Exploring trust in decentralized finance intermediaries: a taxonomy and archetypes for guiding blockchain-based investment decisions on the web

Why has blockchain-based investing not yet become mainstream? Despite its technical potential, it is often perceived as complex and difficult to reconcile with traditional investor behavior, factors that, following the Diffusion of Innovation (DOI) theory, inhibit adoption by lowering perceptions of relative advantage, compatibility and trialability (Rogers et al., 2014). Trust plays a pivotal role, as perceptions of risk and uncertainty in complex technologies strongly influence how potential adopters evaluate technology's attributes. Highlighting the impact of word-of-mouth on mobile banking acceptance, negative experiences tied to dis-adoption can erode trust, while positive stories from re-adopters restore it (Saeed and Xu, 2020). As shown in financial technology (FinTech) innovations, such as facial recognition payments, trust built through transparency, quality and security is essential not only for adoption but also for achieving sustained, mainstream use (Lim et al., 2025).

In light of these challenges, this study examines decentralized finance (DeFi) intermediaries, which offer innovative investment opportunities through blockchain technology to create transparency, thereby promoting mass adoption and trust. While the growth of DeFi platforms — measured by market capitalization — is rapidly increasing (Alamsyah and Muhammad, 2024; Voskobojnikov et al., 2021) and the number of new assets and transactions via blockchain-based platforms is already substantial (Alamsyah and Muhammad, 2024), broad mass acceptance remains a challenge (Król and Zdonek, 2022; Zeiβ et al., 2024).

Adoption is hindered by a lack of awareness and perceived trust deficit among investors (Zeiβ et al., 2024). Additionally, the plethora of new crypto investment opportunities, primarily driven by emerging web applications, creates a complex environment for private investors (Caxton et al., 2022; Voskobojnikov et al., 2021). The accessibility and functionality of financial services have undergone significant evolution in recent years (Alamsyah and Muhammad, 2024). Ennew and Sekhon (2007) argue that customers generally find it challenging to understand financial products and are therefore more likely to trust digital providers. Consequently, private investors face heightened risks when purchasing financial products (Ennew and Sekhon, 2007), especially in the blockchain domain, where well-known financial providers are less prevalent due to new intermediaries (Chen and Bellavitis, 2020; Grassi et al., 2022). Numerous negative reports, such as those concerning fraud in the blockchain sector, also adversely affect acceptance (Valeonti et al., 2021; Zeiss et al., 2024). The inherent complexity of blockchain complicates matters (Glomann et al., 2019), making it challenging for individuals with limited investment experience to understand the nuances of decentralized investments compared to traditional banking (Litterscheidt and Streich, 2020; Zeiβ et al., 2024).

To facilitate the acceptance of blockchain and crypto assets among the broad masses, it is crucial to create a supportive environment for newcomers and inexperienced users (Glomann et al., 2019). In this context, trust plays a pivotal role in driving acceptance (McKnight et al., 2002; Zeiβ et al., 2024). As evidenced by extant literature, mitigating trust-related concerns can reduce perceived risk, thereby diminishing users' uncertainties or apprehensions regarding potential outcomes (McKnight et al., 2002; Slade et al., 2015). However, little attention has been paid to blockchain-based investment intermediaries, critical to establishing trust. This underscores the need for a comprehensive overview of the existing investment opportunities, as market transparency for customers remains insufficient (Langley and Leyshon, 2021; Zeiss et al., 2024). It is essential to identify the characteristics that crypto intermediaries possess and those that platform providers should transparently display to increase trust levels.

This study contributes to the overarching goal of blockchain acceptance in the Internet market by enhancing the transparency and comprehensibility of the multitude of blockchain-based investment opportunities while recognizing trust as a vital factor for adoption. We focus on the apparent differences between investment intermediaries and the derivation of investment archetypes, which are identified, explained and discussed in the context of trust, specifically concerning technology-based and institution-based trust as well as crypto-assets-related research findings.

Therefore, Section 2 examines existing research on blockchain-based financial service intermediaries, trust in DeFi platforms and related work. Subsequently, Section 3 outlines the research design based on the taxonomy development and cluster analysis. The final taxonomy and the derived archetypes are presented in Section 4, followed by a discussion in Section 5. Finally, Section 6 addresses the conclusions, limitations and future research.

