From Impact to Value: A Comprehensive Framework for Evaluating Digital Service Sustainability

1. Introduction

The evolution of digital services is profoundly transforming the ways in which individuals, organizations, and institutions produce, share, and utilize information. In this context, digital services that process and disseminate technical data — such as those related to earthquake monitoring — assume a crucial role not only from a technological perspective, but also in relation to sustainability across its three fundamental pillars: environmental, economic, and social. However, while consolidated models have been developed over the past decades for assessing the sustainability of physical products and infrastructures, the evaluation of digital service sustainability remains a relatively recent and still conceptually evolving field.

If we consider the commonly adopted scientific definition of digital sustainability — “the process of designing, developing, deploying, and also disposing (end-of-life) digital resources with the aim of improving environmental and economic well-being, considering the entire life cycle” [1] — the analytical focus predominantly concerns the sustainable management of the technological and physical infrastructures enabling the service. Within this perspective, sustainability assessment is largely confined to the evaluation of direct impacts associated with infrastructure components (e.g., data centers, hardware, network systems, and energy consumption), thereby emphasizing the environmental footprint of digital technologies.

A broader and more enabling interpretation emerges from the definition reflected in the Manifesto of the “Fondazione per la Sostenibilità Digitale”[2], which conceptualizes digital sustainability as “the way in which digital transformation and technological innovation are oriented so that technology becomes a tool for sustainable development, contributing to meeting the needs of present generations without compromising those of future generations.” In this view, digital technologies are not merely objects of sustainability assessment but strategic levers capable of generating systemic sustainable value.

Accordingly, the sustainability analysis of a digital service should be grounded in the assumption that any service operates within a technological, environmental, and organizational life cycle. Even an intangible service — such as an online platform providing seismic data — relies on a complex physical infrastructure to acquire, process, store, and distribute information, including sensors, data centers, servers, telecommunication networks, software systems, databases, and end-user devices. Moreover, such services generate effects that extend well beyond energy consumption or carbon footprint. They influence risk response capacity, the quality of technical and institutional decision-making processes, operational efficiency, and the safety of communities.

Under this framework, sustainability analysis must simultaneously address two distinct yet interdependent dimensions:

• Sustainability of the digital service, referring to the impacts generated by the operation of the system itself, including emissions associated with servers, data processing, network traffic, equipment manufacturing, and hardware life-cycle management.
• Sustainability enabled by the digital service, referring to the positive systemic value generated within the context in which the service operates, such as reduced physical travel, enhanced risk prevention, improved emergency response support, mitigation of environmental damage, and strengthened public safety.

This distinction is particularly relevant for digital services delivering critical data, such as earthquake-related information. Accurate and timely information can substantially reduce economic costs, environmental emissions, and indirect impacts that may significantly exceed the direct footprint of the service itself. For instance, enabling remote technical assessments, supporting earthquake-resistant design, reducing on-site inspections, and facilitating rapid emergency interventions can produce sustainability gains that are not adequately captured by traditional environmental or economic metrics. Conventional assessment tools, such as Life Cycle Assessment (LCA), are designed to quantify impacts but are not inherently structured to evaluate the systemic value generated by digital enablement [3].

To address this conceptual gap, a new methodological construct is required: the Digital Sustainability function. This function can be defined as a mathematical model capable of integrating heterogeneous indicators — environmental, economic, and social — into a single aggregated value that simultaneously captures both the operational impacts of a digital service and the systemic value it enables. Such a function must be flexible and adaptable to different service typologies, and capable of representing non-linear dynamics, where benefits may increase at rates that differ from variations in individual indicators.

The objective of this approach is not solely to quantify the environmental or energetic “cost” of a digital service, but rather to evaluate its broader “value” in terms of risk reduction, process optimization, informed decision-making support, and contribution to collective well-being. This shift represents a transition from an impact-based assessment paradigm to a value-based analytical framework. Such a transition is essential for understanding the strategic role of digital services in achieving global sustainability objectives. Within this perspective, the Digital Sustainability function becomes a foundational methodological framework for evaluating, comparing, and optimizing complex digital services, formally integrating their enabling dimension into sustainability assessment models.

2. Methods and Functions for Sustainability: A Comparative Analysis

The increasing digitalization of contemporary society is profoundly modifying how services are produced, distributed, and used. Data-driven digital services—such as those aimed at monitoring or analysing complex natural phenomena—pose new challenges to sustainability assessment. Classical methodologies, primarily developed for material products and industrial systems, prove only partially adequate when applied to computational systems, cloud infrastructures, algorithmic processes, or immaterial value chains. Consequently, there is a critical need to scrutinize existing methods and explore the possibility of introducing synthetic functions that account for the unique characteristics of digital services, not only from the perspective of environmental impacts but also regarding the economic, social, and systemic value generated.

