Customer value associated with 

electric smart meters in South Africa. 

 

 

Desmond Mabilo 

0410840F 

 

A research article submitted to the Faculty of Commerce, Law and 

Management, University of the Witwatersrand, in partial fulfilment of the 

requirements for the degree of Master of Business Administration 

 

Johannesburg, 2023 

 

Protocol number: WBS/BA0410840F/335 

(April 2023) 

 

 

 



 

ii 

 

DECLARATION 

 

I, Desmond Mabilo, declare that this research article is my own work except as 

indicated in the references and acknowledgements. It is submitted in partial fulfilment 

of the requirements for the degree of Master of Business Administration in the 

Graduate School of Business Administration, University of the Witwatersrand, 

Johannesburg. It has not been submitted before for any degree or examination in this 

or any other university. 

 

 

Desmond Mabilo 

Signed at ……… Plettenberg Bay (Western Cape)……………………………………… 

 

On the ………28th…………………….. day of ……April………………… 2023…. 

 



 

iii 

 

DEDICATION 

This research is dedicated to my wife Itumeleng Lepelesana Mabilo and my daughter 

Onalerona Mabilo.  

To my wife Itumeleng, thank you for your sacrifice and bearing with me through my 

studies, thank you for believing in me and continuously encouraging me, I love you. 

To my daughter Onalerona, whatever you do trust in God always, and go all the way. 

Daddy loves you. 

 

 

 

 

 

 

 

 

 



 

iv 

 

ACKNOWLEDGEMENTS 

• I would like to thank the Lord Jesus Christ, for granting me strength and 

wisdom through the research process. 

• I would like to acknowledge the invaluable contribution made by my 

supervisor Dr Pius Oba for his guidance and willingness to assist as required.  

• I would like to express my gratitude to all persons who took their valuable time 

to respond to my survey.  

 

 

 

 

 

 

 

 

 

 

 

 



 

v 

 

SUPPLEMENTARY INFORMATION 

   

Nominated journal: 
 
 
Supervisor / Co-Author: 
 
Word Count †: 

Manufacturing and Service Operations 
Management Journal. 
 
Dr Pius Oba 
 
13 823 
 

 

   

† Including abstract references, etc. 

 

 

 

 

 

 

 

 

 

 



 

1 

 

ABSTRACT 

Problem definition: Electric smart meters are gaining popularity in utilities and 

municipalities across the globe as a smart grid driver introducing various benefits 

to utilities such as security of revenue and the reduction of resources waste, 

among other factors. However, various studies across the globe indicate mixed 

responses on the value that electric smart meters bring to end customers. In this 

study, the customer value associated with electric smart meters is interrogated in 

the context of residential South African customers. 

Academic/Practical relevance: The theory of customer value which is premised 

on the difference between the benefits and costs of electric smart meters is 

adopted. The benefits and costs of electric smart meters are looked at from an 

operations management perspective, where the benefit elements are the five 

external operations performance objectives (Speed, Cost, Quality, Flexibility and 

Dependability) and the cost element refers to the costs that customers incur to 

access or utilise the capabilities of smart meters.  

Methodology: A quantitative research methodology is utilised to determine the 

customer value of electric smart meters. Because operations performance 

objectives cannot be measured directly as well as the costs, a Structural Equation 

Model (SEM) is utilised to set the benefits and cost elements as latent variables 

indicated by various responses from a survey distributed to a total of 365 

respondents.  

Results: The resulting analysis through SEM indicated that South African 

residential customers find value in electric smart meters. The value is driven 

mainly by the Speed, Cost and Dependability capabilities of electric smart meters. 

Customers are found to be neutral with regard to the Flexibility and Quality 



 

2 

 

capabilities of electric smart meters. The main cost element of smart meters 

affecting customer value is observed as the internet or data fees in South Africa, 

which is a requirement to fully benefit from the capabilities of smart meters. Lastly, 

it is observed that there is no difference in the customer value for the different 

customer groups based on gender and age.  

Managerial implications: This study provides insight relevant to municipalities, 

utilities and organisations operating, planning to introduce or deploy residential 

electric smart metering infrastructure in South Africa to consider the value that 

electric smart meters bring to customers. These organisations should ensure that 

the dominant electric smart metering functions (Speed, Cost and Dependability) 

are visibly implemented to achieve maximum customer satisfaction and impact. 

Furthermore, it is worth noting the negative impact caused by high internet or 

data costs on the value of electric smart meters to residential South African 

Customers. 

Keywords: Electric Smart Meters, Customer Value, Operations Performance 

Objectives, Structural Equation Modelling (SEM). 

 

 

 

 

 

 

 

 



 

3 

 

1. INTRODUCTION  

1.1 Purpose of the study 

The purpose of this study is to determine the customer value of electric smart 

meters by residential customers in South Africa. Furthermore, the study will 

determine dominant electric smart metering technology functions that create 

value for residential customers in South Africa. Lastly the study will further identify 

whether there are significant differences in values by the different residential 

customers in South Africa. 

