Behavioral Intentions of Online Zakat Users in Indonesia: An Extended TAM with Trust.
Security, and Local Cultural Perspectives Ade Sofyan Mulazid1*.
Irhamsyah Putra2.
Fauzan3.
Deni Pandu Nugraha4 Faculty of Economics and Business.
Universitas Islam Negeri Syarif Hidayatullah Jakarta Indonesia Faculty of Sharia Universitas Islam Negeri Fatmawati Sukarno Bengkulu Indonesia Hungarian University.
Hungaria Correspondency Email: adesofyanmulazid@uinjkt.
Article Info Article History:
Received Reviewed Accepted Published : 27 April 2026 : 11 May 2026 : 9 June 2026 : 13 June 2026 Keywords:
online zakat.
acceptance model.
local culture DOI:
https://doi.
org/10.
JEL:
G29.
L86.
Z12
Abstract This study aims to examine the determinants of behavioral intention among online zakat service users in Indonesia by extending the Technology Acceptance Model (TAM) with the variables of trust and security, and by interpreting the results through the lens of Indonesian local culture.
The rapid digitalization of religious philanthropy in Indonesia has expanded the use of online zakat platforms, yet adoption remains uneven and is shaped by concerns that go beyond the classical TAM.
Data was collected through an online survey distributed to 120 respondents who were selected by purposive sampling from among users of online zakat The data were analyzed using Partial Least Squares Structural Equation Modelling (PLS-SEM) with reliability, convergent and discriminant validity, and common method bias diagnostics.
The results indicate that perceived usefulness, security, and behavioral intention significantly affect adoption, while perceived ease of use shows an unexpected negative effect and trust shows only a marginal The model explains 38.
2 per cent of variance in behavioral intentions.
Read against Indonesian patterns of collectivism, religious authority, and gotong royong, the findings suggest that usersAo priorities tangible benefits and transactional security over interface simplicity, and they call on zakat institutions to combine platform usefulness and data protection with ulama-led and community-based trust-building strategies to widen digital zakat adoption in Indonesia.
International Journal of Islamic Business and Economics (IJIBEC), 10 .
June 2026, 43-55 Introduction The acceleration of digital adoption in Indonesia has reshaped the way Muslims fulfil religious obligations, including the payment of zakat.
Internet penetration in Indonesia has reached around 77 per cent of the population, with most users transacting online (We Are Social & Meltwater, 2.
, and digital zakat platforms have grown in parallel.
Online services widen access, raise accountability, and enable real-time tracking of charitable funds.
At the same time, a body of empirical work on technology adoption shows that the willingness to use a new digital service is rarely a function of interface alone (Davis, 1.
In the context of religiously oriented financial transactions, perceived risk and trust often weigh more heavily than usability, particularly in developing countries where digital trust infrastructure is still maturing (Ajzen & Fishbein, 1.
The Technology Acceptance Model (TAM), introduced by Davis in 1989, has been the dominant lens through which researchers explain user acceptance of information systems.
TAM
proposes that perceived usefulness and perceived ease of use determine behavioral intention, which in turn drives actual use.
The model has been productively applied across e-commerce, internet banking, and digital health, but several scholars argue that the original two-variable model is insufficient when the transaction involves financial value, sensitive personal data, or religious Gefen, et al.
showed that trust must be integrated into TAM to explain online shopping behavior, and Pavlou .
extended that argument by treating trust as a moderator of acceptance in e-commerce contexts.
Online zakat platforms inherit features of e-commerce, including digital payment, customer interface, and transactional records, but operate under additional normative pressures.
The funds are religiously consecrated, transparency expectations are higher, and trust failures carry both legal and reputational consequences.
Abdullah and Suhaimi .
found that transparency and accountability shape the intention to use online zakat platforms more strongly than technical features alone, while Yousafzai.
Pallister, and Foxall .
demonstrated that perceived security has a direct positive effect on the adoption of mobile banking technology.
Building on this line.
Kurniawan and Sutrisno .
reported that the perception of security significantly drives the intention to transact through online zakat platforms in Indonesia, especially in relation to the protection of personal and financial data.
Despite these contributions, four gaps remain.
