated price segmentation based on consumer behaviour, promoting research into peak-load pricing, differential pricing by service shift, and price fences strategies (Webb et al., 2023). Additionally, the ability to update menus in real-time has spurred new research into consumer perception of dynamic pricing, analysing behavioural responses such as loss aversion and fairness perception in price variability (Kalyanaram and Winer, 2022; Herrera and Young, 2023).

1. The rise of AI and CRM in RRM

Finally, the increasing importance of digitalisation has driven research into customer management technologies, particularly CRM systems. Zemlina et al. (2023) highlight that CRM adoption allows restaurants to collect customer data, enhance loyalty, and tailor marketing strategies.

Furthermore, the literature has emphasised the role of artificial intelligence in RM optimisation. Choi et al. (2022) demonstrated how machine learning techniques can enhance demand forecasting and reduce decision-making errors, allowing for more precise pricing strategies, customer segmentation, and inventory optimisation.

This study maps the research trajectories of RRM highlighting the evolution of its strategies from space-based capacity optimisation to data-driven decision-making. Likewise, it reaffirms the five strategic levers and supports the addition of two levers (information and sales management) by Talón-Ballestero et al. (2023) to enhance resource optimisation.

While technology's operational and experiential benefits are praised, its sector-specific impact is often overlooked, revealing a theoretical-empirical gap. Despite the existence of restaurant-specific RM systems, their effects are under-explored, presenting opportunities to assess their impact on industry performance. Forecasting models now incorporate machine learning for better data-driven predictions, and performance metrics have evolved from RevPASH to more nuanced measures like CMPASH, ProPASH, ProPASM, SUR, and SPM.

Effective capacity management involves aligning table layout, reservations, and wait times with demand. Insights from table management and group behaviour can increase revenue through higher turnover. Current research explores overbooking and other techniques to reduce no-shows and cancellations, enhancing both revenue and customer satisfaction. Future studies should develop dynamic reservation policies, optimise peak capacity, and devise strategies to increase spend per minute.

Researchers note that meal duration affects revenue but reducing it may risk customer satisfaction. Accurate mealtime predictions are crucial. Online ordering and payment can shorten meal length and reduce errors. Future research should balance meal-duration management with customer satisfaction and improve prediction accuracy.

The restaurant sector focuses on rate segmentation, demand-driven pricing, and price discrimination. Challenges like customer segmentation and variable pricing communication remain unresolved. The literature does not explore discrimination strategies suitable for restaurants, which differ from those of hotels and airlines, wherein the former dynamic pricing is based on advanced reservations. In contrast, it is still based on peak demand or peak-load discrimination in restaurants.

Sales management has seen growth in delivery services, online sales, and partnerships with third-party platforms. These trends require careful capacity management to address cannibalisation, demand fluctuations, and distribution costs. Future directions include leveraging big data for efficient on-demand delivery decisions. Menu engineering and pricing significantly influence consumer choices, and price presentation is key. Future research should refine economic behavioural assessments and the development of distribution and e-commerce in the sector.

Differing customer responses to online and offline practices (e.g. reservations, ordering, pricing, and payment) require analysis to find a balance. RRM research is currently empirical-heavy, and a balance with conceptual studies is needed. Employing mixed-method designs could significantly benefit both practitioners and scholars.

The impact of COVID-19 has accelerated digital transformation in the restaurant industry, driving a fundamental shift in RM research. Adopting digital menus, consolidating online reservations, and expanding delivery services have prompted new research avenues in data analytics, capacity optimisation, price personalisation, and operational efficiency.

As the industry evolves, academic literature reflects a clear transition toward data-driven RM, leveraging sophisticated tools to maximise profitability and enhance the customer experience in the post-pandemic landscape.

Traditionally, RRM has been structured around three strategic levers: capacity management, meal duration, and pricing (Kimes, 1999). This study supports the addition of information management and sales management (Talón-Ballestero et al., 2023) as crucial components in modern RRM, reflecting the increasing role of big data and digital decision-making tools.

The evolving landscape of RRM underscores the need to expand its theoretical framework, particularly in information management and sales management due to the increasing role of big data and AI-driven decision-making. The increasing reliance on digital menus and platforms necessitates a shift from retrospective data analysis to proactive decision-making, where information serves not only to interpret demand but also to shape customer behaviour and operational strategies. Research on AI-driven predictive models is essential to enhance forecasting accuracy and real-time adaptability, while a deeper understanding of digital consumer behaviour will refine segmentation, pricing, and engagement strategies. The pandemic has also driven a revaluation of RRM performance metrics, with RevPASH evolving into more precise measures like the SUR (Kalan, 2023).