Blockchain-based financial service intermediaries: The financial sector has recently undergone a significant transformation, with banks historically serving as trustworthy contact points for customers (Gramlich et al., 2023; Langley and Leyshon, 2021). With the rise of FinTech, new companies have emerged to compete with established financial service providers (Chen and Bellavitis, 2020; Schwiderowski et al., 2024). Introducing blockchain and DeFi has enabled further transformations, creating novel opportunities for storing and exchanging assets (Zeiss et al., 2024). This has led to new business models, services and products offering various new investment opportunities (Gramlich et al., 2023; Voskobojnikov et al., 2021). The convergence of physical and digital realms is driven by tokenization, enabling the fractionalization of assets, such as artworks, through internet-based applications (Whitaker and Kräussl, 2020). Moreover, these changes and the emergence of new market participants have led to a shift in power distribution within the financial sector – a transformation often characterized by the concepts of disintermediation and re-intermediation (Zeiβ et al., 2024). Technological trends have introduced new intermediaries into the financial system, replacing traditional banks (Cai, 2018; Langley and Leyshon, 2021) and shifting the existing trust structures of investors (Chen and Bellavitis, 2020; Langley and Leyshon, 2021).

Trust in DeFi platforms: Although blockchain technology is perceived as more secure due to its tamper-proof nature (Gramlich et al., 2023; Valeonti et al., 2021), there is a notable trust deficit in applications among the general public, particularly in financial domains (Zeiβ et al., 2024), which hinders the DOI. Studies show that trust cuts uncertainty and boosts user confidence in digital and finance systems (Gan and Lau, 2024; Jalan et al., 2023). Trust has been shown to strongly influence technology adoption, as demonstrated in FinTech, where online or mobile banking and facial recognition payments are prevalent (Shiau et al., 2023; Zhou, 2011), as well as in the area of crypto assets (Voskobojnikov et al., 2021). In the context of the cryptocurrency ecosystem, trust issues such as price manipulation, security, governance and transparency are relevant (Rehman et al., 2020). To establish a trust relationship, it is imperative that the trustor (crypto asset investor) possesses sufficient reasons to place their trust in the trustee (financial intermediary) and that the trustee is perceived as trustworthy by the trustor (Smits and Hulstijn, 2020). Years ago, trust in the conventional banking system was predominantly founded on reputation or prior interactions (party-based or person-based trust). With the advent of technological advancements, this paradigm has undergone a shift. Consequently, the crypto literature has begun to explore institution-based and technology-based trust (Sas and Khairuddin, 2017; Smits and Hulstijn, 2020; Voskobojnikov et al., 2021). Institution-based trust relies on third-party verification and is particularly relevant in the online sector, where transactions occur between new or anonymous parties (Pavlou and Gefen, 2004; Zhao et al., 2023). Therefore, certifications, control mechanisms and feedback features are particularly relevant (Pavlou, 2002). Due to the unregulated nature of blockchain technology, third-party institutions like governance or regulatory authorities are challenged to provide an adequate response, protect investors and mitigate risks while fostering technological development (Ferreira and Sandner, 2021; Gramlich et al., 2023). Given the crypto market's global orientation and the original exclusion of intermediaries, the use of institutional trust is currently limited. Technology-based trust places greater emphasis on the integrity and transparency of the technology itself, assuming that the mechanisms will perform as intended (Sas and Khairuddin, 2017; Smits and Hulstijn, 2020). Accordingly, it encompasses a range of elements, including technical standards, security procedures and protection mechanisms (Pavlou and Gefen, 2004; Smits and Hulstijn, 2020). Intermediaries can position these elements more clearly, making them appear more trustworthy to the trustor. However, these elements must be elaborated and presented to the platform user.

Related work: Studies on crypto investments focus on adoption (Zeiβ et al., 2024) or explore opportunities, including user-centered non-fungible token (NFT) marketplace design (Caxton et al., 2022) and metaverse platforms (Lee et al., 2024). Further studies explore concepts such as security (Valeonti et al., 2021), user-focused design (Lee et al., 2024; Litterscheidt and Streich, 2020) and usability of platforms, websites and apps (Rehman et al., 2020) to discuss the specific challenges of blockchain-based financial ecosystems. Taxonomies address digital platforms (Derave et al., 2024) and their decentralization (Perscheid et al., 2020), as well as blockchain-based business models (Weking et al., 2020). Additionally, they handle DeFi on a broad level (Puschmann and Huang-Sui, 2024) or, more specifically, analyze business models (Beinke et al., 2024), token structures (Hartwich et al., 2024; Schwiderowski et al., 2024) or risk and trust issues (Rehman et al., 2020). However, no approaches have yet been developed to address the aforementioned research questions.