First, much of the technology-acceptance literature in Indonesia focuses on commercial digital services rather than on religious philanthropy, and therefore misses the social, cultural, and theological pressures that shape zakat behavior (Masrom, 2011.
Venkatesh.
Thong, & Xu, 2.
Second, while Gefen and colleagues have integrated trust into TAM, fewer studies pair trust and security as joint extensions in the zakat domain, even though both factors are theoretically and empirically distinct (Gefen et al.
, 2.
Third, although AlMaghrabi and Dennis .
stress that culture and religiosity moderate adoption, and Saputra and Prasetyo .
have applied TAM to online zakat at a local scale, no single national-scale study integrates trust, security, and the standard TAM constructs in a unified PLS-SEM model.
Fourth, the existing TAM-zakat literature rarely interprets its findings through the lens of distinctive Indonesian cultural patterns, namely collectivism, high power distance, the religious authority of ulama and kyai, and the ethic of gotong royong (Hofstede.
Hofstede, & Minkov, 2010.
Geertz, 1960.
Hefner, 2.
, even though such patterns are known to shape technology adoption across cultures (Straub.
Keil, & Brenner, 1.
The Indonesian case illustrates these gaps in practical terms.
The national zakat potential is estimated at more than IDR 300 trillion per year, yet actual collection through formal channels remains a fraction of that figure, hovering below IDR 30 trillion annually (BAZNAS, 2.
significant share of muzakki still prefer to pay zakat directly to local mosques, pesantren, or individual mustahiq rather than through digital platforms, even though licensed amil zakat institutions such as BAZNAS.
Dompet Dhuafa, and Rumah Zakat have invested heavily in online User surveys consistently report that concerns about data security, doubts about whether the platform is supervised by credible religious authorities, and unfamiliarity with the disbursement International Journal of Islamic Business and Economics (IJIBEC), 10 .
June 2026, 43-55 process deter adoption more than interface difficulty does.
In several documented cases, public trust in digital zakat was shaken when platform operators failed to publish timely disbursement reports or when media reports questioned the shariah compliance of fund allocation.
These empirical realities underline the need for a model that goes beyond perceived usefulness and ease of use to incorporate the trust and security dimensions that are salient in the Indonesian religiousphilanthropic context.
Trust in online financial services Trust is a recurring determinant of adoption in online environments, where users surrender sensitive information without face-to-face interaction.
Gefen, et al.
treat trust as a precondition for the meaningful use of TAM in e-commerce, and McKnight.
Choudhury, and Kacmar .
disaggregate trust into beliefs about reliability, competence, and integrity that together shape transactional In the context of online zakat, trust extends beyond the platform to encompass the institutional integrity of the zakat manager .
, the audit chain of disbursement, and the alignment of operations with shariah norms.
Security in online transactions Security captures the user's belief that the platform protects against unauthorized access, fraud, and data misuse.
Pavlou .
demonstrates that perceived security reduces transactional uncertainty in e-commerce, and Kurniawan and Sutrisno .
show that the same logic applies to online zakat.
When users are confident that their data and contributions are safe, they are markedly more willing to transact through the platform.
Behavioural intention and actual use Behavioral intention occupies a central place in TAM and its extensions.
Venkatesh and Davis .
argue that intention is a strong proximal predictor of system use, and Venkatesh.
Morris.
Davis, and Davis .
confirm this finding across organizational settings.
In the present study, behavioral intention serves as a mediator linking the four TAM constructs .
erceived usefulness, perceived ease of use, trust, securit.
to the actual use of online zakat services.
Indonesian local cultural context for digital zakat Indonesia presents a culturally distinctive setting for the study of digital adoption.
Hofstede's cultural-dimension model places Indonesian society in the cluster of strongly collectivist .
ndividualism index of .
, high-power-distance .
, and moderately uncertainty-avoiding societies (Hofstede.
Hofstede, & Minkov, 2.
Three implications follow for online zakat behavior.
First, in collectivist settings, technology adoption travels along social and family networks rather than through individual exploration, so peer endorsement and the recommendation of community figures weigh heavily on intention (Straub.
Keil, & Brenner, 1.
Second, the high-power-distance ranking elevates the role of religious authorities such as ulama, kyai, and ustadz in legitimizing platforms and shaping perceptions of institutional integrity.