The Life Cycle Assessment (LCA) represents one of the most established tools for environmental sustainability evaluation [3]. Based on ISO 14040 [4] and ISO 14044 [5] standards, LCA allows for the analysis of the entire life cycle of a product or service, from the production phase to disposal, through a structured and internationally recognized method¹. The strength of LCA lies in its ability to provide a complete picture of emissions and resource consumption. However, its application to digital services raises methodological questions: the definition of the functional unit, the allocation of impacts in multi-tenant cloud environments, and the rapid obsolescence of digital technologies contribute to significant variability in results².

The Product Environmental Footprint (PEF) [6] and the Product Carbon Footprint (PCF) represent attempts to make environmental assessment more homogeneous and comparable [7,8]. PEF, developed by the European Commission, proposes Category Rules (PEFCRs) aimed at standardizing the analysis, thereby limiting the analysts' modeling freedom³. However, the lack of specific PEFCRs for digital services reduces its applicability in the short term. PCF, simpler and more intuitive, focuses on CO₂ equivalent emissions but risks neglecting significant environmental dimensions such as raw material extraction, impacts on biodiversity, and land use.

Environmental Product Declarations (EPDs), based on third-party verified LCAs and harmonized standards ISO 14025 [9], are useful tools for environmental communication⁴. Nevertheless, their application to digital services remains complex: the definition of the functional unit and the lack of specific product categories limit the comparability of results.

A fundamental contribution to the debate comes from the ICT for Sustainability (ICT4S) literature, particularly the work by Hilty and Aebischer [10]. The authors distinguish three levels of impact:

• Efficiency: the reduction of direct ICT impacts.
• Enabling: the capacity of ICT to reduce impacts in other sectors (e.g., energy optimization or improved territory management).
• Transformational: structural changes in the functioning of socio-technical systems.

This distinction is particularly relevant for digital services that provide real-time data or decision support, which can generate indirect benefits far greater than their direct impacts. However, ICT4S lacks a formalized mathematical model to systematically quantify these benefits.

The debate on the definition of sustainability, addressed by authors such as Pezzey [11], highlights the tension between "strong" and "weak" approaches—that is, between biocentric visions prioritizing natural resource conservation and anthropocentric visions oriented toward balancing economic development and environmental protection. This tension is also reflected in sustainability indices, as observed by Mayer [12], who points out the typical critical issues of composite indicators: indicator selection, normalization, weighting, and aggregation. The OECD, in its Handbook on Constructing Composite Indicators, also underscores the risks associated with building synthetic indices, including loss of transparency and potential distortion of information [13].

In the context of urban and territorial policies, comparative studies such as that by Shen et al. [14], highlight how sustainability indicators vary enormously across contexts, making it difficult to obtain a truly comparable framework. This observation also applies to digital services, which exhibit high technological and operational heterogeneity.

The global reference framework is provided by the United Nations' 2030 Agenda, which proposes a multidimensional approach to sustainability articulated in 17 Sustainable Development Goals (SDGs) [15]. While offering a holistic vision, the 2030 Agenda does not provide operational quantitative tools for assessing digital services.

In summary, no single existing method alone manages to capture the complexity of digital services:

• LCA/PEF offer rigor but are limited to the environmental dimension.
• PCF is simple but reductive.
• EPD communicates well but does not easily adapt to digital services.
• ICT4S includes socio-technical dimensions but lacks mathematical formalization.
• Composite indices offer integration but present methodological criticalities.
• SDGs provide a macro-strategic but not operational framework.

This leads to the necessity of developing flexible synthetic functions—such as the Digital Sustainability Functions and the proposed Digital Sustainability Value (VDS)—capable of integrating the impacts, benefits, systemic transformations, and immaterial dimensions of digital services. Such functions are not intended to replace consolidated methods but to provide a complementary tool for interpreting sustainability in complex digital contexts.

3. Digital Sustainability Function and the Definition of the Digital Sustainability Value

The increasing digitalization of public, scientific, and infrastructural services has necessitated the development of new methods to evaluate the sustainability of processes supported by digital technologies. Classical sustainability models were conceived for physical products and infrastructures, whereas digital services possess radically different characteristics: they produce relatively modest direct impacts (electrical energy, hardware infrastructure) but can generate extremely significant indirect impacts by modifying processes, enabling new information flows, enhancing territorial resilience, optimizing resources, or increasing community safety.

