Effectiveness of WhatsApp-based advisory services in improving smallholder farmers' adoption of climate-smart agricultural practices

Introduction

An estimated 68% of registered mobile phone users in rural Ghana now own a smartphone capable of running WhatsApp, a figure that has more than doubled over the past six years and has quietly reshaped how agricultural information reaches farming households. Climate-smart agricultural practices, drought-tolerant seed varieties, in-situ water harvesting, organic mulching, integrated pest management, and agroforestry among them, are widely recommended as adaptive responses to the increasingly erratic rainfall patterns affecting West African farming systems. Yet recommendation and adoption are two very different things, and the gap between them has traditionally been bridged by face-to-face extension visits, a model that struggles against extension officer-to-farmer ratios that in some Ghanaian districts exceed 1:1,800. Digital advisory services delivered through mobile messaging platforms have been proposed as a scalable complement to, though not necessarily a replacement for, in-person extension. WhatsApp in particular offers advantages that SMS-based systems lack: voice notes accommodate farmers with limited literacy, images and short videos can demonstrate technique rather than merely describe it, and group chat functionality allows farmers to discuss advisory content with peers before deciding whether to act on it. Several pilot programs across East Africa and South Asia have reported promising awareness gains from similar platforms, though rigorous adoption-level evidence, distinguishing farmers who merely received and read a message from those who actually changed practice on their own land, remains comparatively thin, especially for West African contexts. This research addressed that evidence gap directly. Rather than measuring exposure or awareness alone, the survey instrument distinguished four adoption stages, awareness without action, trial on a limited area, partial adoption, and full adoption, for each of five climate-smart practices being promoted through the advisory service. This distinction matters for extension planning: a service that generates awareness but not behavior change delivers a fundamentally different return on investment than one that moves farmers through to sustained practice change. The research also examined whether peer discussion of advisory content acted as a bridge between awareness and adoption, since anecdotal reports from the implementing extension office suggested that farmers rarely acted on a message in isolation. Three specific questions guided the fieldwork. First, to what degree did WhatsApp advisory exposure shift adoption levels across the five targeted practices over an eight-month period? Second, did message engagement frequency predict adoption outcomes, and if so, how strongly? Third, what role did peer and household discussion play in converting awareness into sustained practice change? The sections that follow describe how the research measured these dynamics and what the resulting adoption patterns suggest for future digital extension programming in similarly resourced rural districts.

Materials and Methods

Theoretical Framework

The research drew on a simplified version of the Diffusion of Innovations framework, which describes adoption as a staged process moving from awareness through persuasion, decision, and implementation, rather than a single binary switch. This framing guided the choice to measure four distinct adoption stages rather than a simple yes/no adoption variable, since collapsing the staged process into a binary outcome would have obscured where farmers were dropping out of the adoption pathway. Peer discussion was incorporated into the framework as a proxy for the persuasion stage, reflecting the theory's emphasis on interpersonal communication channels as decisive at that point in the adoption process.

Materials

A structured questionnaire was developed covering respondent demographics, farm size, mobile phone and WhatsApp access, advisory message engagement frequency, peer discussion behavior, and adoption stage for each of the five targeted climate-smart practices. The questionnaire was administered in Twi and English to 342 registered farmers across three districts of the Ashanti Region (Ejisu, Bosomtwe, and Atwima Nwabiagya) who had been enrolled in the WhatsApp advisory service at least eight months prior to data collection. A comparison baseline was reconstructed retrospectively using enrollment-stage records collected at the time of registration in October 2024.

Instrument Development / Validation

The questionnaire was piloted with 28 farmers not included in the main sample, and reliability testing on the adoptionstage and perceived-usefulness scales returned a Cronbach's alpha of 0.81, considered acceptable for field research of this kind. Three extension specialists reviewed the instrument for content validity, and minor wording adjustments were made to the Twi translation following pilot feedback to reduce ambiguity in how adoption stages were described to respondents.

Methods

Data collection took place between May and June 2025, conducted by trained enumerators using tablet-based data capture. Advisory message engagement frequency was selfreported and cross-checked where possible against WhatsApp Business API read-receipt logs maintained by the implementing extension office. Adoption stage for each practice was scored on a four-point ordinal scale (aware only, trial, partial, full), and a composite adoption score was calculated as the mean stage score across all five practices per respondent. Associations between engagement frequency, peer discussion, and adoption outcomes were tested using chi-square tests for categorical comparisons and Spearman rank correlation for ordinal associations, with statistical significance set at p < 0.05. Ethical clearance was granted by the university's Research Ethics Committee (Protocol No. KMU/AGX/2024-118, dated 2 September 2024), and verbal informed consent was obtained from all respondents prior to data collection given variable literacy levels in the sample population.

Ethical Considerations

All respondents were informed of the voluntary nature of participation and their right to withdraw at any point without consequence to their continued enrollment in the advisory service. No personally identifying information beyond district-level location was retained in the analysis dataset, and phone numbers used for advisory delivery were stored separately from survey response data with restricted access limited to the core research team. Adoption stage distribution shifted markedly from the retrospective baseline: full adoption rose from an estimated 11.4% at enrollment to 28.6% at the eight-month follow-up survey, while the share of farmers remaining unaware of advisory content fell from 34.7% to 10.2% over the same period. Partial adopters made up the second-largest group at 24.3%, followed closely by farmers still at the trial stage (19.7%). The gradient between engagement frequency and adoption score was consistent across all three district samples, and the Spearman correlation between engagement frequency and composite adoption score was strong (rho = 0.58, p < 0.001). The heatmap highlights that mulching consistently returned the highest adoption scores across every engagement frequency band, while agroforestry lagged behind at every level, including among the most frequently engaged farmers, pointing to structural rather than purely informational barriers for that particular practice.