Third, when the transacted good is religiously consecrated, the threshold for tolerable uncertainty falls sharply, raising the salience of visible security features over interface ease.
Beyond the Hofstede frame, three indigenous cultural elements further bear on online zakat The principle of gotong royong, which translates the value of cooperation into everyday practice, frames zakat as a communal religious-economic obligation rather than a private individual This in turn nudges users to favor platforms that visibly tie individual contributions to collective Long-standing pesantren and mosque-based zakat collection networks supply a dense traditional infrastructure that competes with, and sometimes complements, digital platforms (Beik & Arsyianti, 2.
The Shafi'i jurisprudential mainstream and the everyday religiosity documented in the anthropology of Indonesian Islam (Geertz, 1960.
Hefner, 2.
further generate normative expectations that the online disbursement chain remains transparent, auditable, and aligned with shariah practice.
These cultural patterns, taken together, frame how Indonesian users read the four extended-TAM constructs in the present study.
This study addresses those gaps by testing an extended TAM in which perceived usefulness and perceived ease of use are augmented with trust and security, and by interpreting the resulting International Journal of Islamic Business and Economics (IJIBEC), 10 .
June 2026, 43-55 coefficients through the lens of Indonesian local culture.
The contribution is threefold.
Theoretically, the paper extends TAM into the religious-philanthropy domain and shows how the relative weight of its variables shifts when the transaction is ethically charged.
Culturally, it reads the model coefficients against indigenous Indonesian patterns, including collectivism, religious authority, and gotong royong, which gives the statistical findings a substantive interpretive depth.
Practically, the findings give zakat institutions and platform developers a calibrated picture of which design and governance investments translate into wider adoption.
The objectives of this research are to identify the determinants of behavioral intention among online zakat users in Indonesia, to measure the relative weight of trust and security against the standard TAM constructs, to interpret the resulting pattern through Indonesian cultural frames, and to derive practical recommendations for zakat institutions and platform developers.
Method This study uses a quantitative cross-sectional design with an online survey administered through Google Forms.
The target population consists of Indonesian users of online zakat services.
Respondents were drawn through non-probability purposive sampling so that only users with firsthand experience of an online zakat platform were included (Hair.
Hult.
Ringle, & Sarstedt, 2.
total of 120 valid responses were obtained, which exceeds the minimum sample size required for PLS-SEM with the present model complexity.
The target population of this study comprises all Indonesian Muslims who have used at least one online zakat platform to pay zakat, infaq, or sadaqah within the twelve months preceding data The sampling frame was accessed through social media networks and the mailing lists of three nationally licensed amil zakat institutions.
A non-probability purposive sampling technique was applied with two inclusion criteria: .
the respondent must be an Indonesian citizen aged 18 years or older, and .
the respondent must have completed at least one zakat payment through an online The questionnaire link was distributed through WhatsApp groups.
Instagram, and the institutional websites of the partnering zakat organizations during a four-week data collection window in 2023.
A total of 138 responses were received, of which 18 were excluded because the respondents did not meet the inclusion criteria or left more than ten per cent of items unanswered, yielding 120 valid responses for analysis.
This sample size satisfies the minimum threshold for PLSSEM, which Hair.
Hult.
Ringle, and Sarstedt .
set at ten times the largest number of structural paths directed at any latent variable in the model.
Data were analyzed with Partial Least Squares Structural Equation Modelling (PLS-SEM) because the technique handles non-normal data, accommodates relatively small samples, and is robust for complex models with multiple latent variables (Henseler.
Ringle, & Sarstedt, 2015.
Chin.
Reliability and validity were tested through Cronbach's alpha.
Composite Reliability.
Average Variance Extracted (AVE), and the Heterotrait-Monotrait Ratio of Correlations (HTMT) (Hair.
Risher.
Sarstedt, & Ringle, 2.
Common Method Bias (CMB) was screened using the Full Collinearity Variance Inflation Factor (FCVIF) procedure proposed by Kock .
, with a threshold of 3.
The structural model was estimated through bootstrapping with 1,000 resamples to obtain stable tstatistics.