We applied a taxonomy development approach proposed by Nickerson et al. (2013) to examine the differentiation of blockchain-based investment intermediary platforms. Following the categorization of real-world objects based on the taxonomy, a cluster analysis (Ward, 1963) was conducted to identify archetypes. The taxonomy and archetypes were subsequently evaluated through a survey, as depicted in Figure 1, and discussed in the trust context afterwards.

A taxonomic system relies on classifying meta-characteristics (MCs), which define its fundamental structure and are established in the initial stage of taxonomy development (Nickerson et al., 2013). Since MCs must align with the taxonomy's purpose, our MCs focus on functionality, highlighting overall investment attributes of intermediary platforms, architecture and security. As a result, our MCs incorporate the previously mentioned interplay of institution-based and technology-based trust elements, which are relevant in the digital financial world. We collected several taxonomies, analyzed their structures in workshops and derived our MCs, e.g. functionality business model perspectives or platform characteristics (Perscheid et al., 2020; Puschmann and Huang-Sui, 2024; Weking et al., 2020).

Additionally, taxonomy development is an iterative process requiring predefined ending conditions (EC). We adopted the eight objective and five subjective EC from Nickerson et al. (2013). The finalization of taxonomy development necessitates fulfilling all EC.

Figure 1 illustrates that we chose each iteration between conceptual and empirical procedures. Moreover, Figure 1 provides a structured overview of the data collection methods and the identified real-world objects. We collected these real-world objects using a multi-stage strategy. This approach involves a literature review of scientific papers (e.g. Shah et al., 2023; Zeiss et al., 2024), searching on websites (e.g. coinmarketcap.com) or following non-scientific articles (e.g. financial institutions). The two criteria that guided our selection process were that the intermediary platforms must be operational and that they must represent a viable business. As our scope is on private investment, we excluded objects with only business-to-business solutions.

• Iteration I: Initially, a conceptual-empirical approach was chosen due to the research team's substantial expertise in this domain (Nickerson et al., 2013). The objective of this iteration was an initial taxonomy draft (Herm et al., 2022). We performed a systematic mapping database search for data collection (Petersen et al., 2015), as this method aims to provide a comprehensive overview. A search was conducted using the query terms “DeFi”, “Decentralized Finance”, “Taxonomy” and “Systemization”. Finally, a corpus of 13 taxonomies served as a literature base for the first taxonomy draft.

• Iteration II: The second iteration contains a conceptual-empirical approach, incorporating a structured literature review (Vom Brocke et al., 2009). This deductive procedure involves retrieving knowledge from scientific literature and combining it with empirical data in the taxonomy (Nickerson et al., 2013). This review concentrated on “investment platforms” and “decentralized finance”, resulting in 34 articles.

• Iteration III: This iteration also drove the conceptual-empirical approach, extending our knowledge base. Therefore, to extend our knowledge base, we applied a systematic mapping and snowballing method in the third iteration (Petersen et al., 2015). This method forms a corpus of basic literature (De Reuver et al., 2018), and a forward search was initiated from this point. The resulting 51 papers were used to refine the taxonomy.

• Iteration IV: To gain practical knowledge in addition to the theoretical foundation, we employed the empirical-conceptual approach. By this inductive procedure, we analyzed a set of 15 intermediary platforms to recognize attributes and categorize them, as outlined by Puschmann and Huang-Sui (2024). This approach is grounded in document analysis, which entails collecting and examining documents– such as white papers, websites, documentation and scientific as well as non-scientific articles – for each platform (Kundisch et al., 2022).

• Iteration V: In the fifth iteration, we selected an empirical-conceptual approach. The objective was to analyze and classify intermediaries based on empirical data. We contacted employees of the platforms identified in Iteration III (n = 15) and conducted five interviews using the semi-structured approach of Kvale and Brinkmann (2008). We presented the taxonomy, and the experts reviewed the MCs as well as categorized the intermediary platforms. Additionally, we reduced the complexity and scope of the taxonomy, as mentioned by the experts.

• Iteration VI: For the final review of the current taxonomy version, we applied the empirical-conceptual approach. This review assessed the current state based on 35 identified intermediary platforms. As all objects were successfully assigned to their respective dimensions, no additional characteristics or dimensions were identified for inclusion in the taxonomy. The changes made were limited to nominal adjustments. Consequently, we initiated a comprehensive review of all EC and confirmed that all subjective and objective criteria were met without exception.

To identify archetypes, validate the taxonomy and derive archetypal patterns, the research team employed an agglomerative hierarchical clustering algorithm (Ward, 1963). For this, a descriptive analysis of 50 intermediary objects (from Iterations IV and VI: n = 15, n = 35) was conducted (Herm et al., 2022), and each object was classified within the taxonomy's dimensions. The frequency of occurrence for each characteristic was used in the descriptive analysis and subsequent hierarchical pattern (see Figure 3).

