Kenya faces a severe plastic waste challenge that highlights the need for reuse and circular solutions. National estimates indicate that the country generates about 22,000 tonnes of solid waste per day, of which 20% is plastic. The daily per capita plastic consumption is estimated in 0.03 kg per person, amounting to 0.5–1.3 million tonnes of plastic waste annually (Kenya Plastic Pact, 2026). However, reports indicate that only 27% of plastic waste is collected in Kenya, of which just 8% is recycled (Pucino et al., 2020; Kenya Plastic Pact, 2026). The remaining plastic waste is landfilled, incinerated, or leaked into the environment, contributing to widespread pollution of urban, rural and natural areas (Pucino et al., 2020; Kenya Plastic Pact, 2026).

Kenya’s waste profile and plastics landscape have attracted substantial policy and private-sector attention over the past decade. Notable efforts include the 2017 nationwide ban on plastic carrier bags, the 2020 prohibition of single-use plastics in protected areas, and the launch of the Kenya Plastics Pact Roadmap to 2030 (UNEP, 2020; Kenya Plastic Pact, 2021; NEMA, 2025). The Kenya Plastics Pact is a multi-stakeholder alliance comprising government, industry, civil society, and informal sector actors working toward a circular plastics economy by 2030. Its objectives include eliminating problematic single-use plastic items, ensuring that all packaging is reusable or recyclable, and increasing recycling rates in the population (Kenya Plastic Pact, 2021; Kenya Plastic Pact, 2026). Participating stakeholders commit to a shared set of ambitious and time-bound targets, where progress is monitored and publicly reported on an annual basis to strengthen accountability and accelerate the transition toward sustainable outcomes (Kenya Plastic Pact, 2021).

Moreover, consumer research indicates growing public awareness of plastic pollution and increasing willingness to adopt reusable alternatives. Studies have reported that 67% of consumers supported the 2017 plastic bag ban, and have documented notable increases in reusable bag ownership and consumer consciousness around plastic waste reduction following its implementation (Omondi and Asari, 2021; Getanda, 2023). However, despite high awareness and supportive policy frameworks, sustainable practices such as reuse, refilling, and source separation remain uneven across communities (Getanda, 2023). Economic constraints, limited waste management infrastructure, and inconsistent regulatory enforcement weaken the translation of consumer awareness into sustained behavior change (Oguge et al., 2021; Getanda, 2023). While formal community-based and private-led initiatives for plastic reuse and recycling are emerging, the persistence of single-use plastics in informal markets continues to undermine reuse efforts (Oguge et al., 2021; Getanda, 2023). This contextual backdrop highlights the urgency of addressing plastic pollution in Kenya and the opportunity for novel interventions to complement existing policy and infrastructure initiatives in advancing the country’s transition toward a circular economy.

Set against this background, the present research investigates the behavioral and structural barriers and enablers of reusable-on-the-go systems among low-income consumers in Kenya. Latest data from the Kenya National Bureau of Statistics (KNBS) (2024) provide an indicative estimate of the low-income consumer base in Kenya. In Nairobi, the Lower Income Group comprises approximately 3.38 million people living in households with monthly expenditures below KSh 82,000 as of May 2025 (Kenya National Bureau of Statistics (KNBS), 2024).¹ This group includes poor, near poor, and low-income segments especially relevant to this study. At the national level, conservative estimates indicate that 20.2 million people (39.8% of the population) were living below the poverty line in 2022, including 15 million in rural areas and 5.4 million in urban areas. Taken together, these figures highlight the large consumer base and market potential of the target population for this research.

Furthermore, the present qualitative study was conducted as part of a larger project led by Base-of-the-Pyramid Innovation Center (Bopinc) in collaboration with Mount Kenya University (MKU) with funding from TRANSFORM. TRANSFORM is a collaborative initiative led by Unilever, the United Kingdom’s Foreign, Commonwealth & Development Office, and Ernst & Young. It aims to test and scale new solutions that tackle environmental challenges, improve health and wellbeing, and build inclusive economies (TRANSFORM, 2025). TRANSFORM partners are interested in understanding the behavioral and structural determinants shaping the adoption of reusable behaviors in Kenya to best support and scale-up the circular economy in low-income markets.

