Front-Loaded or Balanced? The Mechanism through Which Review Order Affects Overall Ratings in Premium Service Settings

He Wang, Ziyu Zhou, Hanxiang Liu · 2026

Read Front-Loaded or Balanced? The Mechanism through Which Review Order Affects Overall Ratings in Premium Service Settings by He Wang, Ziyu Zhou, Hanxiang Liu (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-00008.txt.

Introduction

He Wang and colleagues test whether rating first or writing first changes consumer scores. Across three trials and a large second data study, rating first pushed scores apart: high quality service scored higher, low quality scored lower. The authors trace this split through feeling based shortcuts and mental effort, and they find fun products show it more than useful ones. Data from Yelp and Letterboxd match the pattern, with Yelp scores spread wide and Letterboxd scores bunched.

Catalog record

CREATORHe Wang, Ziyu Zhou, Hanxiang Liu
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE51936
LICENSEhttp://creativecommons.org/licenses/by-sa/4.0/
RIGHTSCC BY SA
SOURCEarXiv

Source: https://arxiv.org/abs/2602.00008

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Front-Loaded or Balanced? The Mechanism through Which Review Order Affects Overall Ratings in Premium Service Settings He Wang, Ziyu Zhou, Hanxiang Liu Abstract In the increasingly prevalent landscape of high-quality service contexts, whether consumer evaluation interfaces adopt a rating-first or review-first sequence has become a critical factor shaping rating authenticity and feedback quality. While prior research has primarily examined review content and sentiment, systematic investigation into how evaluation order influences rating outcomes remains limited. Through exploratory analyses, we find that Letterboxd— which employs a review-first, rating-after mechanism— exhibits a more centralized rating distribution with fewer extreme scores, whereas Yelp— which adopts a rating-first, review-after mechanism— shows a pronounced bimodal distribution with more polarized ratings. Three controlled experiments further demonstrate that in high-quality service contexts, a rating-first (vs. review-first) interface significantly elevates consumers’ overall ratings. Mechanism analyses indicate that cognitive effort and affective heuristics serve as dual pathways: a rating-first (vs. review-first) sequence reduces cognitive effort and heightens affective heuristics, thereby increasing rating scores. Moreover, service quality moderates this process. When service quality is low, the rating-first (vs. review-first) sequence instead leads to lower ratings. This research reveals the psychological mechanisms through which evaluation order affects consumer ratings via cognitive and affective pathways. It extends theoretical understanding of online rating formation and offers practical implications for optimizing platform interface design to enhance rating authenticity and credibility. Keywords: evaluation order, consumer ratings, affective heuristics, cognitive effort, service quality 1 Introduction In recent years, online consumption and digital platform economy have been developing rapidly around the world. It has been a significant part of modern world economy. European Commission carries market investigation to ensure their Digital Market Act keep up with the fast pace of digital markets. Xue et al. define the platform economy as an ecosystem built upon digital technologies—including the Internet, cloud computing, big data, and the Internet of Things—in which platform enterprises play a central coordinating role. These enterprises design and operate integrated systems that connect platforms, consumers, and service providers, thereby shaping the activities and interactions of upstream and downstream market participants [ 1 ] . As the global leader in digital economy, the scale of U.S. digital economy had reached 17.2 trillion US dollars in 2022, accounting for 64.6% of its GDP. It has effectively supported the stable growth of the macroeconomy, further underscoring the increasing significance of the digital sector within the broader economic system. Against this backdrop, and amid the continued expansion of the digital economy, related segments such as online food delivery platforms, e-commerce platforms, and consumer review platforms have also experienced rapid growth. According to McKinsey’s report on food delivery, global online food delivery transactions reached approximately $150 billion in 2021, and the market is growing exponentially. The scale of e-commerce is also massive, with the global e-commerce market size reaching about $27 trillion in 2022 and the number of online shoppers exceeding 2.3 billion [ 2 ] . Major e-commerce platforms like Amazon and Alibaba dominate the industry, and future development will become more multidimensional and in-depth. Review platforms like Yelp, TripAdvisor, and Google Reviews have permeated every aspect of daily life— from dining and shopping to travel and accommodation— becoming essential sources for consumers to gather information and make decisions. It is fair to say that modern life has become inseparable from the online consumption and platform service ecosystem built by the digital economy. Social psychologist Robert B. Cialdini systematically introduced the concept of "social proof" in his book Influence: Science and practice . He posits that when individuals lack information and cannot make accurate judgments, they tend to identify with and mimic others’ behaviors, believing that actions already taken by others reflect the correct choice. Thus, referencing others’ behaviors can reduce decision-making risks [ 3 ] . In today’s internet landscape, users are faced with a bewildering array of choices. Often, they struggle to make decisions due to a lack of specific, precise information or the inability to compare the merits of different options. In such situations, they turn to others’ recommendations to assist their decision-making. This reliance on peer endorsement is a crucial factor in the establishment and continuous evolution of user review systems. As platforms continue to evolve and refine their systems, user review platforms have become a vital medium for building trust, facilitating transactions, and optimizing services. The ratings and comments consumers leave on digital platforms now serve as crucial reference points for merchants to enhance their services, for platforms to allocate exposure to merchants, and for consumers to make informed decisions. The platform evaluation system provides users with a highly interactive and intuitive window to express their consumption experiences, gradually becoming a crucial pillar of the digital economy’s trust framework. Currently, nearly all service platforms invite consumers to rate and review their experience after completing a transaction, thereby providing helpful guidance for subsequent customers. Most platforms’ review sections adopt a "rating-first, review-after" design approach. Users must first provide a star rating or numerical score before optionally submitting text or image reviews based on their experience. To showcase product or service features more comprehensively, some merchants incentivize users with coupons or small gifts to encourage more detailed written reviews. The design of review sections typically places the rating section above the comment section, or displays the comment section only after the rating section is completed. This design prioritizes the reviewer’s experience, aiming to enable faster completion of reviews and enhance the overall evaluation process’s user experience and interaction convenience. Simultaneously, it maximizes the preservation of intuitive, structured data, offering significant value to platforms, businesses, and potential consumers. However, such designs may exert a potential guiding effect on users. According to the anchoring effect proposed by psychologists Amos Tversky and Daniel Kahneman in the 1970s, people tend to be influenced by the initial information when making decisions [ 4 ] . Theory transferred to the review interface reveals that consumer feedback tends to reinforce their initial ratings, potentially compromising the objectivity and accuracy of evaluations as consumers strive to maintain consistency with previous scores. The order effect in questionnaires further indicates that the sequence of questions systematically influences respondents’ outcomes. When overall impression questions are answered first, consumers often rely on abstract impressions, thereby reducing the rationality of responses to subsequent questions. Beyond mainstream rating systems, a handful of platforms position the rating section below the review section. This design directs reviewers’ attention first to textual and image-based comments, guiding them to complete the review section first. This approach largely mitigates the aforementioned effects. However, this review model is rarely adopted. Is this due to platforms overlooking the impact of sequence, or does this model itself have drawbacks? Based on these observations, this paper explores how different review sequences influence rating outcomes. 2 Related Work 2.1 The Importance of Ratings Ratings on social media exert a critically significant influence on consumers’ decision-making processes. Specifically, consumers can select superior merchants by relying on high ratings or avoid establishments with suboptimal consumption environments through low ratings, thereby optimizing their overall consumption experience. Additionally, following their consumption behaviors, consumers can provide post-purchase ratings based on personal satisfaction, leaving verifiable traces within social networks and evaluation systems. Bass (1969) [ 5 ] proposed

Read the complete text