Five constructs were measured with multi-item Likert-type scales adapted from prior validated instruments: perceived usefulness, perceived ease of use, trust, security, and behavioral intention, alongside actual use.
Items with outer loadings below 0.
4 were eliminated, in line with the recommendation of Hair and colleagues.
Discriminant validity was checked through the Fornelarcker criterion (Fornell & Larcker, 1.
, complemented by HTMT, which is more sensitive to subtle violations than the classical criterion.
Result and Discussion Respondent profile and measurement model Of the 120 respondents who completed the questionnaire, 45 per cent fell within the 25 to 34 age band, 52 per cent were men, and 65 per cent held a tertiary qualification (S1 or S.
About 40 per International Journal of Islamic Business and Economics (IJIBEC), 10 .
June 2026, 43-55 cent worked in the private sector and 30 per cent were civil servants, and 55 per cent lived in the major urban areas of Jakarta.
Surabaya, and Bandung.
The composition is consistent with the demographic profile of Indonesia's digital zakat user base, which leans towards educated urban The respondent profile is summarized in Table 1.
Table 1 Respondent Profile Profile Classification Age 25 to 34 years old Gender Male Female Education Higher education (S1.
Profession Private sector employee Civil servant Urban areas (Jakarta.
Surabaya.
Bandun.
Other regions Geography Source: primary data, processed .
Reliability was assessed with Cronbach's alpha.
All six constructs exceeded the 0.
90 threshold, which indicates excellent internal consistency, as reported in Table 2.
Table 2 Internal Consistency Reliability Testing Variable Remarks Current Use Excellent Behavioral Intention Excellent Perceived Ease of Use Excellent Perceived Usefulness Excellent Security Excellent Trust Excellent Source: primary data, processed using SmartPLS .
Convergent validity was inspected through outer loadings.
Composite Reliability, and AVE.
Most indicators loaded above 0.
70 on their assigned construct.
The exception is Perceived Ease of Use, where indicators PEU2.
PEU3, and PEU4 loaded below 0.
70, with PEU3 falling below the 0.
elimination threshold.
This drove the construct's AVE down to 0.
3087, which is below the Fornelarcker recommended minimum of 0.
The authors are therefore encouraged to revisit the PEU items, drop indicators below 0.
40, and re-estimate the model before publication.
The remaining constructs satisfy the AVE > 0.
50 rule, as detailed in Table 3.
Table 3
Convergent Validity Testing Variable Behavioral Intention
Indicator Loading AVE
BI1
BI2
International Journal of Islamic Business and Economics (IJIBEC), 10 .
June 2026, 43-55
Variable Perceived Ease of Use Perceived Usefulness Security Trust
Current Use Indicator Loading BI3
BI4
BI5
PEU1
PEU2
PEU3
PEU4
PEU5
PU1
PU2
PU3
PU4
PU5
PU6
SC1
SC2
SC3
SC4
SC5
SC6
SC7
TR1
TR2
TR3
TR4
TR5
AU1
AU2
AU3
AU4
AU5
AVE
International Journal of Islamic Business and Economics (IJIBEC), 10 .
June 2026, 43-55 Source: primary data, processed using SmartPLS .
Discriminant validity was checked first through the Fornell-Larcker criterion, where the diagonal values .
quare root of AVE) exceed the inter-construct correlations in the same column (Henseler.
Ringle, & Sinkovics, 2.
The result is presented in Table 4.
Table 4 Discriminant Validity (Fornell-Larcke.
Variable Behavioral Intention Perceived Ease of Use Oe0.
Perceived Usefulness Security Trust Oe0.
Current Use Oe0.
Source: primary data, processed using SmartPLS .
Because the Fornell-Larcker criterion can fail to detect subtle discriminant validity issues (Franke & Sarstedt, 2.
, the analysis was complemented by the HTMT ratio (Gefen.
Straub, & Boudreau, 2.
All HTMT values fall below the 0.
85 threshold for conceptually distinct constructs, which confirms discriminant validity, as shown in Table 5.
Table 5 Discriminant Validity (HTMT) Variable Behavioral Intention Ai Perceived Ease of Use Ai Perceived Usefulness Ai Security Ai Trust Ai Current Use Source: primary data, processed using SmartPLS .
