The impact of green claims on consumer responses in bio-based paper and packaging: A digital content and sentiment analysis of Amazon and Trendyol reviews

Okay, S. (2026). "The impact of green claims on consumer responses in bio-based paper and packaging: A digital content and sentiment analysis of Amazon and Trendyol reviews," BioResources 21(1), 2176–2185.

Abstract

This study was conducted to examine how sustainability-oriented (“green”) claims influence consumer responses to bio-based paper and packaging products in digital marketplaces. A total of 611 verified consumer reviews were analyzed, including 230 from Amazon and 381 from Trendyol, covering the period between 2020 and 2025. Through digital content and sentiment analysis, the relationship between eco-communication and perceived value, satisfaction, and trust was explored. A cross-platform analytical pipeline was developed to ensure transparency and reproducibility. Daily review frequencies were aggregated into seven-day rolling averages so that temporal patterns such as enthusiasm bursts and stabilization trends could be identified. It was observed that consumer sentiment was predominantly positive but asymmetrically distributed. Explicit “green” claims such as biodegradable, recyclable, and eco-friendly were found to stimulate higher engagement while also inducing greater polarization, resulting in a characteristic J-shaped rating distribution. Broader variance in Amazon reviews was attributed to the platform’s diverse cultural composition, whereas Trendyol reviews exhibited faster stabilization, indicating stronger normative coherence and trust in localized sustainability messages. Overall, these findings suggest that sustainability has shifted from an optional feature to a baseline expectation in digital consumption. “Green” attributes are now perceived as authenticity and reliability cues that merge ethical commitment with functional quality.

Download PDF

Full Article

The Impact of Green Claims on Consumer Responses in Bio-Based Paper and Packaging: A Digital Content and Sentiment Analysis of Amazon and Trendyol Reviews

Suat Okay *

DOI: 10.15376/biores.21.1.2176-2185

Keywords: Bio-based packaging; Green marketing; Consumer behavior; Digital marketplaces; Sustainability communication; Sentiment analysis

Contact information: Department of Accounting and Tax Applications, Accounting and Tax, Vocational School of Social Sciences, Batman University, Batman, Türkiye;

*Corresponding author: suat.okay@batman.edu.tr

INTRODUCTION

Over the past decade, sustainability has been transformed from a peripheral marketing theme into a central determinant of consumer value creation and corporate legitimacy. In both mature and emerging economies, environmentally responsible communication has been regarded as essential for establishing brand credibility and market differentiation (White et al. 2019; Peattie and Belz 2021). Among various sustainability signals, bio-based paper and packaging represent concrete examples of circular-economy practices, bringing together material innovation and environmental communication (Kılınç et al. 2023; Kılınç and Korkmaz 2024). Consumers view these products not only as functional items but also as indicators of ecological authenticity and corporate responsibility (Iovino et al. 2023).

Digital marketplaces such as Amazon and Trendyol amplify these symbolic cues through algorithmic visibility, user ratings, and peer-generated reviews. In this environment, online reviews act as both informational and emotional signals: they convey product performance while also reflecting users’ affective reactions (Beugelsdijk et al. 2015; Chen et al. 2022).

Despite the growing interest in sustainability communication, two major research gaps have been identified. First, most prior studies have emphasized corporate-side messaging, examining how companies construct or exaggerate green credentials (Delmas and Burbano 2011; de Freitas Netto et al. 2020; Szabo and Webster 2021), while consumer-side feedback processes where authenticity and credibility are continuously negotiated have received limited attention. Second, much of the existing empirical evidence has been derived from controlled surveys and experiments that isolate stimuli from their natural linguistic and social contexts (Huang and Rust 2021). Consequently, empirical evidence remains sparse regarding how genuine consumers articulate, endorse, or challenge sustainability narratives within uncontrolled digital ecosystems.

Understanding how consumers respond to green claims in online marketplaces is important for advancing both marketing theory and sustainable business practice. Prior research shows that eco-attributes such as biodegradable or recyclable materials can influence satisfaction, perceived value, and repurchase intentions. However, the effectiveness of these claims varies across contexts; cultural values, platform features, and social-proof mechanisms shape whether consumers view sustainability information as credible or exaggerated (Hussain and Huang 2022; Peña-García et al. 2024; Yum et al. 2024). Emerging-market consumers offer useful insights into how digital trust, moral framing, and local identity shape responses to green claims (Kutaula et al. 2024; Zhao et al. 2024).