1.2 Context of the study 

Smart metering technologies are described as a technology composed of a near 

real-time two-way communication between smart electricity metering devices at 

consumer premises and data management systems residing at utility premises 

or the cloud (US Department of Energy, 2016). Furthermore, smart metering 

technologies enable utilities the capability to provide visibility to their customers 

their energy consumption and associated events in a near real-time fashion. This 

data is mainly provided to consumer using web portals, mobile applications and 

In-home displays using capabilities of the internet to access data residing in the 

utility data centres or cloud infrastructure (US Department of Energy, 2016).  

Following global trends, several South African municipalities have been 

considering the deployment of electricity smart meters and smart metering 

systems (Advanced Metering Infrastructure) to improve revenue management, 

reduce electricity losses, address billing and credit control issues, promote 

efficiency and improve service delivery to customers (SEA, 2015). The 

implementation of these smart metering projects in South Africa has been seen 

in the City of Johannesburg, the City of Tshwane, the Nelson Mandela Bay 

Municipality, and the eThekwini Municipality (SEA, 2015; Zyl, 2018). 



 

4 

 

According to a report published by Grand View Research in the year 2018, the 

smart meter market is expected to grow at the Compound Average Growth Rate 

(CAGR) of about 5.1% from 2018 to 2025 (from $54.1 million in 2017 to $79,5 

million by 2025) (Nhede, 2018; Zyl, 2018). Furthermore, (Nhede, 2018; Zyl, 2018) 

state that smart meters are gaining popularity among South African users due to 

introduced benefits such as prepayment and easy-to-use functions. On the other 

hand (Ball, 2019) states that according to energy professionals in a survey by the 

Energy Institute, smart meters only benefit companies and not consumers.  

Smart meters are not understood to add value to consumers as they do not 

provide opportunities for consumers to reduce their bills (Ball, 2019). Furthermore 

(Völker, Reinhard, Faustine, Pereira, 2021), state that the use-cases of electric 

smart meters have customer benefits that are far less than the benefits 

experienced by utilities or the grid in general. Based on this uncertainty on the 

value that smart metering technologies bring to consumers; research is 

undertaken to identify Customer Value associated with electric smart meters in 

South Africa for residential customers. 

1.3 Problem statement 

The main research problem is to empirically identify the customer value 

associated with electric smart meters in South Africa. Identifying the customer 

value associated with electric smart meters in South Africa will assist utilities, 

municipalities and city planners in deploying electric smart metering technologies 

in a manner that not only benefits utilities as stated by (Ball, 2019; Völker et al., 

2021b) but also benefits customers in South Africa.  

1.3.1 First sub-problem 

The first sub-problem of the study is to determine the customer value of electric 

smart metering functions in South Africa. 



 

5 

 

1.3.2 Second sub-problem 

The second sub-problem of the study is to identify the dominant customer value 

amongst the different electric smart metering technology functions.  

1.3.3 Third sub-problem 

The third sub-problem of the study is to identify if there is a difference in customer 

value of smart metering technologies for different customer types.  

1.4 Significance of the study 

The literature reviewed did not identify research investigating the value 

associated to electric smart meters for residential customers in South Africa. This 

research, therefore, contributes to the body of knowledge by providing insight on 

the identified gap. The research findings are aimed to provide insight to managers 

and leaders of electricity utilities and municipalities to adopt smart metering 

technologies that also address the specific needs of customers in South Africa. 

Prior to the roll-out of smart metering technologies, South African utilities and 

municipal leaders can communicate these benefits that align to the needs of 

customers, thus eliminating any scepticism that may arise as that observed in 

similar economies such as Indonesia (Chou & Yutami, 2014).  

1.5 Delimitations of the study 

The following do not form part of the study: 

● Research on other types of smart meters, such as water, gas etc. Only 

electric smart meters are considered. 

● Utilities and municipalities outside the borders of South Africa. 



 

6 

 

Furthermore the study does not cover any research methodology outside the 

following: 

• Quantitative empirical study based on survey questionnaire distributed to 

South African residential customers. 

1.6 Assumptions of the study 

The following assumptions are made in the study: 

● Respondents to the survey questions will answer honestly. 

2. LITERATURE REVIEW  

2.1 Introduction 

The aim of this section is to outline from the literature the definition of electric 

smart meters and associated functions. This discussion will also cover how 

electric smart metering technologies are implemented by various nations across 

the globe, the benefits realised, and the challenges encountered. This discussion 

is, thereafter, followed by a presentation of management theories from the 

literature that is applicable to addressing the research problem identified in this 

study. 

The management theories identified are the theory of customer value and the five 

operations management performance objectives. Through the literature review of 

electric smart meters, their functions and implementation across the globe, 

research questions are developed. The literature review of management theories 

are to develop a guiding principle to answer the research questions and address 

the problem statement.  