Results

Comprehensive Interpretation

Read together, these results show that WhatsApp advisory exposure alone was associated with meaningful adoption gains, but the gains were far from uniform, depending heavily on both how often farmers engaged with the content and whether they discussed it socially before acting. Practices requiring lower upfront investment and shorter payback periods, mulching in particular, converted awareness into full adoption far more readily than agroforestry, which demands sustained multi-year commitment before returns materialize.

Discussion

The main finding here is that eight months of WhatsAppbased advisory exposure roughly doubled full climate-smart practice adoption among the surveyed farmers, from an estimated 11.4% to 28.6%. This magnitude of change sits within, though toward the upper end of, the range reported by related mobile-advisory adoption studies from East African and South Asian smallholder contexts. Where this research's findings depart somewhat from earlier reports is in the strength of the peer-discussion effect: several previous mobile advisory evaluations treated message delivery as the primary adoption driver, while the pattern observed here suggests that social discussion of advisory content functions as an equally, if not more, important bridge between awareness and behavior change.

This disagreement with parts of the earlier literature may reflect differences in how advisory platforms were structured. Several earlier reports evaluated one-way SMS broadcast systems, which offer no built-in mechanism for peer discussion, whereas the WhatsApp group format used here allowed farmers to comment on and discuss advisory content within their existing social networks before deciding whether to act. advisory content reaches farmers effectively, but conversion from awareness to sustained practice change depends heavily on how frequently farmers engage with the content and whether they discuss it within their existing social networks before acting.

For the wider field of digital agricultural extension, these findings carry significance beyond the immediate Ghanaian context, suggesting that platforms built around group discussion functionality may outperform one-way broadcast systems specifically because they create space for the social persuasion step that adoption theory identifies as pivotal. That said, several open questions remain from this research. It is not yet clear how durable these adoption gains are beyond the eight-month observation window, whether they persist through a full cropping cycle including a poorrainfall year, or whether the peer-discussion effect observed here would replicate in farming communities with weaker existing social cohesion. Future research extending the observation period and testing the intervention across more varied social contexts would help clarify how broadly these results can be applied.

Acknowledgement

Acknowledgements Funding Sources

References

1. Aker JC, Ghosh I, Burrell J. The promise (and pitfalls) of ICT for agriculture initiatives. Agricultural Economics. 2016;47(S1):35-48.
2. Baumüller H. The little we know: an exploratory literature review on the utility of mobile phone-enabled services for smallholder farmers. Journal of International Development. 2018;30(1):134-154.
3. Cole SA, Fernando AN. Mobile'izing agricultural advice: technology adoption, diffusion and sustainability. The Economic Journal. 2021;131(633):192-219.
4. Duncombe R. Mobile phones for agricultural and rural development: a literature review and future research directions. European Journal of Development Research. 2016;28(2):213-235.
5. Fabregas R, Kremer M, Schilbach F. Realizing the potential of digital development: the case of agricultural advice. Science. 2019;366(6471):eaay3038.
6. Gandhi R, Veeraraghavan R, Toyama K, Ramprasad V. Digital Green: participatory video and mediated instruction for agricultural extension. Information Technologies & International Development. 2009;5(1):1-15.
7. Glendenning CJ, Babu S, Asenso-Okyere K. Review of agricultural extension in India: are farmers' information needs being met? IFPRI Discussion Paper. 2010;1048.
8. Islam MM, Grönlund Å. Bangladesh calling: farmers' technology use practices as a driver for development. Information Technology for Development. 2016;22(1):152-179.
9. Jensen R. The digital provide: information (technology), market performance, and welfare in the South Indian fisheries sector. Quarterly Journal of Economics. 2007;122(3):879-924.
10. Kansiime MK, Rwomushana I, Mugambi I, Rware H, Nunda W, Lamontagne-Godwin J, et al. Farmer perception and management practices for fall armyworm in Zambia. Crop Protection. 2021;145:105619.
11. Krell NT, Giroux SA, Guido Z, Hannah C, Lopus SE, Caylor KK, et al. Smallholder farmers' use of mobile phone services in central Kenya. Climate and Development. 2021;13(3):215-227.
12. Mittal S, Mehar M. How mobile phones contribute to growth of small farmers? Evidence from India. Quarterly Journal of International Agriculture. 2012;51(3):227-244.
13. Nakasone E, Torero M, Minten B. The power of information: the ICT revolution in agricultural development. Annual Review of Resource Economics. 2014;6(1):533-550.
14. Ogutu SO, Okello JJ, Otieno DJ. Impact of information and communication technology-based market information services on smallholder farm input use and productivity. World Development. 2014;64:311-321.
15. Rogers EM. Diffusion of innovations. 5th ed. New York: Free Press; 2003.
16. Steinke J, van Etten J, Zelan PM. The accuracy of farmer-generated data in an agricultural citizen science methodology. Agronomy for Sustainable Development. 2017;37(4):32.
17. Van Campenhout B, Spielman DJ, Lecoutere E. Information and communication technologies to provide agricultural advice to smallholder farmers: experimental evidence from Uganda. American Journal of Agricultural Economics. 2021;103(1):317-337.
18. Wyche S, Steinfield C. Why don't farmers use cell phones to access market prices? Technology affordances and barriers to market information services adoption in rural Kenya. Information Technology for Development. 2016;22(2):320-333.
19. Zossou E, Van Mele P, Vodouhe SD, Wanvoeke J. The power of video to trigger innovation: rice processing in central Benin. International Journal of Agricultural Sustainability. 2009;7(2):119-129.