After deriving archetypal patterns, both artifacts underwent evaluation. A quantitative survey (Szopinski et al., 2019) was conducted across five key aspects, including functionality, architecture, security, the taxonomy itself and the archetypes. The survey was completed on the Prolific platform, where we configured filters for our recruitment, with 115 individuals professionally engaged in the financial sector, including financial advisors, traders and portfolio managers. Owing to their active roles in investment decision-making and financial consulting, these participants were domain experts. Figure 1 presents an overview of the demographics. The study was structured as follows: First, the topic was introduced to provide the necessary context. After addressing general questions, the taxonomy was briefly explained. Furthermore, measurement items by Sonnenberg and Vom Brocke (2012) (see Figure 1) were combined with a 7-point Likert scale for each item. The findings are summarized in Section 4.3.

Developing a taxonomy and archetypes supports the differentiation of blockchain-based investment intermediaries. This market transparency is crucial for establishing trust and is pivotal to the acceptance of crypto. It also reveals fundamental characteristics relevant to technology-based and institution-based trust that must be demonstrated to the market for greater acceptance.

This section presents the taxonomy of blockchain-based investment intermediaries (see Figure 2). It describes its 3 MCs, 18 dimensions and 63 characteristics that can be assigned to the trust literature, as described in Section 5, and which are relevant to RQ1 and the groundwork for answering RQ2.

Functionality: This MC encompasses dimensions that delineate intermediaries' utility features and value-creating functions. The main investment focus is differentiated into income generation, exchange, asset management, in-game payouts and other categories (Shah et al., 2023; Valeonti et al., 2021). Upon closer examination of intermediary platforms, financial features such as lending, staking or yield farming can be distinguished (Schär, 2021), as well as derivative options, insurance, tokenized physical assets (Litterscheidt and Streich, 2020; Whitaker and Kräussl, 2020) and NFTs (Hartwich et al., 2024). Intermediaries can create their own native tokens, which are an integral part of their ecosystems. These tokens are grouped into asset, utility and payment tokens (Schwiderowski et al., 2024). Regarding deposit options, the taxonomy distinguishes between fiat currencies, cryptocurrencies and proprietary tokens (Zeiss et al., 2024). With new regulatory approaches (Ferreira and Sandner, 2021), growing technical maturity and promised value gains, corporate clients are increasingly investing in the crypto market. The target groups of intermediaries are divided into private investors and businesses (Beinke et al., 2024; Derave et al., 2024). A critical factor is how users gain access to the intermediaries, with mobile applications and Internet websites playing a particularly central role (Caxton et al., 2022; Lee et al., 2024). Intermediaries should provide financial learning, as financial literacy is a critical factor for adopting crypto investments. An analysis was conducted to determine whether the platforms offer integrated learning resources such as video courses (Król and Zdonek, 2022; Rehman et al., 2020).

Architecture: This MC entails the fundamental principles and structural configurations of the intermediary platforms' operational, relational and systemic properties, demonstrating essential elements for technology- and institution-based trust. Anonymity distinguishes platforms based on the degree of identity obfuscation, ranging from pseudo-anonymity, where identities are partially concealed, to no anonymity (Benedetti, 2021; Perscheid et al., 2020). The blockchain network differentiates intermediaries based on whether they use a single-chain, which limits functionality to a specific ecosystem (Meyer et al., 2022; Puschmann and Huang-Sui, 2024), or a multi-chain, which expands transaction options across various blockchains but does not allow full interoperability (Grassi et al., 2022). Transaction processing varies among platforms based on whether investments are conducted on-chain (Puschmann and Huang-Sui, 2024), off-chain (Schär, 2021) or through centralized intermediaries without direct blockchain integration (Shah et al., 2023). Consensus includes Proof of Work and Proof of Stake for decentralized verification, Proof of Authority for private blockchains, hybrid approaches combining protocols, alternative algorithms under development and centralized systems without consensus (Benedetti, 2021; Cavallaro and Mathieu, 2024). Interoperability categorizes intermediaries by their ability to transfer assets and data (Gramlich et al., 2023), ranging from no interoperability in centralized systems to seamless cross-chain operability (Chen and Bellavitis, 2020; Ray, 2023). Furthermore, the market mechanisms encompass intermediaries within the business logic of transaction processes, including centralized order books managed by intermediaries, on-chain order books that rely entirely on smart contracts for decentralized transactions and market makers utilizing liquidity pools to facilitate exchanges (Schär, 2021). Additional mechanisms include reserve aggregation for optimal pricing, investment pooling for collective returns and hybrid approaches (Shah et al., 2023).