 

7 

 

2.2 Electric Smart Meters 

2.2.1 Background 

Electric smart meters are an evolved technology which began as devices installed 

in private households, commercial buildings, industrial sites and other areas 

measuring electricity with rotating disc devices with mechanical displays (Völker 

et al., 2021). This technology required utilities to deploy persons to collect 

readings from different sites using paper-based capturing of consumer electricity 

consumption, leaving room for human error. The rise of digital technologies gave 

birth to smart meters, with capabilities of measuring electricity accurately and 

transferring electricity consumption to utility IT systems through 

telecommunications technologies (US Department of Energy, 2016; Völker et al., 

2021).  Electric smart metering technologies enable utilities the capability to 

provide visibility to their customers their energy consumption and associated 

events in a near real-time fashion. This data is mainly provided to consumer using 

web portals, mobile applications and In-home displays using capabilities of the 

internet to access data residing in the utility data centres or cloud infrastructure 

(US Department of Energy, 2016). On Figure 1 is an end-to-end utility information 

technology systems and load control capabilities enabled by an electric smart 

meter technology, introducing various benefits to both utilities and their 

customers. 



 

8 

 

 

Figure 1: Electric smart meter enabled utility information technology 

functions. Adapted from Advanced Metering Infrastructure and customer systems. 

Results from the Smart Grid Investment Grant Program. By US Department of Energy, 

2016 

(https://www.energy.gov/sites/prod/files/2016/12/f34/AMI%20Summary%20Report_09-

26-16.pdf) 

2.2.2 Benefits 

Smart metering technologies are considered the first step that needs to be taken 

by countries in pursuit of sustainable development and energy efficiency (Chou 

& Yutami, 2014). In a report by Gill & Judith (2008), smart meters provide the 

benefits listed below. 



 

9 

 

● Accurate customer billing 

● Reduced electricity bills due to better feedback on usage to customers 

● Flexibility to use either pre-pay or post-pay 

● Less carbon emission 

● Advanced tariffs 

Many economies globally are seen investing in smart metering technologies that 

report energy usage of the consumers and provide feedback to consumers on 

their energy usage with the aim of changing their behaviour to reduce energy 

consumption (Echeverri et al., 2017; Mogles et al., 2017). In economies such as 

India, smart meters have been deployed to overcome the challenges of the Indian 

power system, such as billing inefficiencies, theft, unreliable energy delivery and 

distribution companies’ financial problems (Chawla, Kowalska-Pyzalska & 

Skowrońska-Szmer, 2020). These features show that smart meters not only 

benefit utilities, but also provide an advantage for consumers by allowing them to 

save costs, on electricity usage by being nudged to behave differently with 

regards to their electricity usage. Consumers can also be left at ease, by being 

provided an efficient billing systems and early detection of any electricity supply 

problems. 

In a report by a House of Commons Science and Technology Committee, 

published in the year 2016, it is outlined that smart meters realise benefits for 

both consumers and utilities and energy providers as well as the society 

(Sovacool, Kivimaa, Hielscher & Jenkins, 2017). Table 1 below summarises the 

different benefits. 

 

 



 

10 

 

Table 1: Smart meter benefits 

Customers 
Utilities and Energy 

Providers 
Society 

Easy switching between 

energy suppliers 

Savings from 

eliminated sites visits 

for data collection 

 

Enable optimisation of 

electricity generation and 

network management 

Accurate billing Reduced call centre 

traffic, due to less 

inaccurate-billing query 

issues 

 

Visibility of the network, 

eliminating the need for 

increased reserve generation 

Reduced debt 

accumulation, realised by 

near real-time billing 

Theft detection and 

optimal management of 

debt 

Innovation and development 

of new business models and 

entrants 

 

    Enabling the initiatives to 

meet climate change 

requirements (e.g., reduced 

carbon emissions). 

Adapted from Vulnerability and resistance in the United Kingdom’s smart meter 

transition. By Sovacool, Kivimaa, Hielscher & Jenkins, 2017, Energy Policy, 109, p. 767–

781 

Although the benefits are presented in the three categories, the benefits realised 

by the utilities and energy producers can be of value to consumers. Such may 

include reduced call centre traffic, resulting in reduced waiting time for customers 

when making queries. Meeting climate change needs could be a valuable need 

for customers as well. (Sovacool et al., 2017) state that there are more than 67 

benefits of smart meter functions that overlap across the personas listed in Table 

1. Based on this, research on customer value of different smart meter functions 

by residential customers in South Africa is found to be necessary to understand 

the need with a South African context. 