Structural Model and Hypothesis Testing The R-square value for Behavioral Intention is 0.
3823 (Adjusted RA = 0.
, which Chin .
classifies as moderate explanatory power.
In contrast, the R-square for Actual Use is only 0.
(Adjusted RA = 0.
, which is in the weak range and suggests that variables outside the present model carry significant additional explanatory weight on actual use behavior.
The R-square results are summarized in Table 6.
Table 6 R-Square Results Dependent Variable R-Square R-Square Adjusted Status Current Use Weak Behavioral Intention Moderate Source: primary data, processed using SmartPLS .
International Journal of Islamic Business and Economics (IJIBEC), 10 .
June 2026, 43-55 Common Method Bias was screened with the Full Collinearity VIF approach.
The FCVIF for Actual Use is 1.
075 and for behavioral Intention is 1.
Both values lie well below the 3.
3 threshold proposed by Kock and Lynn .
, indicating that the data are not contaminated by significant common method bias.
The substantive concern raised by Podsakoff.
MacKenzie.
Lee, and Podsakoff .
, namely that single-source self-report data inflate shared variance, is therefore not borne out in this dataset.
Hypothesis testing was conducted with bootstrapping .
,000 resample.
The path coefficients, standard errors, and t-statistics are reported in Table 7.
Table 7 Hypothesis Testing Results Hypothesis Status Behavioral Intention Ie Actual Use Significant Perceived Ease of Use Ie Behavioral Intention Oe0.
Oe2.
Significant Perceived Usefulness Ie Behavioral Intention Significant Security Ie Behavioral Intention Significant Trust Ie Behavioral Intention Marginal .
%) Source: primary data, processed using SmartPLS .
Discussion The structural results converge on a coherent picture of online zakat adoption in Indonesia.
behavioral intention translates into actual use ( = 0.
2638, t = 2.
, which confirms TAMAos mediating logic but with a relatively modest magnitude.
The largest direct effect on intention is Perceived Usefulness ( = 0.
3726, t = 3.
, suggesting that Indonesian online zakat users adopt a platform primarily when they believe it improves the efficiency and accountability of their religious In practical terms, this encompasses the speed of payment processing, the availability of electronic receipts, and confirmation that funds have reached the intended beneficiaries.
Two findings depart from standard TAM expectations.
First.
Perceived Ease of Use carries a negative path coefficient ( = Oe0.
4654, t = Oe2.
The most plausible interpretation is that a segment of the sample associateAos simple zakat platforms with limited functional depth.
Users who value detailed nisab calculation, fund earmarking across asnaf categories, and auditable disbursement reporting perceive overly simplified interfaces as a signal of thin institutional capacity.
The negative sign does not refute TAM but indicates that ease of use operates through a different cognitive route in religiously charged transactions, where richer disclosure outweighs interface minimalism.
Second.
Trust shows only a marginal effect ( = 0.
2771, t = 1.
78, significant at the 10 per cent leve.
, in contrast to Security, which exerts a robust positive effect ( = 0.
3129, t = 2.
In the present sample, users appear to anchor their willingness to transact in concrete, verifiable security features such as encryption, authentication, and transaction logs, while institutional trust functions as a softer, slower-moving factor that has not yet fully consolidated in the Indonesian online zakat Read jointly, the four extended-TAM coefficients suggest that online zakat adoption in Indonesia is currently driven by a benefit-and-protection logic rather than a comfort-and-familiarity The four coefficients can be read more sharply when placed against Indonesian cultural The negative coefficient on perceived ease of use is consistent with a collectivist preference for rich, community-validated features over individual interface simplicity (Hofstede.
Hofstede, & Minkov, 2010.
Srite & Karahanna, 2.
In a setting where zakat is practiced within the family, the mosque, and the neighborhood, an overly streamlined interface can read as a stripping away of the institutional and shariah cues that users expect.
The robust positive effect of perceived International Journal of Islamic Business and Economics (IJIBEC), 10 .
June 2026, 43-55 security tracks the high uncertainty avoidance that surrounds religiously consecrated funds.
Within ShafiAoi jurisprudential expectations and the everyday religiosity that characterizes Indonesian Muslim practice (Hefner, 2.