Security: This MC outlines the mechanisms and structures that ensure integrity, accountability and equitable participation within the intermediary platform, particularly related to institutional and technology-based trust. Governance differentiates intermediaries based on how decisions are made (Beck et al., 2018), ranging from centralized models controlled by a single entity to partially decentralized systems with shared decision-making, and fully decentralized structures where decentralized autonomous organizations facilitate collective decision-making by the community (Greiner et al., 2025; Shah et al., 2023). Asset control differentiates whether investors directly manage their assets through self-custody via wallets and smart contracts or relinquish control to intermediaries, resulting in centralized management and increased vulnerability (Alamsyah and Muhammad, 2024). When it comes to legal accountability, intermediaries can be regulated, adhering to licensing guidelines such as Virtual Asset Service Provider License (VASP) and Markets in Crypto-Assets Regulation (MiCA) (Ferreira and Sandner, 2021), or unregulated, where platforms operate outside direct regulatory oversight (Heimburg and Wiesche, 2023). The trust structure categorizes platforms' foundation of user trust, including intermediary-based trust in central entities, code established through smart contract transparency, distributed community trust managed by decentralized communities and hybrid structures (Kölbel et al., 2022). Furthermore, the distribution of power distinguishes between token-based and no distribution (Perscheid et al., 2020).

Figure 3 provides an overview of the descriptive analysis and the cluster dendrogram, illustrating six resulting clusters from the algorithmic approach. Data from 50 intermediaries were collected and categorized along the taxonomy dimensions. The value distribution of each cluster is shown in the subsequent table and addresses RQ2. Finally, Figure 4 presents the archetypes with brief descriptions, highlighting the current intermediary market relevant for private investments.

The DeFi Gaming Intermediary focuses on in-game payouts with NFTs, as well as features primarily related to staking and exchanging. Intermediary access is provided through web-based or multi-channel, targeting a shared customer group. Intermediaries implement native tokens for multiple purposes (e.g. AAVEGOTCHi). The architecture recognizes interoperability, on-chain transaction processing and pseudo-anonymity. For security, decentralized governance involves decision rights and power distribution, paired with an unregulated environment and asset control through self-custody.

The second cluster is defined as Multi Application Web Intermediary and unites around a main focus on primary exchange. The archetype includes the shared functionality of yield farming (e.g. Yearn) and web-based access. However, most platforms also offer staking or derivatives. The user base mainly consists of private investors, but corporate clients for capital allocation are also included. Most intermediaries implement a multi-blockchain architecture. Transaction processing occurs both on and off-chain, with pseudo-anonymity for users. Platform governance is based on decentralized decision-making with token-based power distribution. Regulation is typically absent, and trust is provided through a combination of intermediaries, code or distributed communities.

The Specialized Staking Web Intermediary is characterized by offering staking options (e.g. liquid staking on Lido). The focus is on income generation as these intermediaries are highly specialized. The typical target audience consists of customers with web-based access, though options for businesses may also exist. Most platforms implement native tokens, but there are few learning opportunities, as these intermediaries usually cater to experts rather than new users. This is accompanied by self-custody asset control and partial decentralized governance, where users have limited decision-making rights.

The objects of the Learning-driven Exchange Intermediary combine the main focus on exchange functions with the provision of staking options. Financial features such as lending, yield farming, derivatives, insurance or NFTs may also be integrated. As these exchanges often operate as major players and serve as primary entry points for beginners, they are predominantly offered across multiple channels, including smartphone apps and web-based platforms (e.g. Binance). In most cases, businesses are also addressed in the target group. Typically, these intermediaries do not implement native tokens. However, a key functionality is the availability of financial education resources, as these platforms are prominent spots for new users. The governance is centralized, with no implemented power distribution or decision-making rights for users, and operating in regulated environments. The intermediary itself ensures trust, while third-party services primarily handle asset control.

The Tokenized Asset Intermediary is characterized by its focus on asset management and the implementation of tokenized physical assets (e.g. Timeless). Both private customers and business investors are part of the target audience with mixed access. Deposits are typically made via fiat currency transfer. Most intermediaries within this archetype do not offer native tokens and generally do not provide financial learning opportunities. They often operate on a single blockchain without interoperability or user anonymity. In terms of market mechanisms, they integrate centralized order books. Governance is based on a centrally conceived model with no power distribution. Asset control is handled by third-party solutions, while an intermediary establishes trust.