The study developed in-depth interviews with nine key expert entrepreneurs with long trajectories working in Kenya’s reusable-on-the-go market. All interviews were developed between November 2024 and January 2025, and each interview was held privately and individually to ensure the confidentiality of responses and to minimize social desirability biases. The approach ensures transparency and promotes trust to allow participants to safely express their sincere perceptions and best respond to the questions based on their lived experience. The initial recruitment strategy aimed to collect information from at least seven experts and to continue interviewing experts until data saturation was reached, as indicated by the absence of any new insights into additional interviews.

The study employed purposive sampling to recruit expert informants, leveraging recommendations from collaborators and local networks who have previously worked with Bopinc, MKU and Unilever in Kenya. The three organizations nominated experts with substantial field experience and sectoral insight. Experts were selected based on their in-depth understanding of Kenya’s reusable-on-the-go market by demonstrating a minimum of 3 years of market experience working with products especially tailored to the low-income consumer segment (such as personal and home care items, and food and beverage products). The recruitment strategy offered advantages to identifying informants with the most specialized expertise and contextual knowledge relevant to the market and population of interest (Maruster, 2013).

A subsequent snowball sampling strategy was included to enhance the potential breadth and diversity of experts’ perspectives (Gierczyk et al., 2023). The last question of the interview asked participants about the opportunity to recommend other experts who met the inclusion criteria (see Question 31 in Section D of the semi-structured interview guide, Annex 2). Two additional experts were recruited through this method as most participants recommended experts who were already included in the initial mapping, and data saturation was achieved in the ninth interview. This combined sampling approach ensured the inclusion of rich insights from key informants with a long trajectory of industry experience in Kenya’s ROTG market.

The final sample of nine experts is appropriate for this qualitative study. Methodological research states that data saturation can be achieved in small and information-rich populations, with frameworks showing that as few as six interviews may be sufficient to capture core themes in focused inquiries (Malterud et al., 2016). Consistent with this approach, prior studies in sustainability, public health and related fields have employed small samples (N < 9) where analytic depth and information power justified their use (Sauter et al., 2023; Puiu and Velickovic, 2025; Manninen and Makkonen, 2025).

Collecting data from key expert informants provides three methodological and ethical advantages in this research. First, interviewing expert informants allows access to rich, domain-specific and historical knowledge that is usually unavailable through consumer interviews, large-scale surveys or public datasets. Experts can reveal broader and cross-cutting dynamics about the market, user behaviors, and structural factors that are difficult to capture through standard methods (Patton, 2002; Bogner et al., 2009; Patton, 2015). This approach is useful to investigate behavioral and structural factors of a rapidly evolving market in which contextual understanding is vital. In addition, experts are also more equipped to capture nuanced perspectives that can be overlooked in group settings. Thus, conducting individual and private interviews help to mitigate both social desirability biases and dominance effects, which allows informants to express their views without the influence of peers (Kvale and Brinkmann, 2009).

The second advantage is that it mitigates placing additional psychological and financial burdens on a highly sensitive low-income group. The approach taken respects the time and circumstances of this vulnerable group by not asking them sensitive questions (about their feasibility to access, afford and use basic living necessities), disrupting their work schedules, or requiring them to travel to other locations to conduct the interview (Bogner et al., 2009; Rubin and Rubin, 2011). And third, doing interviews with experts provide more timely and actionable feedback to inform the development of more cost-effective local strategies to scale up in Kenya. This is especially informative in a sector characterized by rapid piloting, persistent experimentation, and continuous innovation (Patton, 2002; Bogner et al., 2009).