Table 1. Analytical Workflow Applied to Consumer Reviews of Bio-Based Packaging Products

This study addresses the gaps just described by examining 611 verified online reviews of bio-based paper and packaging products collected from Amazon and Trendyol between January 2020 and October 2025, using a standardized cross-platform pipeline that encompasses text cleaning, linguistic normalization, and sentiment scoring. In line with the theoretical and empirical gaps identified in the literature, the study focuses on three core research questions: RQ1: How are consumer ratings for bio-based paper and packaging products distributed across digital marketplaces? RQ2: How do consumer sentiments evolve over time, and what temporal patterns characterize the stabilization of sustainability perceptions? RQ3: How does sustainability-related discourse differ between Amazon and Trendyol, and what factors explain cross-platform heterogeneity? The conceptual and analytical process guiding this investigation is illustrated in Table 1.

EXPERIMENTAL

Materials and Methods

By integrating digital consumer analytics with sustainability-marketing theory, the relationship between claim explicitness, sentiment volatility, and cultural context was examined to reveal how these factors shape consumer trust within digital marketplaces.

A cross-platform digital content analysis approach was adopted to examine how consumers respond to sustainability-oriented packaging communications. All data were collected exclusively from publicly accessible product-review pages on Amazon and Trendyol, ensuring full compliance with research ethics and data-protection principles. No private or personally identifiable information was accessed or utilized during any stage of data processing.

The unified dataset consisted of 611 verified consumer reviews: 230 from Amazon, representing a global and linguistically diverse audience, and 381 from Trendyol, reflecting an emerging-market context rooted in Türkiye. All sampled products were identified as including explicit sustainability descriptors such as eco-friendly, biodegradable, recyclable, organic, or bamboo-based. Each review entry included the verified star rating (1 to 5 scale), review text, and timestamp, allowing both textual and temporal analyses of consumer evaluations to be conducted.

Data preprocessing followed a transparent and reproducible workflow that standardized the text across both languages and platforms. The HTML data were parsed with Python’s BeautifulSoup library, and the text was then normalized through UTF-8 encoding, lowercasing, and the removal of punctuation, stop-words, and duplicate entries. These steps ensured a consistent structure across the dataset before sentiment analysis was applied. Language-specific cleaning routines were applied separately for English (Amazon) and Turkish (Trendyol) datasets using tailored stop-word dictionaries to preserve semantic precision.

Sentiment polarity was estimated on a continuous scale (–1 to +1) using a hybrid VADER TextBlob ensemble. VADER was used to compute an initial polarity score due to its suitability for short, informal review text, and TextBlob was used to generate a secondary polarity estimate that adds complementary lexical coverage. The final hybrid score was calculated as a weighted average (0.6 × VADER + 0.4 × TextBlob) to balance robustness and interpretability across platforms. Because both tools were originally developed for English, their use for the Turkish corpus was intended to provide a consistent comparative index rather than fine-grained emotion detection; therefore, Turkish-specific preprocessing and aggregate-level interpretation were emphasized. For categorical summaries, hybrid scores greater than +0.05 were labeled positive, scores lower than –0.05 were labeled negative, and remaining scores were labeled neutral. To capture consumer-attention dynamics, review frequencies were transformed into daily time series, and rolling seven-day means were calculated for both sentiment and rating averages. This smoothing procedure was used to identify short-term fluctuations, enthusiasm bursts, and stabilization patterns in consumer evaluations.

While the overall dataset covers the full period from 2020 to 2025, shorter time windows (e.g., 180 days) were used exclusively for rolling-window analyses and graphical visualization purposes, in order to illustrate short-term dynamics without altering the underlying sampling horizon. The rolling-window formula was applied to smooth short-term volatility while preserving behavioral inflection points. Through this temporal aggregation procedure, an interpretable representation of how initial bursts of enthusiasm evolve toward equilibrium in consumer perceptions was obtained. Descriptive statistics summarizing review distributions, mean ratings, sentiment proportions, and textual characteristics are presented in Table 2.

Table 2. Descriptive Statistics of Review Data

To ensure transparency and replicability, all cleaning and analysis scripts were version-controlled and documented within an open-source workflow (e.g., Python 3.12 and Jupyter Notebook). Cross-platform review panels were merged through standardized brand and product identifiers so that direct comparisons of sentiment stability, linguistic tone, and rating heterogeneity could be performed. A complementary keyword co-occurrence analysis was also conducted to identify thematic clusters such as softness, durability, absorbency, and eco-friendliness, thereby highlighting the intersection between functional and ethical dimensions of consumer perception.