2.2.3 Challenges 



 

11 

 

Due to limited awareness, knowledge and understanding of smart meter 

technology, consumers in economies such as Indonesia have shown some 

hesitance in the adoption of smart meters in their buildings (Chou & Yutami, 

2014). The scepticism and hesitation have hindered progress in a journey for 

energy efficient lifestyles by consumers by not leveraging the benefits of smart 

meter technologies. To ensure progress, smart metering technologies need to be 

presented in a way that addresses the needs (including latent) of consumers.  

In a study by Lineweber (2011), a survey was undertaken on the important 

benefits of smart meter functions. It was discovered that there is large gap in the 

implementation of smart grid technologies, where the benefits are not clearly 

communicated to and understood by the consumers in the United States. The 

limitation in communication of the smart grid technologies resulted from an 

approach that sought to educate consumers about smart grid technologies in a 

shallow fashion, not focussing on communicating the downstream benefits of 

smart metering technologies. As a result, consumers could only see a fancy 

technology without deep insights on the benefits they can realise from it. This gap 

contributed to consumers not finding value and rejecting smart grid and smart 

metering technologies developed not only for the utility benefits but also 

customers’ benefits as well. Based on this, a clear communication with deep 

insights on the downstream benefits of smart metering technologies relevant to 

the consumer group is necessary for the success of smart metering technologies 

implementation. Understanding of this relevant downstream functions is therefore 

essential. 

2.3 Customer Value 

Customer value emerges from the perspective of marketers and customers, 

described as a theory which determines the value that customers see, 

experience, and perceive of a product (Woodside, Golfetto & Gibbert, 2008). 

Customer value is further described as the ability for companies to create and 



 

12 

 

add value to goods and services they deliver to customers, this value can be 

delivered at four value points namely, Service, Quality, Image and Price (SQIP) 

(McFarlane, 2013). The success of this value lies in the reaction of the customer 

to the value delivered, which makes the customer central to the success of the 

value. Customer value is also influenced by the costs that customers incur in 

acquiring goods or services, and this results in customer value described by the 

formula below (Woodside et al., 2008). 

Customer Value = Sum of benefits – Sum of Costs incurred 

Customer value is also described as not just a marketing and customers’ concept, 

but it also can be adopted as an organisation-wide strategy and management 

approach to gain competitive advantage in the market (Setijono & Sandberg, 

2005). They further argue that quality and operations management activities in 

organisations have a great influence on customer perceived value. This implies 

that the performance of quality management and operations management 

initiatives can influence the value perceived by customers. 

In the case of the deployment of electricity smart meters by utilities, it is essential 

that a positive perceived value by customers is realised, such as creating 

opportunities for electricity savings and meeting other customer preferences and 

needs (Albani, Domigall & Winter., 2017). Although utilities may realise benefits 

from electricity smart meters, not meeting customer preferences and needs 

poses a danger to utilities such as the rejection of electricity smart meters, as 

observed in economies such as Indonesia, USA etc. (Chou & Yutami, 2014; 

Lineweber, 2011). In a study on German household customers, Gerpott & 

Paukert (2013) performed an empirical analysis to determine the customers’ 

willingness to pay for smart meters based on three perceived smart meter 

benefits listed below: 

● Usefulness of consumption feedback 

● Trust in data protection 



 

13 

 

● Environmental awareness 

This resulted in trust in data protection and consumption feedback as main drivers 

for changing consumer behaviour in electricity usage and increased the 

willingness to pay for smart metering technologies. This study focussed on only 

a few electricity smart metering benefits, leaving the question of whether the 

willingness to pay would increase or decrease if more of the electricity smart 

metering benefits were considered. Therefore, a more comprehensive study 

considering much more if not all smart metering benefits is considered here, 

amongst these functions the dominant smart metering functions need to be 

determined as well. From a South African context similarly to other economies, 

determining the dominant smart metering functions will help with addressing 

critical needs of consumers, and as a result deliver them ensuring customer 

satisfaction in a focussed manner. 

In a study by Barnicoat & Danson (2015), it was discovered that the older 

generation in Scotland, who are on the low-income band, have high value for 

smart meters due to the consumption feedback enabling them to change their 

behaviour to save electricity. Based on this, determining the value perceived by 

different customer types (based on varying age and gender) is required in the 

South African context to understand if there would be any challenges of varying 

customer values by different customer types in South Africa. Varying customer 

types add to the complexity in implementation, if not prepared the situation can 

be much unfavourable. 

2.4 Performance measurement 

Introducing smart metering technologies is an exercise that requires substantial 

capital investment that involves the replacement of existing equipment and the 

acquisition of new technology (Echeverri et al., 2017). To ensure benefits and 

prevent losses from substantial capital investments such as smart metering 

technologies, the adoption of operations management principles is necessary, to 



 

14 

 

ensure that the operations meet the needs intended by measuring the 

performance using concepts such the 5 operations performance objectives. The 

ability to respond to customer needs is a crucial part in ensuring operations 

preparedness (Slack, Alistair, Robert, Hemmanth & Khomotso, 2017). 