The data collection protocol employed a semi-structured interview guide organized in four sections (see Annex 1). Section A collected basic background data from the expert and the interviewer, and Section B obtained informed consent to collect participants’ data and record the interview. Section C asked questions related to the Theoretical Domain Framework, collecting information on the barriers and enablers to adopting reusable-on-the-go products among low-income consumers in Kenya. This section asked one specific question per TDF domain, and the answers to all questions were used to complement findings in all other domains when relevant data emerged during the interview. Section C also captured information related to customer experience, contextual realities of stores and suppliers, the success of past strategies to promote reusable behavior, and recommendations for future interventions. Section D ended the interview by allowing participants to ask questions and to refer to other potential expert informants. Data from all sections informed the cross-domain analysis when relevant data emerged.

All nine interviews were conducted in English or Swahili (or a combination of both in practice) and scheduled to last for a maximum of 90 min. Translation support was offered to participants who preferred another local language, although none of the experts required it. Interviews were done virtually (e.g., Google Meets) or in-person at various locations in Nairobi, Kenya. Every interview was audio-recorded with prior consent of the participant and transcribed to ensure accuracy and facilitate thematic analysis.

Semi-structured interviews were conducted in English, audio-recorded, and transcribed verbatim. Transcripts were anonymized (by CM and DN) and independently analyzed by two researchers (LAM and RD) to reduce potential biases and enhance objectivity. Data analysis was carried out using Excel as the main qualitative research tool, and it followed a two-phase coding process (Meyer and Avery, 2009; Bree and Gallagher, 2016). In the first phase, inductive codes were generated through reflexive thematic analysis, allowing themes to emerge organically from the data without predefined categories (Braun and Clarke, 2006; Clarke and Braun, 2017). In the second phase, deductive coding was applied using the TDF, with each domain (e.g., Knowledge) treated as a distinct code. The results from both phases were compared and discussed until consensus was achieved. Both inductive and deductive codes were mapped to the COM-B model to support a structured and theory-informed interpretation of the findings.

The study obtained ethical approval and informed consent from every participant prior to data collection. Ethical approval was granted by the National Commission for Science, Technology and Innovation (license No: NACOSTI/P/24/414319, issued on November 28, 2024). Informed consent for participation and audio recording was requested before the start of each interview, emphasizing each participant’s comprehension and autonomy to participate. Interviewers explained the purpose of the study, potential risks and benefits, and gave time to participants to ask for clarifications before seeking consent. Participants were informed of their right to withdraw at any point and were assured about the privacy and confidentiality of their responses. All personal identifiers were removed and replaced with unique codes to maintain anonymity before the analysis phase (e.g., Expert #1 to #9, see Table 2). Data was stored on a secured and password-protected platform that was accessible only to the research team. Audio recordings were made only with participants’ consent, having the option to pause or stop recording at any time.

As shown in Table 2, the nine key expert informants demonstrate extensive experience in Kenya’s reusable-on-the-go market. The average age of the experts is 39 years (ranging from 28 to 51) and 56% of them are men. Seven experts were residing in Kenya (78%) at the time of the interview; one was living in South Africa and one in The Netherlands. Collectively, the experts combine a total of 127 years of direct professional market experience in Kenyan reusable-on-the-go systems (with an average of 14 years per informant and a median of 9 years, ranging between 3 and 27 years). All informants currently hold senior roles in private companies and nonprofit organizations directly involved in the ROTG market including (dual) positions as Founder, Executive Director, CEO, Head of Impact and Partnerships, Senior Program Manager, Innovation Lead, and Product and Service Designer. In addition, they have experience in a wide range of fast-moving consumer goods that include personal and home care (e.g., powder soap, sunscreen, and detergent), beverages (e.g., water, and milk), food (e.g., salad, and oil), and household energy solutions (e.g., gas canisters). Six participants hold a bachelor’s degree (67%) and three obtained a postgraduate degree (33%).

This section presents the results of the study. Following the methodological approach, Figure 1 displays the inductively identified themes in italics and maps them to the Theoretical Domain Framework and the COM-B model (Cane et al., 2012).