This dual nature—material and enabling—requires a mathematical model capable of representing both the service's footprint and the systemic value it produces. From this requirement arises the general function (1) of Digital Sustainability (S_D):

$${\mathit{S}}_{\mathit{D}}=\sum _i{W}_i{\ast f(}_{N_i})$$

(1)

This function is composed of three fundamental elements:

• N_i – Normalized Indicator (0–1). Each indicator is transformed into a dimensionless value between 0 and 1. The indicator can represent environmental aspects (energy, CO₂), economic aspects (cost reduction, efficiency), social aspects (accessibility, safety), quality of service (latency, uptime), or systemic value (enabled benefits).
• f(N_i) – Non-linear Transformation (optional)The transformation allows for the modeling of non-linear effects, such as:

  ○

  Decreasing marginal growth (logarithmic).

  ○

  Activation thresholds (sigmoid).

  ○

  Penalization of low indicators (e.g., quadratic functions).

In the context of digital services, many relationships are non-linear: an improvement in the data update frequency can yield an increase in social value much greater than the percentage variation of the indicator.

• W_i – Weight (summation = 1). Each weight indicates the relative importance of the indicator within the application context. The normalization of weights enables consistent comparisons between different services.

In this form, the function S_D allows for the combination of heterogeneous metrics into a single synthetic indicator, but it only represents the "internal" sustainability of the digital service—that is, how efficient, optimized, and responsible it is from an environmental, economic, and social standpoint. It fails to capture the external value, or the positive impact that the service enables within the system in which it operates.

To address this gap, particularly in digital services with high public or environmental value, a conceptual extension is necessary: the Digital Sustainability Value (V_DS) function (2).

$${\mathit{V}}_{\mathit{D}\mathit{S}}=\sum _i{W}_i{\ast f(}_{N_i})$$

(2)

Mathematically, the structure appears similar, but the meaning of the indicators Ni changes radically: they no longer represent “the sustainability of the service but the sustainability that the service enables”. In other words, V_DS measures what the service facilitates in terms of the overall sustainability of the system.

In digital services that provide critical data—such as those for earthquake monitoring and analysis—this enabled value is often far greater than the direct impact of the service itself. Typical examples include:

• Reduction of physical travel for site inspections, resulting in CO₂ emission reduction.
• Improved timeliness of civil protection interventions.
• Support for anti-seismic design and risk prevention.
• Reduction of environmental damage and economic losses through informed decisions.
• Increased safety for communities exposed to risk.
• Enabling large-scale scientific analysis.
• Transparency and accessibility of seismic data for citizens and administrations.

These effects cannot be described by a traditional LCA, nor by a mere analysis of the service's energy consumption. They require specific, normalized, value-oriented indicators that can be integrated into the V_DS function.

The N_i indicators thus become measures of the enabled capabilities:

• Environmental N_i: reduction in kilometers traveled, reduction in emissions, prevention of environmental damage.
• Economic N_i: reduction in monitoring costs, emergency management costs, and structural damage costs.
• Social N_i: service accessibility, safety improvement, support for public decisions.
• Technical N_i: data quality, timeliness, interoperability.
• Institutional N_i: level of adoption and integration into decision-making processes.

Depending on the context, these indicators may undergo non-linear transformations to represent thresholds, cumulative effects, or resilience dynamics. The value of the V_DS function does not replace the S_D function; the two coexist:

• S_D measures “how sustainable the digital service is”;
• V_DS measures “how sustainable the digital service makes the system”.

A service may have low energy consumption (high S_D) but limited social impact (low V_DS). Conversely, a more energy-intensive service that is crucial for territorial safety and resilience may have a very high V_DS despite significant energy costs. The model, therefore, allows for distinguishing efficiency from generated value, introducing an objective and comparable evaluation framework.

In summary, the V_DS function extends traditional sustainability models to include the value enabled by digital services, representing an essential step toward a Digital Sustainability methodology capable of reflecting the real impact of technologies in contemporary society.

3.1. Disaggregation of the Digital Sustainability Value

The comprehensive assessment of digital services requires the V_DS function to be disaggregated into the fundamental pillars of sustainability (Environmental, Economic, and Social). This decomposition allows for a granular analysis of the systemic value enabled by the service and ensures that the final aggregated value is transparent and interpretable.