, users expect explicit assurance that zakat reaches the eight asnaf without intermediation by unverifiable digital actors.
The marginal effect of trust is also legible through the local cultural frame.
Institutional trust in collectivist and high-power-distance societies is rarely a property of the platform alone.
it is endorsed and transferred through ulama, kyai, and family networks.
Until those endorsements are visibly inscribed inside the digital channel, the platform sits at the edge of the userAos circle of The path coefficient of 0.
2771 may therefore reflect an institutional trust that is still in transition rather than absent.
The principle of gotong royong adds a further interpretive layer: zakat is not only a private religious obligation but also a communal economic mechanism that knits muzakki and mustahiq into a circuit of solidarity.
Platforms that visibly aggregate individual contributions into communal outcomes are therefore likely to align more closely with usersAo cultural expectations than platforms that frame giving as a purely individual transaction (Beik & Arsyianti.
The positive and dominant effect of perceived usefulness is consistent with the original TAM proposition (Davis, 1.
and with recent applications in Islamic fintech.
Usman.
Chairy, and Projo .
& Alfani.
Yuniarto, & Handrito .
similarly reported that perceived usefulness is the strongest predictor of intention to use Islamic mobile banking in Indonesia, outweighing both ease of use and subjective norms.
Likewise.
Kasri and Yuniar .
found that functional benefit perceptions drive the willingness to use digital zakat services among Indonesian millennials, a finding that resonates with the present result.
The negative coefficient on perceived ease of use diverges from the mainstream TAM literature but is not unprecedented.
Alalwan.
Dwivedi, and Rana .
observed that in Jordanian mobile banking, ease of use became non-significant once trust and risk entered the model, and Venkatesh .
noted that the salience of ease-of-use declines as the normative stakes of the transaction rise.
Saputra and Prasetyo .
applied TAM to online zakat at a local Indonesian scale and found ease of use to be positive but weak, which is directionally aligned with the present finding that users of religiously charged platforms weight functional depth over interface simplicity.
The robust effect of security and the marginal effect of trust mirror the pattern reported by Amin.
Hamid.
Lada, and Anis .
in Malaysian Islamic internet banking and by Kurniawan and Sutrisno .
in the Indonesian online zakat context.
Pavlou .
and Gefen.
Karahanna, and Straub .
treat trust and security as conceptually adjacent but functionally distinct, and the present data confirm that distinction: users anchor their willingness to transact in concrete, verifiable security features before institutional trust has fully matured.
The cultural moderation of this pattern is consistent with Srite and Karahanna .
, who demonstrated that collectivism and uncertainty avoidance moderate the relative strength of TAM constructs across national settings.
Taken together, the findings suggest that online zakat in Indonesia functions less as a standalone digital service and more as a digital extension of an inherited communal institution.
Designers and zakat managers therefore face a clear ordering of priorities.
The priority is to increase functional usefulness through richer reporting, integrated payment channels, and traceable disbursement The second is to visibly strengthen security through advanced encryption, two-factor authentication, and audited security disclosures.
The third is to build trust through institutional transparency and partnership with religious authorities whose endorsement carries weight in collectivist, high-power-distance settings.
These technical interventions need to be paired with culturally embedded strategies.
Visible endorsement by ulama and kiyai, integration with mosque- and pesantren-based collection networks, and the framing of digital giving as an extension of gotong royong are likely to amplify the impact of platform investments on user adoption.
The R-square for behavioural intention .
indicates moderate explanatory power, meaning that roughly 62 per cent of variance remains unexplained and invites future research to incorporate additional predictors such as religiosity.
International Journal of Islamic Business and Economics (IJIBEC), 10 .
June 2026, 43-55 digital literacy, and social influence.
The weak R-square for actual use .
further signals that the pathway from intention to behavior is mediated by contextual factors, including access to digital infrastructure, income, and competing offline zakat channels, that the present model does not Conclusion This study extended the Technology Acceptance Model with trust and security to identify the determinants of behavioral intention among Indonesian users of online zakat platforms and interpreted the resulting pattern through the lens of Indonesian local culture.
Three substantive conclusions emerge.
First, perceived usefulness and perceived security are the strongest determinants of behavioural intention, and behavioural intention is in turn a significant predictor of actual use.