Overall, the methodological design combines computational rigor with marketing-focused interpretability and offers a reproducible framework for evaluating sentiment across different cultural contexts. In the subsequent section, comparative results are presented and the observed behavioral regularities are discussed in relation to the digital diffusion of sustainability meanings across marketplace contexts.

RESULTS AND DISCUSSION

A comparative analysis of consumer evaluations across Amazon and Trendyol was conducted to identify behavioral regularities in how sustainability-oriented paper and packaging products are perceived. Consistent, though contextually distinct, behavioral patterns were evident across both marketplace environments. In each dataset, a well-known J-shaped rating distribution was identified, indicating that positive evaluations substantially outnumber neutral or negative ones. To numerically illustrate this skewness, positive reviews accounted for 78.3% of Amazon ratings and 82.9% of Trendyol ratings, whereas negative evaluations remained comparatively low (10.0% and 6.3%, respectively). The distribution also exhibited right-skewed characteristics, reflected in the higher frequency of 5-star ratings and the low density of mid-range evaluations, further confirming the presence of a J-shaped pattern. This skewness, which is common in digital marketplaces, appears to arise from factors such as positivity bias, self-selection, and social-proof dynamics (Chevalier and Mayzlin 2006). From a marketing psychology perspective, highly satisfied consumers are more likely to share their experiences, which strengthens perceptions of product reliability and contributes to social validation.

Fig. 1a. Rating distribution for Amazon reviews of bio-based packaging products

Fig. 1b. Rating distribution for Trendyol reviews of bio-based packaging products. Distributional comparison illustrates the J-shaped pattern across both marketplaces.

Upon closer examination of Fig. 1a, a wider dispersion was observed in Amazon reviews, with noticeable tails near mid-level ratings (three stars). This pattern was interpreted as an indication of greater heterogeneity in consumer expectations across a global audience that varies in price sensitivity, delivery standards, and environmental familiarity. Such dispersion is often regarded as a credibility-testing phase, during which diverse users interpret sustainability cues through distinct cultural heuristics. In contrast, the curve displayed in Fig. 1b appeared more condensed for Trendyol, suggesting a higher convergence of evaluative norms. Within emerging-market contexts, eco-attributes were found to be more strongly integrated into consumers’ moral self-concepts, transforming sustainability signals into expressions of collective trust rather than skepticism (Hofstede 2022).

Fig. 2a. Rolling seven-day mean of ratings for Amazon, based on a rolling-window visualization applied to the full review dataset (2020 to 2025)

Fig. 2b. Rolling seven-day means of ratings for Trendyol. The graph shows the temporal evolution of mean ratings, highlighting early volatility followed by convergence and stabilization.

A temporal analysis was subsequently conducted to extend this interpretation. The temporal evolution of consumer sentiment was modeled using a seven-day rolling mean, calculated as the average rating of the current day and the preceding six days. This smoothing approach minimized short-term volatility while retaining major behavioral inflection points, revealing two distinct phases in the time-series patterns: (i) an initial period of heightened affective engagement, during which daily ratings exhibited sharper fluctuations, and (ii) a subsequent stabilization phase in which sentiment converged toward an equilibrium level. The slower convergence observed in Amazon’s data reflects the platform’s higher demographic and linguistic heterogeneity, which tends to delay the formation of collective trust, whereas Trendyol displayed a faster and smoother stabilization due to more homogeneous evaluative norms and a shared normative orientation toward domestic and eco-symbolic values. This divergence provides evidence that platform ecology the structural configuration of digital marketplaces serves as a moderating factor in the translation of sustainability narratives into lasting consumer confidence, consistent with expectation–performance alignment and information diffusion theories, according to which emotional extremity gradually diminishes as perceived reliability increases (Moe and Trusov 2011; White et al. 2019; Singh et al. 2024).

Table 3. Descriptive Statistics for Amazon and Trendyol Review Datasets

As shown in Table 3, higher average satisfaction accompanied by lower variance was observed among Trendyol users, indicating that trust and perceived authenticity were more consistently maintained within this market. In contrast, a broader standard deviation was detected in Amazon reviews, suggesting the presence of evaluative pluralism in which sustainability messages coexist with diverse functional expectations. From a marketing-behavioral perspective, this distinction has been interpreted as evidence that cultural familiarity and message proximity jointly shape the effectiveness of eco-claims. When environmental messaging is perceived as being aligned with local values, consumer evaluations tend to cluster around positive anchors; conversely, when such claims are filtered through global skepticism, evaluations are found to remain more dispersed (Srisathan et al. 2024).