The Digital Sustainability Value V_DS is therefore defined as the weighted summation of its three core components (3): Environmental Value (V_env), Economic Value (V_eco), and Social Value (V_soc):

$${\mathit{V}}_{\mathit{D}\mathit{S}}=W_{env}\ast V_{env}{+W}_{eco}\ast V_{eco}{+W}_{soc}\ast V_{soc}$$

(3)

where W_env, W_eco, and W_soc are the main weights (4) applied to the three sustainability dimensions with:

$$\left(W_{env}+W_{eco}+W_{soc}=1\right)$$

(4)

Each of these components (Table 1) is, in turn, defined as a composite function reflecting the aggregation of specific normalized indicators (N).

Operationalization of the Components

• Indicator Selection (N_i, N_j, N_k): Indicators are specifically chosen to measure the enabled benefits (the output of the service), not just the direct footprint (the input). For example, instead of measuring server power consumption (part of S_D), V_env measures the reduction in CO₂ emissions due to the remote execution of an analysis facilitated by the service.
• Internal Weighting (W_i, W_j, W_k): These weights define the relative importance of specific indicators within their own pillar. For instance, within V_soc, the weight assigned to 'Safety Improvement' may be higher than that assigned to 'Data Accessibility'.
• Non-linear Transformation f(N): This function is essential to capture the fact that the benefit generated by the digital service is often non-linear. In fact, in the proposed model, the non-linear function is not necessarily an equation applied a posteriori, but is embedded in the very structure of the scoring classes, as also applied in the indicators of the case study presented in Appendix A. For example, reducing data latency from 5 seconds to 1 second (a small technical change).
• The non-linear distribution of the scores assigned to the indicators is driven by three dynamics:
• decreasing marginal returns (logarithmic function), so that some indicators give greater importance to the early stages of adoption;
• thresholds of accuracy and resolution, in which case the value scale can be calibrated on critical physical thresholds (i.g. V_env);
• synergy effects and "value jumps" that are necessary to capture the combined impact of multiple factors (i.g V_soc).
• Aggregation: The three primary components (V_env, V_eco, V_soc) are then aggregated using the main weights W_env, W_eco, W_soc to obtain the final Digital Sustainability Value (V_DS).

This structured approach ensures that the V_DS framework integrates the multidimensionality of sustainability, moving beyond mere environmental impact to quantify the strategic value of digital services as enablers of systemic sustainability.

4. How to Calculate Digital Service Sustainability Value

The estimation of digital sustainability value—that is, the value generated by digital services—does not rely on direct measurements of emissions or resource consumption. Instead, it must be conceptualized as a form of indirectly enabled environmental value, emerging from enhanced capabilities for environmental damage prevention, improved planning processes, and the reduction of secondary or cascading impacts.

Within this framework, the indicators incorporated into the valuation function must be constructed on the basis of systematic scoring procedures applied to responses elicited through targeted assessment questions. These questions are designed to evaluate the extent to which a digital service can prevent, mitigate, or reduce environmental, economic, and social impacts.

Accordingly, the definition of the indicator structure requires the formulation—tailored to each specific case study—of a set of evaluative questions aimed at assessing:

(i) the service’s capacity to support environmental and economic monitoring activities;

(ii) its contribution to advancing scientific and societal knowledge related to environmental value; and

(iii) its spatio-temporal responsiveness within systems dedicated to preventing territorial damage or reducing secondary effects associated with natural processes or anthropogenic pressures.

To illustrate the operationalization of this methodological framework, the following section presents a case study developed for the evaluation of digital services that rely on environmental data, computational models, and decision-support systems for knowledge generation, monitoring, and the prevention of natural hazards and risks linked to natural resources.

4.1. Evaluation Example of an Earth Science Digital Service

In this specific context, we analyze how to estimate the weights of the three components that constitute the digital sustainability value of a service, using as a case study a macro-service focused on seismic hazard and risk of TCS Seismology within the EPOS Platform [16,17]. This service provides data/service delivery and interactive products that are highly relevant for the analysis of seismic hazard and risk in Europe.

In order to assign a value to digital service, it was necessary to define a structured set of indicators across the three dimensions of sustainability—environmental, social, and economic. The methodological approach adopted begins with a detailed description of the digital service and an initial qualitative analysis of its objectives, functionalities, and areas of application. This preliminary analysis enabled a systematic mapping of the service to the targets of the Sustainable Development Goals (SDGs) [15] and, subsequently, to the indicators associated with those targets, with the aim of identifying which sustainability aspects are potentially addressed by the service.

The analysis showed that the macro-service focused on “Earthquake hazard and risk” contributes, either directly or indirectly, to the following SDG targets: 1.5, 4.4, 9.5, 11.5, 11.b, 13.1, 16.6, and 17.7. Based on the indicators defined at the UN level for each of these targets, a structured matrix of questions was developed to collect more detailed and specific information on the characteristics, uses, and impacts of the macro-service. The information gathered through this matrix was subsequently integrated with the metadata of the individual digital services composing the macro-service, allowing for a comprehensive and coherent reconstruction of the overall information framework.