The results advance the theoretical understanding of ROTG adoption by demonstrating how consumers’ capability, cognitive load, and social meaning jointly shape sustained reuse behavior in low-income contexts. While prior work on reuse systems has emphasized awareness, environmental attitudes, and infrastructure availability, the present findings show that these factors are insufficient when operational complexity and trust barriers remain high (Berger et al., 2022; Hong et al., 2024; Oliveira et al., 2024). Despite widespread awareness of plastic pollution and familiarity with the concept of reuse, Kenyan low-income consumers frequently disengage from ROTG systems due to difficulties operating resupply machines, concerns about hygiene and contamination, and the cumulative cognitive effort required to sustain repeated use. These findings reinforce and extend prior evidence showing that barriers to reuse adoption stem primarily from behavioral frictions in system design rather than from a lack of consumer motivation (Miao et al., 2023; Hong et al., 2024; Oliveira et al., 2024; Herweyers et al., 2024).

By highlighting the role of planning, remembering, cleaning, transporting, and resupplying containers, this study aligns with theories of scarcity in behavioral economics, which posit that poverty-related cognitive load reduces individuals’ capacity to engage in more effortful and future-oriented behaviors (Mani et al., 2013; Schilbach et al., 2016). In the Kenyan low-income market, single-use plastic packaging may function as a low-effort default, even when reuse options are financially or environmentally advantageous. The findings therefore contribute to theoretical debates by positioning ROTG adoption as a capability-constrained behavior rather than a purely preference-driven choice (Collis et al., 2023; Miao et al., 2023).

The study also extends theoretical insights on social identity and stigma in sustainable consumption. While reuse systems in Kenya are not perceived as inherently undesirable; rather, stigma emerges from design and branding choices that signal low socioeconomic status. This supports broader evidence across health, finance, and environmental domains showing that policies and interventions perceived as “for the poor” can undermine uptake by threatening dignity and social image (World Bank, 2015; Artavia-Mora, 2020). Importantly, this study shows that stigma is malleable and design-dependent, indicating that reuse adoption is shaped by functional attributes and symbolic meanings attached to systems, products and practices.

From a practical perspective, the findings suggest that successful ROTG systems in low-income markets must prioritize ease of use, trust, and dignity alongside affordability. Operational complexity emerged as a critical barrier (particularly for first-time users), which highlights the importance of minimizing the number of steps required to participate in reuse systems. Concerns about container cleanliness, especially for milk and edible oil, also highlight the need for designing solutions that reduce perceived contamination risk, cleaning difficulties and convenience. Without addressing these concerns, reuse systems are unlikely to achieve sustained adoption regardless of price incentives (Borg et al., 2022; Collis et al., 2023; Miao et al., 2023).

The study also demonstrates the effectiveness of experiential and interpersonal interventions. In-store demonstrations, brand ambassadors, and hands-on guidance help consumers navigate unfamiliar systems, reduce anxiety, and build procedural knowledge and trust. These findings are consistent with prior research showing that experiential learning, credible institutional signals, and point-of-purchase support are particularly effective in fostering behavioral change toward green systems (Sarkis et al., 2010; Berger et al., 2022; Buckel et al., 2024). Such interventions appear especially important in low-income settings, where consumers face very high opportunity costs and limited tolerance for risky experimentation. These findings parallel recent experimental evidence from Kenya showing that experiential and cost-salient interventions significantly increase uptake of reuse solutions by reducing risk and uncertainty at the moment of choice (Howell et al., 2025). Hence, ROTG systems should be embedded within broader market ecosystems that combine affordability with visible legitimacy, user support, and social normalization.