The Private Investor NFT Marketplace Intermediary archetype entails NFT marketplaces (e.g. OpenSea), which are distinguished by their prominent shared feature of NFTs. The primary focus may be on exchange or asset management functions. The target group comprises private investors who can access the platform via web or apps. Deposits can be made using both fiat and cryptocurrencies. These platforms do not integrate native tokens, although some intermediaries provide learning opportunities. Most platforms utilize a multi-blockchain architecture with on-chain transaction processing. They ensure either one-chain or cross-chain interoperability. Governance follows a centralized model without power distribution among users. The environment is regulated, and the trust structure is based on either code or centralized intermediaries. In most cases, asset control is conducted via self-custody.

The survey with 115 participants provided feedback on the entire taxonomy, the MCs and the archetypes. Measurement tools from Sonnenberg and Vom Brocke (2012) were used (see Figure 2). The results show a high level of agreement with the research findings, although differences between the MCs are noticeable:

1. The Entire Taxonomy achieved the highest average agreement (5.61 on a 7-point Likert scale), reflecting its clarity, accessibility and comprehensiveness. Participants appreciated its ability to effectively organize complex financial processes. However, qualitative feedback highlighted opportunities to simplify overly detailed elements, which seem to be a typical limitation of comprehensive taxonomies.

2. MCs: Functionality aligns with practical applications. Participants found the taxonomy's usability and operational applicability favorable, but suggestions included improving the balance between functionality and simplicity for broader adoption. This can be underlined as it “provides sufficient transparency and ease to navigate and build a good portfolio and accessibility” (I19). Nevertheless, some also mentioned that it might be congested (I97) or that the layout might not be easy to read for non-scientific experts (I88, I98).

Architecture's feedback emphasized feasibility and operational applicability, with conceptual clarity identified as an area for improvement. Some participants noted the complexity (I3, I7, I8), while others praised “the design is understandable and is not multi-layered in terms of the dimensions, make it understandable too” (I15). One participant commented that the MC “breaks down the technical backbone of blockchain platforms” (I27). From this, we can deduce that granularity is important, but users' understanding still needs to be sharpened.

The Security dimension (5.61) was reckoned for its clear structure and accessibility, reflecting participants' strong emphasis on effective risk management frameworks. Suggestions included refining detailed elements to better align with regulatory and industry-specific needs (I68). Qualitative feedback highlighted that this MC “helps users and stakeholders better understand security features and risks” (I100). Additionally, it enables a “comprehensive assessment of security measures, facilitating the identification of vulnerabilities and areas for improvement” (I101). This shows a higher added value for technology-based trust mechanisms.

1. Archetypes garnered strong support, with an average agreement of 87.6% based on cumulative responses (“Somewhat Agree,” “Agree” and “Strongly Agree”). Participants endorsed their relevance and feasibility across various decision-making scenarios. The ability of the archetypes to operationalize abstract elements of the taxonomy was particularly appreciated. Statements such as “very clear” (I13, I34, I38) and “very applicable” (I13) confirm this. Furthermore, the “brief, straight-to-the-point explanations make everything well-rounded” (I38). One participant commented, “The various archetypes are revealing and easy to understand, allowing me to get the entire overview” (I111), which is particularly interesting for diversification and market transparency.

In summary, the taxonomy received broad approval across all dimensions and derived archetypes. Results highlight its clarity, accessibility and applicability while identifying potential improvement areas, particularly in simplification and design. This information proves crucial in providing newcomers to the market with the necessary knowledge and transparency.

The adoption of blockchain technology in the Internet market is predominantly driven by the transparency and comprehensibility of the numerous blockchain-based investment opportunities while acknowledging trust as a fundamental element in the context of digital platforms and transactions (Gan and Lau, 2024; Jalan et al., 2023; Sas and Khairuddin, 2017). Consequently, market transparency should be created (Zeiss et al., 2024), while intermediary platforms must provide secure and reliable environments to protect users from potential fraud and technical vulnerabilities (Voskobojnikov et al., 2021). In this context, trust not only needs to be established but also consistently reinforced. This requires a clear orientation that supports trust-building efforts and provides a structured understanding of the dynamics within blockchain ecosystems. While our taxonomy offers a comprehensive context for understanding blockchain intermediaries, its connection to trust is deeply embedded in its various dimensions. The archetypes derived from the taxonomy highlight specific characteristics that have become dominant in blockchain-based investments, fostering market transparency. The results provide a clear overview of the current investment landscape within the DeFi market, fostering a more comprehensible and accessible investment environment.