In other words, the proxy value of the enabling function provided by the digital service was translated into measurable indicators, in accordance with the definitions outlined in Table 1 and considering specific value drivers such as: the informational capacity of the service, its level of use and adoption, and its relevance for risk management and disaster response.

Once identified, the indicators N_i were calculated for the Earthquake hazard and risk macro-service by means of an aggregated assessment of the information provided by each individual sub-service (a total of twelve sub-services were identified). This assessment combined the analysis of sub-service metadata with a quantitative coding of the responses provided by the macro-service producer community, collected through a standardized questionnaire administered to them. The final list of indicators is presented in Table 2, while Appendix A provides the complete list of indicators, including the calculation methods which, in some cases, through variable discrete classes, reflect the use of the non-linear function described in the methodology for calculating the enabled value.

After this process, we moved on to evaluating the weights to be assigned to each component of the sustainability value. In general, the method for weight attribution should be based on a multi-criteria analysis framework (e.g., the Delphi method). However, in this particular case, it is evident that the weight assigned to the social component—namely territorial safety and the preservation of human lives—is greater than that attributable to the economic component (building and infrastructure damage reduction) and, simultaneously, to the environmental component (mitigation of environmental impacts in terms of ecosystem disruption and the release of pollutants into soil or water).

It is evident that, when evaluating the three components within their respective domains of expertise, each domain tends to assign the highest weight to its own component, while consistently attributing a medium-to-high weight to the social dimension, particularly with regard to the protection of human lives. By averaging the resulting assessments, one can obtain a set of weights such as the following: W_soc = 0.45; W_env = 0.29; W_eco = 0.26.

Subsequently, following the evaluation of the weights and the definition of indicators for the individual components, the Digital Sustainability Value of the “Earthquake Hazard and Risk” macro-service was computed using Equation V_DS presented in paragraph 3.1.

In this phase, the value of the components was obtained as the sum of their respective indicators (Nᵢ), which, as detailed in Appendix A, had previously been normalized within a range between 0 and 1.

$$\mathit{E}\mathit{a}\mathit{r}\mathit{t}\mathit{h}\mathit{q}\mathit{u}\mathit{a}\mathit{k}\mathit{e} \mathit{H}\mathit{a}\mathit{z}\mathit{a}\mathit{r}\mathit{d} \mathit{a}\mathit{n}\mathit{d} \mathit{R}\mathit{i}\mathit{s}\mathit{k} {\mathit{V}}_{\mathit{D}\mathit{S}}=0.29\ast \left(4.2\right)+0.26\ast \left(2.8\right)+0.45\ast \left(3.8\right)$$

(5)

The final result for the “Earthquake Hazard and Risk” service is V_DS = 3.66. This value should be compared to a theoretical maximum score of 5.9, which would be achieved if all indicators across the three dimensions (environmental, economic, and social) reached the maximum normalization score. This benchmark of 5.9 is derived from the weighted sum of the maximum values of each component, considering the number of indicators mapped for each pillar.

5. Goal and Conclusion Archived

The objective of this study is to shift the paradigm for evaluating digital or information services, which are often interpreted in a restrictive manner primarily in terms of their impact on environmental, social, and economic sustainability. By introducing the concept of digital sustainability value, this work seeks to assign a positive weight to digital services, recognizing them as enabling elements and as producers of information that is essential for the assessment of sustainability indicators related to the SDGs.

Within this context, digital services assume a systemic and enabling role in achieving environmental, economic, and social sustainability objectives. Their benefits are therefore assessed in relation to their conscious and responsible use. The digital sustainability value of a service should be contrasted with the impacts associated with energy consumption and the use of natural resources required by the underlying information infrastructure, including both hardware and software components.

The balance between these two dimensions—namely, the value generated and the associated impacts—allows for a meaningful assessment of how a digital service, or a set of services provided by a technological infrastructure, represents a tangible benefit for the community and contributes effectively to the achievement of the planet’s sustainable development goals.

The subsequent step consists of identifying a technological infrastructure that delivers digital services, such as the EPOS platform, and assessing not only the digital sustainability value of all the services it provides, but also their environmental and economic impacts. This approach enables a comprehensive evaluation of the actual benefits generated by the services as a whole. Indeed, it can be argued that as long as the digital sustainability value of an infrastructure exceeds the impacts it produces, the digital services delivered generate net benefits and contribute positively to the achievement of the SDGs and to overall sustainability.