Furthermore, Setijono & Sandberg (2005) state that customer value has a direct 

relationship with quality and operations management activities in organisations. 

The output of quality and operations management activities influence the 

customer value. Knowing customers’ needs and challenges is very crucial for 

operations and quality managers to enable them to determine what they need to 

do to deliver a successful operation service (Slack et al., 2017). 

Due to the substantial capital investment required to introduce smart metering 

technologies, it is crucial that utilities possess an optimal level of operations 

management preparedness. Operations preparedness is achieved when the 

smart metering technologies meet the performance objectives of both the utilities 

and customers referred to by Slack et al. (2017) as the internal and external 

performance objectives.   Due to the perceived limited focus and benefits of smart 

metering technologies to customers espoused by Ball (2019) and Völker et al. 

(2021), this study is undertaken with a focus on the external performance 

objectives which address the benefits to customers.  

2.4.1 Performance objectives 

Both external and internal performance objectives in an operations environment 

such as in the adoption of smart metering technologies can be determined by five 

elements, namely Speed, Quality, Flexibility, Dependability and Cost as depicted 

on Figure 2 (Slack et al., 2017).  



 

15 

 

 

Figure 2: 5 Performance Objectives of Operations Management. Adapted from 

Operations Management: Global and Southern African Perspectives (3rd ed.), by Slack, 

N., Alistair, B.J., Robert, J., Hemmanth, S., & Khomotso, P, 2017, Pearson. 

Speed refers to the elapsed time between a customer requesting a product or 

service and the delivery thereof (Slack et al., 2017). From a smart metering 

perspective, speed can be considered as the time it takes for a consumer to 

receive feedback from the smart meter on their consumption (Echeverri et al., 

2017; Mogles et al., 2017). Traditionally consumers did not have the opportunity 

to get feedback on their electricity consumption, but only received one reading at 

the end of the month mainly for the purpose of billing (Gill & Judith, 2008).  

Quality refers to consistent conformance to customer expectations on the 

product or service delivered resulting in either satisfaction or dissatisfaction 

(Slack et al., 2017). From a smart metering perspective, quality can represent the 



 

16 

 

quality of consumption readings by the smart meter, this can include less 

estimations in readings used for billing (Gill & Judith, 2008). 

Dependability refers to doing things on time for customers, exactly when they 

are needed and in the correct format (Slack et al., 2017). From a smart metering 

perspective, this element can look at on-time delivery of their electricity bills (Gill 

& Judith, 2008). Dependability can also be represented by the accuracy of 

consumption readings which customers use to change their behaviour in 

electricity consumption to reduce their bills (Echeverri et al., 2017; Mogles et al., 

2017). Incorrect reporting can cause consumers to be frustrated when their 

change in behaviour does not produce an impact on electricity bills. 

Flexibility refers to the ability of an operation to accommodate changing 

customer needs depending on circumstances (Slack et al., 2017).  From a smart 

metering perspective this considers the opportunity for customers to change their 

usage patterns, such as the ability for customers to choose whether to use 

electricity in a post-paid or prepaid fashion, as well as flexible tariffs such as Time 

of Use (TOU) (Gill & Judith, 2008), where customers have the flexibility to access 

different electricity rates based on the hour or days when electricity is used.  

Cost refers to the provision of opportunities for customers to acquire goods and 

services and receiving good value (Slack et al., 2017). From a smart metering 

perspective, opportunities can be created for consumers to pay less for electricity 

by creating tariffs that have reduced pricing during off-peak periods (Echeverri et 

al., 2017; Mogles et al., 2017). Consumers can plan and reserve their heavy load 

electricity usage for off-peak periods and save on electricity costs. Through 

feedback on their electricity usage, consumers can also identify appliances that 

contribute to high electricity usage and change to more energy efficient 

appliances.  



 

17 

 

2.5 Customer value and performance measurement 

In an article by (Amini, Falk, Schmitt, 2014), a quantitative analysis of consumer 

perceived value deviation was undertaken. In this article consumer perceived 

value (CPV) of two different tablet PCs is analysed wherein the CPV construct is 

derived from benefits and costs elements. The CPV components of the benefit 

and the cost elements are depicted Figure 3. 

 

Figure 3: CPV Construct and its Contributing Elements. Adapted from 

Quantitative Analysis of the Consumer Perceived Value Deviation. By Amini, P, Falk, 

B, & Schmitt, R, 2014, Procedia CIRP, 21, 391–396.  

In this study the Customer Perceived Value of electric smart meters is analysed, 

the benefits elements are considered as the identified five performance 

objectives (Speed, Quality, Flexibility, Dependability and Cost) of electricity smart 

metering technologies. The costs element is composed of the costs associated 

with using smart metering technologies such data usage in viewing and 

downloading historical data on mobile devices, the cost associated with using 

electricity services from utilities, instead of using other alternatives such the 

installation of solar and battery storage at customer premises (Naidoo, 2021).  