Second, perceived ease of use exerts a negative effect on intention, which suggests that in religiously charged transactions users associate richer disclosure and functional depth with institutional capacity, and read overly simple interfaces as a signal of thin governance.
Third, trust shows only a marginal effect, indicating that institutional trust in online zakat is still being built and currently operates as a slower-moving variable than transactional security.
Read against Indonesian patterns of collectivism, religious authority, and gotong royong, this asymmetric pattern reflects a market in which digital zakat operates as an extension of inherited communal institutions rather than as a stand-alone digital service.
Practically, the results recommend a sequenced intervention for zakat institutions and platform developers.
Investment in functional usefulness, namely real-time tracking, reporting, and integration with widely used payment rails, delivers the largest gain.
Visible security investments, with transparent communication of encryption and audit practices, come second.
Trust-building through accountability publications, partnership with credible religious authorities, and accessible financial reporting then sustains adoption over the longer term.
Read through the Indonesian cultural lens, these technical interventions need to be paired with culturally embedded strategies.
Visible endorsement by ulama and kiyai, integration with mosque- and pesantren-based collection networks, and the framing of digital giving as an extension of gotong royong are likely to amplify the impact of platform investments on user adoption.
Educational campaigns via social media and religious communities can complement these interventions by translating platform capability into perceived benefit for diverse user groups.
The study is limited by its sample size .
= .
, urban skew, and the weakness of the Perceived Ease of Use construct, whose AVE falls below the recommended threshold and whose indicators require re-specification in future work.
The cultural reading offered in the Discussion is interpretive rather than measured, since religiosity, collectivism, and adherence to ulama authority were not modelled as latent variables in the present survey.
Subsequent research should expand the sample geographically beyond major urban areas, refine the PEU instrument, and add cultural moderators such as religiosity, generational cohort, exposure to pesantren networks, and prior digital banking experience, to test directly how the trust and security balance evolves as the online zakat market matures within the Indonesian cultural setting.
Acknowledgment The authors thank the respondents who took part in the online survey, the management of the online zakat institutions that facilitated participant recruitment, and the colleagues at the Faculty of Economics and Business.
Universitas Islam Negeri Syarif Hidayatullah Jakarta, who provided constructive feedback on earlier drafts of this manuscript.
________________________
How to Cite Mulazid.
Putra.
Fauzan, & Nugraha.
Behavioral Intentions of Online Zakat Users in Indonesia: An Extended Tam with Trust.
Security.
And Local Cultural Perspectives.
International Journal of Islamic Business and Economics (IJIBEC), 10.
, 43-55.
https://doi.
org/10.
28918/ts4ma476 International Journal of Islamic Business and Economics (IJIBEC), 10 .
June 2026, 43-55
________________________
Conflict of Interest Statement: The authors declare that the research was conducted without any commercial or financial relationships that could be construed as a potential conflict of interest.
________________________
Author Contributions Statement: A.
contributed to conceptualization, methodology development, supervision, project administration, and manuscript review and editing.
contributed to investigation, data collection, data curation, formal analysis, and preparation of the original draft.
contributed to methodology refinement, validation, visualization, and critical review of the manuscript.
and D.
contributed to data curation, formal analysis, visualization, and original draft preparation.
All authors contributed to the interpretation of the findings, reviewed the manuscript, and approved the final version.
________________________
AI Usage Statement: The authors declare that Artificial Intelligence (AI) tools were used only for language editing, grammar checking, and improving manuscript clarity.
AI was not used for research design, data analysis, interpretation, or conclusion development.
All intellectual and scientific contributions were conducted solely by the authors, who take full responsibility for the originality, accuracy, and integrity of this manuscript.
AI tools are not listed as authors or contributors.
________________________
Copyright A2026 Mulazid.
Putra.
Fauzan, & Nugraha.
This article is distributed under the terms of the Creative Commons Attribution 4.
0 International License (CC BY 4.
, which permits use, distribution, and reproduction in any medium or forum, provided that appropriate credit is given to the original author.
and copyright holder.
, and the original publication in this journal is properly cited in accordance with accepted academic standards.
Any use, distribution, or reproduction that does not comply with these terms is not permitted.
References