Theoretical implications: Our research sheds light on societal transformation and hence innovation diffusion in investments, exemplified by the increasing role of intermediaries (Zeiβ et al., 2024). Through web technologies, these intermediaries facilitate broader public access, thereby promoting the democratization of finance (Gramlich et al., 2023). Emphasizing technology-based and institution-based trust, our taxonomy presents relevant elements, particularly in the dimensions of architecture and security, which are pertinent in the scientific discourse on blockchain trust. Our taxonomy incorporates regulatory aspects (legal accountability) within institution-based trust, and several dimensions concerning technology-based trust (e.g. consensus, governance).

A thorough examination reveals a discernible interconnection between the identified dimensions within the taxonomy and the core components of Mayer's trust theory – ability, integrity, benevolence – thereby underpinning its contribution to well-established trust principles (Mayer et al., 1995). This connection is demonstrated by including market mechanisms and transaction processing, which reflect ability through their emphasis on system reliability and technical competence. Moreover, governance and legal accountability illustrate integrity through transparent and consistent rule enforcement. Benevolence is indicated by financial learning and access, which enable user-centered design and foster inclusive access. Concerning trust transfer theory, the transfer of trust from a known object, perceived as reliable, to a new, initially unknown object is possible (Cao et al., 2018). However, for such a transfer to occur, a certain level of initial trust must first be established. Our work aims to foster the construction of this initial trust, which serves as the foundation for further trust development and enables its transfer to new blockchain-based intermediaries. Strengthening initial trust not only facilitates trust transfer but also increases adoption readiness and lowers barriers to market entry.

While much of the existing literature focuses on trust-building mechanisms, the concept of trust-free systems offers a complementary perspective, particularly relevant in blockchain-based environments (Hawlitschek et al., 2018). These systems aim to minimize reliance on interpersonal or institutional trust by embedding trust directly into technological protocols. Rather than depending on a trusted third party, users rely on the verifiability and immutability of code and distributed architectures. Nevertheless, blockchain-based platforms need to be aware that blockchain technology, in and of itself, cannot create an environment that eliminates the need to build trust outside of the closed blockchain ecosystem (Hawlitschek et al., 2018). This underscores the notion that, while technology-driven trust is imperative, it is insufficient in isolation. To fully leverage the advantages of trust-free blockchain-based platforms, researchers and practitioners must develop approaches to overcome the trust frontier between closed technical systems and the complex realities of the physical world. The mainstream diffusion of blockchain investing will remain stalled until both institutional safeguards (e.g. clear regulation and insured custodians) and user-centric, reliable interfaces co-elevate institutional and technology-based trust, thereby restoring the DOI attributes of compatibility, relative advantage and trialability essential for mass adoption (Rogers et al., 2014).

Although the initial aim of this study was to serve as a foundation for analyzing the intermediary market within the DeFi network and identifying conceptual discrepancies, the evaluation and survey feedback demonstrated that it covers a far broader spectrum of research on information systems and the internet. Additionally, the survey results indicate that the findings of this study contribute to investment decision-making, financial risk analysis, and the accessibility of intermediaries via the internet (I19, I36, I97). Participants viewed it not merely as an analytical framework but as a practical tool that could support investment decision-making and financial risk assessment (I19, I36, I97). While the taxonomy was not initially designed for these purposes, this feedback highlights its perceived scientific and practical value for the mentioned areas. However, we also contribute to the following fields, which are briefly summarized. We recognize the importance of ethical accountability and power distribution in digital ecosystems (Derave et al., 2024; Heimburg and Wiesche, 2023) by examining governance structures and asset control. By systemizing the financial features of intermediaries, this research supports economic transformation, focusing on value creation and resource allocation in investment markets (Chen and Bellavitis, 2020; Hartwich et al., 2024; Lee et al., 2024). Finally, the results foster the expansion of internet-related policy research and support policy implications by analyzing intermediaries that are not centrally regulated (Ferreira and Sandner, 2021; Gramlich et al., 2023) and digital platforms (Heimburg and Wiesche, 2023).

Practical implications: As blockchain technology and online applications disrupt traditional financial systems (Cai, 2018), trust becomes an indispensable tool for intermediaries to attract new private investors and deepen user engagement (Glomann et al., 2019; Rehman et al., 2020). It functions as a distinctive value proposition for a platform, as reliable and secure web applications can help differentiate themselves from other intermediaries by giving users a sense of security and increasing user confidence, thereby enhancing the overall value of the platform (Lee et al., 2024). Utilizing appropriate tools that underscore pertinent components of trust within the design can favor the adoption process (Voskobojnikov et al., 2021). The taxonomy provides a framework for intermediaries to emphasize the relevant trust factors of blockchain platforms that exert influence on the trustor.