The adoption of overbooking models (Chiang, 2023) reflects a growing convergence between RRM and hotel/airline revenue management strategies. The role of digital platforms in capacity management requires further exploration, particularly in optimising segmentation, offer personalisation, and reservation management. Investigating how these technologies dynamically adjust availability and pricing is crucial to improving efficiency and revenue generation. Likewise, duration management remains central to balancing customer satisfaction and operational performance, necessitating studies on how controlled dining times influence consumer behaviour, table turnover, and revenue maximisation.

In pricing management, the application of AI-driven dynamic pricing models demands further research to ensure profitability while maintaining fairness and customer acceptance. The integration of peak-load pricing and behavioural segmentation presents new opportunities for demand distribution and revenue enhancement. On the other hand, increased reservations encourage intertemporal discrimination, as in the airline and hotel industry, facilitating segmentation and, therefore, dynamic pricing. Furthermore, transparency and familiarity influence fairness perceptions, requiring further theoretical development to refine pricing strategies in a digital environment. Understanding behaviour-based pricing and neuromarketing techniques will also advance revenue optimisation and improve long-term profitability.

Despite its strategic relevance, sales management remains underexplored within RRM theory. Future research should address the impact of distribution and delivery strategies, digital menu engineering, and behavioural pricing in shaping consumer choices and revenue outcomes. Additionally, advancing RRM theory requires a deeper examination of competitive pricing, the integration of no-shows and cancellations into predictive models, and the refinement of AI-based forecasting techniques.

Finally, the increasing role of CRM systems in RRM highlights a shift toward personalised marketing, customer segmentation, and loyalty strategies (Zemlina et al., 2023). The transformation of RRM into a data-driven discipline suggests that future research should explore algorithmic decision-making, predictive analytics, and real-time pricing optimisation to enhance both profitability and customer experience in the post-pandemic restaurant industry.

By identifying these research priorities, this study contributes to the ongoing development of RRM, offering new theoretical avenues to enhance its strategic application in an increasingly digitalised and data-driven environment.

The integration of AI-based RRM systems enhances demand forecasting, optimising pricing, resource allocation, and financial performance. CRM systems support comprehensive data collection, enabling targeted marketing, personalised service strategies, and improved demand management. Additionally, reservations, mobile applications, and location-based services facilitate customer loyalty reinforcement and inter-temporal price discrimination, aligning operational efficiency with revenue maximisation.

Digital platforms play a key role in capacity management, improving table allocation through segmentation and personalised reservations. Stricter booking policies, including deposits and time restrictions, mitigate cancellations and no-shows, stabilising revenue and enhancing planning. Similarly, optimising reservation duration through automated systems reduces inefficiencies, maximises table turnover, and aligns service flow with demand fluctuations while maintaining customer experience.

Sales management benefits from targeted techniques that increase bookings, regulate peak demand, and maximise occupancy. Expanding delivery services diversifies revenue streams, while digital menus incorporating behavioural pricing and neuromarketing techniques encourage higher spending and enhance decision-making. Additionally, structured pricing management through AI-driven dynamic pricing enables adaptive rate adjustments, peak-load pricing, and inter-temporal discrimination, ensuring revenue optimisation. Transparent booking guarantees and penalties further improve capacity planning by reducing uncertainty.

Enhancing service efficiency while maintaining customer satisfaction remains central to sustainable revenue growth. Targeted discounts attract new customers, while data-driven insights support long-term profitability. This review underscores the industry's rapid digital transformation, demonstrating how RRM strategies driven by digital menus and automation improve profitability, customer experience, and industry professionalisation through cost-effective and sustainable revenue practices.

The primary limitations of this study include the narrow database and its focus solely on peer-reviewed articles, inherent to SLR methodology. Expanding the scope to encompass diverse publications like books, conference papers, and theses could yield novel perspectives. Additionally, utilising specialised RRM terminology instead of general keywords may capture overlooked studies. Future RRM reviews should scrutinise publications on each strategic lever for detailed analysis and consider meta-analyses to discern variations in the associations between RM strategies and restaurant performance based on performance metrics.

Funding: Funding for open access charge was provided by Universidad de Cádiz/CBUA. Additionally, this work was partially supported by the State Research Agency of the Ministry of Science and Innovation (reference code PID2022-140786NB-C31).

The supplementary material for this article can be found online.