Figure 4 demonstrates the five performance objectives through which Customer 

Value relationship as a concept can be used to understand and answer the 

research questions of this study. 



 

18 

 

 

Figure 4: Customer Value Relationship with Performance objectives as 

corresponding elements. 

From Figure 4, the Customer Value associated with electric meters can be 

represented by the below equation: 

𝐶𝑉 =
1

𝑛
 ∑ 𝑥𝑖 ∗ 𝐵𝐸𝑖

𝑛
𝑖=1 −

1

𝑚
∑ 𝑦𝑗 ∗ 𝐶𝐸𝑗

𝑚
𝑗=1        (1) 

Where CV = Customer Value; BE and CE are the benefit  and cost elements, 

respectively; n and m are the number of benefit and cost elements respectively; 

and x and y are the weights corresponding to the benefit and cost elements 

relating to the customer value, respectively (Amini et al., 2014). 

In the study by (Amini et al., 2014), the Customer Perceived Value is the key 

construct with a causal relationship with the benefits and costs elements 

described by equation 1. The benefits and costs elements were not directly 

measured, rather other attributes describing the elements could be measured. A 

Structural Equation Model was constructed to describe the key construct (CPV), 

with the benefit and cost elements as latent variables. 

In this study the same applies, the perceived Customer Value is identified as the 

key construct where a Structural Equation Model (SEM) is constructed with the 5 

performance objectives and the cost elements as the latent variables. On Table 

2 is the list of benefit and cost elements and corresponding smart metering 

functions they are described based on the insights gathered. 



 

19 

 

Table 2: Smart metering functions describing the benefit and cost elements 

of the CPV construct 

Benefit Elements 
(Operations 

performance objectives) 
Smart metering functions describing benefit 

elements 

Speed 

Near real-time feedback on electricity usage 

Consistent periodic generation of electricity bills (No 
delays due to missing collection of consumption reading) 

Quality 

Accurate customer billing (significant reduction in 
estimated billing) 

Carbon emission reporting (triggering customers fro 
carbon emission awareness) 

Advanced tariff and visibility to customers 

Automatic meter reading (elimination of manual meter 
reading by utility personnel) 

Flexibility 

Historical electricity usage and forecast, allowing 
consumers to budget electricity usage. 

Flexibility to use either pre-pay or post-pay methods 

Innovation and development of new business models 
and entrants 

Alerts on behaviour change to prevent load shedding 

Advanced tariffs enabling integration of renewable 
energy (such rooftop solar)  

Dependability 

Theft detection and optimal management of debt to 
protect financial health of utilities. 

Better management and optimization of electricity 
generation and network management for reliable 
electricity delivery. 

On time delivery of electricity bills. 

Alerts on electricity outages before they occur. 

Alerts on low electricity credits. 

Cost 

Reduced electricity bills, triggered by near real-time 
electricity usage feedback. 

Advanced tariffs visible to customers with different rates 
based on time of day (Peak, Off-peak) 

  

Cost Elements Smart metering functions describing cost elements 

Costs 

Cost of electricity 

Cost of internet services to access electric smart 
metering functions (Web or Mobile devices) 

Based on the above, the first research question arising from the first sub-problem 

is derived as follows: 



 

20 

 

● What is the customer value associated with electric smart metering 

functions in South Africa? 

This first question will determine if there is positive or negative customer 

perceived value by South African residential customers responsible for 

monitoring and or purchases of electricity. This question aims to determine if 

claims made by (Ball, 2019) and (Völker et al., 2021a) that the negative 

perception on the Customer value of electric smart meters, is applicable to South 

African residential customers. 

Furthermore in a study by (Kaufmann et al., 2013) it was identified that smart 

metering can provide a variety of services and that the different services 

contribution to the customer perceived value and willingness to pay varies. 

Furthermore in a study by (Muchenje & Botha, 2021) to identify Consumer-centric 

factors for the implementation of smart meter, it was identified that there are 10 

consumer centric factors that were significant to smart meter customers. Based 

on this, the second research question is developed which aims to determine if 

smart metering functions grouped according to the 5 performance objectives 

(Speed, Quality, Flexibility, Dependability and Cost) contribution to customer 

value differ amongst South African residential customers. Furthermore the 

questions also seek to determine the dominant smart metering functions 

contributing to the customer value. Below is the hypothesis arising from this 

question: 

H0: Null Hypothesis – Smart metering functions contribute equally 

to customer value. 

H1: Alternate Hypothesis – Smart metering functions’ 

contributions to customer value vary.   