When these empirical patterns are synthesized, a two-phase behavioral model of digital eco-communication can be inferred:

1. Enthusiasm phase characterized by initial affective saturation, during which green claims are perceived to function as attention triggers and emotional cues that amplify short-term positivity.

2. Stabilization phase defined by cognitive recalibration, through which functional performance and message credibility are jointly evaluated to determine sustainable satisfaction trajectories.

Together, these dynamics show how perceptions of sustainability become more stable over time in digital feedback systems. The proposed model contributes to marketing theory by linking short-term emotional variation with longer-term trust formation in online marketplaces. Consistent with sustainability signaling theory, “green” claims are conceptualized not only as ethical indicators but also as performance heuristics, enabling consumers to infer product reliability through moral and symbolic framing (Iovino et al. 2023). Ultimately, the cross-platform findings are understood to demonstrate that digital sustainability narratives function as evolving social learning processes through which collective interpretation transforms communication into legitimacy.

LIMITATIONS

This study has several limitations that should be acknowledged when interpreting the findings. First, the analysis is based solely on publicly available consumer reviews from Amazon and Trendyol, which may not fully capture the broader population of users or offline purchasers. Second, the sentiment analysis relies on a hybrid lexicon-based model, which, despite its cross-linguistic advantages, may miss nuanced emotional expressions or culturally embedded linguistic cues. Third, the study focuses on descriptive and comparative inferences rather than causal relationships; therefore, the patterns observed in sustainability-related evaluations cannot be interpreted as evidence of behavioral causality. Finally, restricting the dataset to bio-based paper and packaging products limits generalizability to other product categories. Future research could extend the analytical framework to experimental designs, additional platforms, or multimodal data sources such as images and verified eco-label metadata.

CONCLUSIONS

Large-scale, cross-platform evidence from this study demonstrates how consumers perceive and evaluate bio-based paper and packaging products in digital marketplaces.

Analysis of 611 verified reviews from Amazon and Trendyol showed that sustainability-oriented packaging is consistently evaluated positively and is perceived as a core indicator of product quality rather than a peripheral ethical attribute.
Consumer ratings display a characteristic J-shaped distribution, indicating that positive evaluations dominate while neutral and negative responses remain limited across both platforms.
Consumer sentiment follows a two-phase pattern in which early affective responses gradually stabilize as product experience accumulates, leading to more consistent credibility-based evaluations.
Platform context influences evaluation dynamics, with more homogeneous and localized marketplaces exhibiting faster convergence of trust compared to culturally diverse global platforms.
Sustainability-related claims are more effective when they are supported by clear and verifiable product information rather than symbolic or purely emotional messaging.
Communicating environmental benefits together with functional performance attributes strengthens perceived authenticity and supports consumer trust in bio-based packaging products.
Platform-level transparency mechanisms contribute to higher consumer confidence and more consistent interpretation of sustainability claims in digital marketplaces.

ACKNOWLEDGEMENTS

A generative artificial intelligence tool (ChatGPT, OpenAI) was used during the preparation of this manuscript. The tool was employed exclusively to enhance language clarity, improve readability, and support minor editorial adjustments. All research design, analysis, interpretations, and conclusions were entirely developed by the author.

REFERENCES CITED

Beugelsdijk, S., Maseland, R., and van Hoorn, A. (2015). “Are scores on Hofstede’s dimensions of national culture stable over time? A cohort analysis,” Global Strategy Journal 5(3), 223-240. https://doi.org/10.1002/gsj.1098

Chen, T., Samaranayake, P., Cen, X., Qi, M. and Lan, Y-C. (2022). “The impact of online reviews on consumers’ purchasing decisions: Evidence from an eye-tracking study,” Frontiers in Psychology 13, article 865702. https://doi.org/10.3389/fpsyg.2022.865702

Chevalier, J. A., and Mayzlin, D. (2006). “The effect of word of mouth on sales: Online book reviews,” Journal of Marketing Research 43(3), 345-354. https://doi.org/10.1509/jmkr.43.3.345

De Freitas Netto, S. V., Sobral, M. F. F., Ribeiro, A. R. B., and Soares, G. R. L. (2020). “Concepts and forms of greenwashing: A systematic review,” Environmental Sciences Europe 32, article 19. https://doi.org/10.1186/s12302-020-0300-3