Age-related differences in adoption also have practical implications. Resistance among consumers over 35 reflects entrenched habits and lower familiarity with emerging technologies, whereas younger consumers show greater openness and flexibility. This pattern aligns with prior findings in reusable food and packaging markets, where younger segments demonstrate higher willingness to experiment with novel systems (Noëth et al., 2024; Hong et al., 2024). Segment-specific strategies, including targeted communication, simplified interfaces, and aspirational branding, may therefore be necessary to broaden adoption across demographic groups.

Situating these findings within the Kenyan and broader African context reveals important implications for sustainability transitions. Consistent with prior studies in Kenya, this research confirms a persistent awareness–action gap in plastic reduction efforts: high environmental awareness does not directly translate into sustained reuse when behavioral and structural barriers remain unaddressed, especially in lower-income consumer segments (Oguge et al., 2021; Getanda, 2023; Kenya Plastic Pact, 2026). Kenya’s ambitious plastic policies, including the plastic carrier bag ban and subsequent single-use restrictions, have raised public awareness but have achieved limited behavioral outcomes due to enforcement gaps and the continued availability of informal alternatives (Omondi and Asari, 2021; Oguge et al., 2021; Getanda, 2023). The present findings mirror this pattern, showing that awareness of ROTG systems is widespread, yet consistent adoption remains limited in the absence of supportive system design.

Affordability emerged as the dominant driver of adoption, reinforcing regional evidence that economic considerations outweigh environmental motivations in low-income settings characterized by income volatility and informality (Behuria, 2021; Dikgang et al., 2012). Moral and environmental appeals alone are insufficient to sustain behavior change when price differences and perceived financial risks persist (Hardisty and Weber, 2009; Bolderdijk and Steg, 2015; Berger et al., 2022; Scheller et al., 2024). These findings indicate that reuse systems must deliver immediate and tangible economic value if they are to compete effectively with single-use packaging in African markets. They also point to strong intertemporal discounting in low-income groups, where immediate affordability and liquidity constraints outweigh longer-term benefits (Hardisty and Weber, 2009; Scharff and Viscusi, 2011; Richards and Hamilton, 2012; Roewer et al., 2015).

The study also aligns with regional evidence highlighting the importance of behavioral frictions and normative cues. In South Africa, research on plastic bag levies and reuse initiatives shows that social norms, perceived legitimacy, and in-store cues often influence behavior more strongly than price signals alone (Hasson et al., 2007; Abiola et al., 2023; Barnes and Gihring, 2025). For instance, Abiola et al. (2023) found that non-monetary nudges and sustained awareness campaigns achieved more consistent increases in reusable bag uptake than subsidies alone, reinforcing the findings in Kenya that behavioral design elements can be more effective than limited price incentives set in isolation. Similar patterns are observed in emerging reuse initiatives across South Africa, where adoption increases with context-sensitive system design and consumer education (Barnes and Gihring, 2025). The present Kenyan findings therefore reflect a broader regional dynamic in which affordability must be complemented by behavioral design elements that reduce effort and normalize reuse as an everyday practice.

Stigma represents a particularly salient sustainability challenge. Across several African contexts, low-cost or informal solutions are often symbolically associated with poverty, which strongly discourages use in public spaces. The present study shows that stigma is not intrinsic to reuse but emerges from design choices that inadvertently reinforce low-status associations. This resonates with broader evidence that dignity and social image are central to behavior change in resource-constrained settings (World Bank, 2015; Artavia-Mora, 2020; Yamin et al., 2023). Designing reuse systems that are aspirational, modern, and socially valued becomes critical to scaling circular economy solutions in African markets.

Finally, evidence from West Africa suggests that reuse is not always a new concept in the region, but it can be rooted in long-standing informal practices of repurposing and resourcefulness. Initiatives such as Trashy Bags Africa in Ghana demonstrate how community-driven reuse can combine waste reduction with livelihood creation, suggesting that formal ROTG systems may benefit from aligning with existing cultural practices of reuse rather than positioning themselves as entirely new behaviors (Trashy Bags Africa, 2026). Leveraging these informal practices may help formal systems achieve greater legitimacy and adoption across diverse African contexts.