In a study by (Barnicoat & Danson, 2015), it was discovered that the older 

generation in Scotland, who are on the low-income band, have high value for 

smart meters due to the consumption feedback enabling them to change their 



 

21 

 

behaviour to save electricity. Based on this insight, a further question is 

developed to understand the customer value of different customer types and to 

determine if there is difference in the customer value of the different customer 

types. In this case the different customer types are identified based on age (18-

24, 25-34, 35-55 and above 55 years) and gender (Male and Female). Below is 

the hypothesis developed for this 3 research question: 

● Is there a difference in the customer value of smart metering technologies 

by different customer types?  

H0: Null Hypothesis – There is no difference in the customer value 

associated with electric smart meters by different customer types. 

H1: Alternate Hypothesis – There is a difference in the customer 

value associated with electric smart meters for different customer 

types. 

2.6 Conclusion of literature review 

To address the problem identified for this study, entailing determining the 

customer value of electric smart meters by residential customers in South Africa, 

a thorough literature review of electric smart meters functions is obtained. Based 

on the obtained functions, benefits and challenges of electric smart meters are 

outlined, furthermore insights based on implementation of smart metering 

technologies in other economies across the globe are obtained.  

To respond to the identified problem, the theory of customer value is adopted 

which derives customer value based on the difference between the benefits and 

costs of electric smart meters to consumers. From the literature review it is 

discovered that introducing smart metering technologies, costs organisations to 

substantial capital investments. In order to ensure optimal results, operations 

management principles must be adopted. Therefore, the benefit elements of the 

customer value of electric smart meters are considered from an operations 

management perspective, described by the 5 external operations performance 



 

22 

 

objectives (Speed, Quality, Flexibility, Cost and Dependability), the cost elements 

are considered as the costs incurred in accessing the capabilities of functions of 

electric smart meters.  

Based on the described literature review and consideration of the identified 

problem, the below research questions are developed. 

● What is the customer value associated with electric smart metering 

functions in South Africa? 

● What is the dominant smart metering technologies functions contributing 

to customer value? 

H0: Null Hypothesis – Smart metering functions contribute equally 

to customer value. 

H1: Alternate Hypothesis – Smart metering functions’ 

contributions to customer value vary.   

● Is there a difference in the customer value of smart metering technologies 

by different customer types?  

H0: Null Hypothesis – There is no difference in the customer value 

associated with electric smart meters by different customer tes. 

H1: Alternate Hypothesis – There is a difference in the customer 

value associated with electric smart meters for different customer 

types. 

3. RESEARCH METHODOLOGY 

3.1 Introduction 

The purpose of this study is to identify electric smart metering technology 

functions that create value for residential customers in South Africa. This section 

discusses the research design and methodology undertaken to fulfil the purpose 

of the study, by answering the questions below: 



 

23 

 

● What is the customer value associated with electric smart metering 

functions in South Africa? 

● What is the dominant smart metering technologies functions contributing 

to customer value? 

● Is there a difference in customer values of smart metering technologies by 

the different customer types?  

Furthermore, this section will outline, the population, sample size, sample 

method, the survey instrument, how data will be collected and analysed as well 

as the validity and reliability of the survey instrument to be used.  

3.2 Research method 

In this study an empirical research methodology is adopted, which is described 

as a systematic observation approach employing naturalism, behavioural and 

explanation to gather evidence, using scientific procedures to answer research 

questions. There are two types of empirical research methodologies, namely the 

qualitative and quantitative methods. In this research a quantitative methodology 

is adopted, which is described as a methodology where the data collected must 

be quantified, this methodology involves numbers (Scherbaum & Shockley, 

2015). A pragmatists paradigm is adopted in this research due to the nature of 

the problem at hand which considers an introduction of technology that demands 

change in how customers can interact with their electricity measurement devices. 

The quantitative methodology is chosen in this study to statistically answer the 

research questions described as follows: 

● What is the customer value associated with electric smart metering 

functions in South Africa? 

● What is the dominant smart metering technologies functions contributing 

to customer value? 



 

24 

 

● Is there is a difference in customer values of smart metering technologies 

by the different customer types?  

3.3 Research design 

Research design entails the  approach that one takes in executing research 

based on data collection and associated tools involved in the process of 

answering the research questions (Miller & Salkind, 2002). Research design is 

composed of two components namely, data collection and analysis. 

In this research a descriptive approach will be undertaken which involves 

developing the research questions before the research is conducted, followed by 

data collection to gather information from respondents to answer the research 

questions developed, outlined in the research method. This method is suitable 

for this study in that it requires one to get an understanding of the perceived value 

of electric smart metering technologies. This study requires the development of 

questions relating to the benefits and costs of smart metering technologies and 

thereafter presenting them to respondents in a manner that it is not leading the 

respondents to specific answers. 

3.4 Population and sample 

3.4.1 Population 

The population for this research is composed of South African household owners 

and tenants responsible for payment and or monitoring and tracking of electricity 

usage in their households. This population is described as 85% of the South 

African households (17.16 million) (World Bank, 2019), translating to a population 

of 14,6 million households.  