Delmas, M. A., and Burbano, V. C. (2011). “The drivers of greenwashing,” California Management Review 54(1), 64-87. https://doi.org/10.1525/cmr.2011.54.1.64

Hofstede Insights. (2022). “Frequently asked questions about updated country scores,” Retrieved from https://www.theculturefactor.com/frequently-asked-questions

Huang, M.-H., and Rust, R. T. (2021). “A strategic framework for AI in marketing,” Journal of the Academy of Marketing Science 49(1), 30-50. https://doi.org/10.1007/s11747-020-00749-9

Hussain, S., and Huang, J. (2022). “The impact of cultural values on green purchase intentions through ecological awareness and perceived consumer effectiveness,” Frontiers in Environmental Science 10, article 985200. https://doi.org/10.3389/fenvs.2022.985200

Iovino, R., Testa, F., and Iraldo, F. (2023). “Do consumers understand what different green claims actually mean? An experimental approach in Italy,” Journal of Advertising 53(2), 200-214. https://doi.org/10.1080/00913367.2023.2175279

Kılınç, İ, Budakçı, M., and Korkmaz, M. (2023). “The use of environmentally friendly abrasive blasting media for paint removal from wood surfaces,” BioResources 18(1), 1185-1205. https://doi.org/10.15376/biores.18.1.1185-1205

Kılınç, İ. and Korkmaz, M. (2024). “Agricultural waste-based composite materials: Recycling processes, technical properties, and industrial applications,” European Journal of Technique (EJT) 14(2), 136-145.

Kutaula, S., Gillani, A., Gregory-Smith, D., and Bartikowski, B. (2024). “Ethical consumerism in emerging markets: Opportunities and challenges,” Journal of Business Ethics 191(4), 651-673. https://doi.org/10.1007/s10551-024-05657-4

Moe, W. W., and Trusov, M. (2011). “The value of social dynamics in online product-ratings forums,” Journal of Marketing Research 48(3), 444-456. https://doi.org/10.1509/jmkr.48.3.444

Peattie, K., and Belz, F.-M. (2021). “Sustainability marketing and the transition toward a circular economy,” Journal of Business Strategy and the Environment 30(8), 3620-3632. https://doi.org/10.1002/bse.2831

Peña-García, N., Losada-Otálora, M., Pérez Auza, D., and Paula Cruz, M. (2024). “Reviews, trust, and customer experience in online marketplaces: The case of Mercado Libre Colombia,” Frontiers in Communication 9, article 1460321. https://doi.org/10.3389/fcomm.2024.1460321

Singh, N., Misra, R., Quan, W., Radic, A., Lee, S., and Han, H. (2024). “An analysis of consumer’s trusting beliefs towards the use of e-commerce platforms,” Humanities and Social Sciences Communications 11, article 899. https://doi.org/10.1057/s41599-024-03395-6

Srisathan, W. A., Ketkaew, C., Jantuma, N., and Naruetharadhol, P. (2024). “Trust and website conversion in consumer responses to green product purchasing: A new perspective through the lens of innovative website design’s technology integration,” Heliyon 10(1), article e23764. https://doi.org/10.1016/j.heliyon.2023.e23764

Szabo, S., and Webster, J. (2021). “Perceived greenwashing: The effects of green marketing on environmental skepticism,” Journal of Business Ethics 171(4), 719-739. https://doi.org/10.1007/s10551-020-04461-0

White, K., Habib, R., and Hardisty, D. J. (2019). “How to SHIFT consumer behaviors to be more sustainable: A literature review and guiding framework,” Journal of Marketing 83(3), 22-49. https://doi.org/10.1177/0022242919825649

Yum, K., Kang, J., and Choi, M. (2024). “Exploring the impact of characteristic-based review systems on customer engagement in map-based platforms,” Humanities and Social Sciences Communications 11, article 1567. https://doi.org/10.1057/s41599-024-04082-2

Zhao, A. L., Dermody, J., Koenig-Lewis, N., and Hanmer-Lloyd, S. (2024). “Cultivating sustainable consumption: The role of harmonious cultural values and pro-environmental self-identity,” Journal of Consumer Behaviour 23(2), 1014-1031. https://doi.org/10.1002/cb.2261

Article submitted: December 10, 2025; Peer review completed: January 9, 2026; Revised version received and accepted: January 10, 2026; Published: January 21, 2026.

DOI: 10.15376/biores.21.1.2176-2185