Additionally, for intermediaries, the taxonomy and archetypes offer critical factors for benchmarking, providing developers with a foundation for pre-planning platforms or reference points for strategic development. Intermediary platform developers (Zeiβ et al., 2024) must recognize the findings and derive features to secure trust and ensure broad access to intermediary platforms and the crypto ecosystem (Rehman et al., 2020). The intermediary must communicate and visualize the trust-relevant topics from the taxonomy to be perceived as trustworthy by the trustor and to enable a trust relationship. This involves adapting strategies and developing features focused on the taxonomy dimensions of functionality, architecture and security, such as financial features, decentralized decision-making in governance or targeted outreach to specific user groups. To strengthen the institutional-based trust of intermediaries in their relationships with private investors, intermediaries should properly implement laws, rules and regulations (such as MiCA or VASP) and thereby transparently demonstrate the implementation of trust mechanisms to users. To fortify their long-term standing, intermediaries must, in contrast to the unregulated nature of blockchain-based investments, address the issue of institution-based trust and proactively establish support structures, initiate them or demand them from the relevant authorities.

However, the visualization of the trust elements and implementation alone is insufficient. In the short term, new market entrants must provide active guidance and support trust-building structures (Glomann et al., 2019) and foster safety nets for exchanges (Voskobojnikov et al., 2021). In the long term, society needs to create sustainable and impartial learning opportunities that make it easier for the general public to understand the basics of blockchain-based intermediary platforms, as users often struggle to fully understand the information and the safeguards put in place. Since financial education and literacy programs in the crypto sector lag behind its developments, intermediaries primarily need to expand their educational offerings to reach a broader audience and build trust among the general public (Jalan et al., 2023; Litterscheidt and Streich, 2020). Based on its characteristics, the Learning-driven Exchange intermediary is particularly relevant for widespread acceptance. The findings also offer new insights into diversification and portfolio management, which are fundamental in finance and must be communicated to private investors (Król and Zdonek, 2022). Intermediaries can select from the developed archetypes tailored to their specific functional, architectural and security-related requirements. The derived archetypes offer clear insights into the diversity of blockchain intermediaries, enabling investors to match platforms to their individual risk profiles, investment goals and technical preferences. In line with this, the taxonomy dimensions serve as a practical checklist when evaluating DeFi platforms. Investors should ask whether a platform offers regulated custody, supports self-custody or provides decentralized governance mechanisms.

Although it remains uncertain whether blockchain-based investments will reach full market penetration, it is crucial to continuously assess the transparency of the evolving landscape of financial intermediaries. This study addresses this challenge by developing a taxonomy that enhances the understanding of blockchain-based financial intermediaries and manages the complexity of DeFi adoption within the context of trust. In doing so, characteristics related to institution-based and technology-based trust are analyzed, particularly considering the digital realities in the cryptocurrency sector. Following the DOI theory, our findings underscore the interplay of technology and a user-centered perspective in the context of adoption. By classifying 50 real-world intermediaries, six archetypes were derived (e.g. specialized staking web intermediary). The findings highlight crypto-based investments and differentiate DeFi intermediaries based on functionality, architecture and security. Trust is positioned as both an outcome and a driver for the success of blockchain-based intermediaries.

While the taxonomy and archetypes provide valuable insights, they represent a time-specific snapshot and are therefore inherently limited in scope and completeness (Hartwich et al., 2024; Herm et al., 2022). We encountered various limitations during the data collection process, which we addressed by employing a combination of qualitative and quantitative methods. This approach entailed the utilization of diverse databases for literature searches and the coding of interviews. The archetypes and clusters are based on platform classifications. To ensure robustness, a multi-level process was employed for platform selection, and classification was conducted collaboratively within the team to ensure high data quality. This study emphasizes these aspects by employing various methods and approaches recognized in IS research. Future research should build upon these findings, focusing on developing risk assessment tools for private investors, decision support structures and enhancing trust in intermediaries. Additionally, financial education about crypto opportunities should be analyzed and expanded. The taxonomy dimensions should be examined for options to operationalize trust-building measures, considering not only technical but also socio-psychological barriers.

Blockchain-based investments are still relatively new, and public acceptance remains low. However, the results point to a paradigm shift in which trust is no longer solely dependent on technological robustness but also user-centric design and accessibility. This is particularly relevant for beginners, who require precise and trustworthy guidelines to navigate blockchain-based investments effectively. The information systems community must provide support, as well as neutral and expert advice. To this end, it is essential to establish a solid foundation alongside the banking sector and develop confidence-inspiring financial instruments to facilitate the adoption and further development of DeFi.