3.4.2 Sample and sampling method 



 

25 

 

A sample is considered in cases where it is not possible to collect data from the 

entire population to answer research questions. This limitation arises mainly due 

to the lack of resources and time to collect data from all cases in a population 

(Taherdoost, 2016). Furthermore, sampling can be achieved through two 

techniques. First, probability sampling or random sampling which is described as 

a sampling method where every item in the population has a chance of being 

included in the population (Taherdoost, 2016). Secondly, there is non-probability 

or non-random sampling which is generally used in experimental or trial research 

and does not represent the target population (Ayhan, 2011). 

In this research, a snowball non-probability/non-random sampling method was 

undertaken ꟷ described as a sample method where sample participants are used 

to encourage others to participate in the study (Ayhan, 2011; Taherdoost, 2016). 

The sampling was undertaken by sending structured questionnaire through an 

email to individuals meeting the criteria of participants (South African household 

owners and tenants responsible for electricity payment and or monitoring and 

tracking). To ensure that the desired sample size is reached, respondents were 

requested to forward the questionnaire email to individuals in their network who 

meet the criteria of respondents. 

3.5 Sample size 

Sample size is a very critical component in making inferences about a population 

from a sample in an empirical research study, free from sample errors or biases. 

This raises the need to calculate accurately the minimum required sample in the 

proposed research (Taherdoost, 2017) for it to be representative of the 

population. 

Below is a formula for calculating sample size for categorical data which is fitting 

for the research that is undertaken in this study, equation 2 is for an unlimited 

sample and equation 3 is for a finite sample: 



 

26 

 

𝑛 =
𝑝(100−𝑝)𝑧2

𝐸2                      (2) 

𝑛′ = 𝑛/(1 + 
𝑝(100−𝑝)𝑧2

𝐸2𝑁
)      (3) 

Where n and n’ are the sample size, P is the percentage occurrence of a state or 

condition, E is the percentage maximum error and Z is the value corresponding 

to level of confidence required (Taherdoost, 2017). 

For this research the following parameters are used, the value of P is set to 50% 

as the output of the research is not obtained, where in such cases the selected 

value is accepted. The value of E is the margin of Error set to 5%. A 95% 

confidence interval is selected, and this corresponds to a value of Z as 1.96 and 

the population size used is 14.6 million. This results in a sample size of 385. 

3.6 The research instruments  

The main goal of the research is to determine the customer value associated with 

electric smart meters in South Africa. The following equation, derived from the 

literature, represents Customer Value (Woodside et al., 2008).  

Customer Value = Sum of benefits – Sum of Costs incurred 

In this study the customer value of electric smart meters as considered as the 

main construct with the benefit elements described as the 5 external performance 

objectives (Speed, Flexibility, Cost, Dependability and Quality) and the cost 

incurred being the identified costs of data and electricity to the consumer. Both 

the benefit and cost elements cannot be directly measured through sending a 

survey questionnaire to respondents. Rather these elements can be described 

by other elements. This introduces complexity to the formula to determine 

Customer Value. 

To successfully determine Customer Value, a multivariate statistical technique is 

proposed. The statistical multivariable technique employed here is the Structural 



 

27 

 

Equation Model (SEM), to describe and determine Customer Value. This is 

similar to a study performed to determine the Customer Perceived Value of tablet 

PCs where the both the benefits and cost elements could not be determined and 

SEM is adopted (Amini, Falk, Schmitt, 2014). SEM is described as a powerful 

multivariate technique increasingly found in scientific investigations to test and 

evaluate multivariate causal relationships, with capabilities to test direct and 

indirect effects on pre-assumed causal relationships (Fan, Chen, Shirkey, John, 

Wu, Park & Shao, 2016). In contrast to regression, the structural model 

represents a causal link rather than a mere empirical association (Stuart, Mednick 

& Bockman, 2000). SEM is a combination of two statistical methods, namely 

Confirmatory Factor Analysis (CFA) and Path Analysis (Fan et al., 2016). 

3.6.1 Path Analysis 

A Path is developed to quantify and explain the causal relationship amongst 

multiple variables (Fan et al., 2016). This is used to confirm and quantify the 

relationship between the customer value and the latent variables described by 

the CFA and the measured variables. From the developed Customer Value 

equation, the structural equation below is developed indicating the path analysis. 



 

28 

 

 

Figure 5: Electric Smart Meter Customer Value Structural Equation Model 

path diagram 

From the structural equation, there are latent variables identified which cannot be 

measured but are described by other measured variables. To determine the effect 

of this relationship a Confirmatory Factor Analysis (CFA) is applied. The identified 

latent variables are Cost, Flexibility, Speed, Dependability, Quality and Costs. 

3.6.2 Confirmatory Factor Analysis (CFA) 

Confirmatory Factor Analysis (CFA) is a method used for estimating the latent 

variables based on correlated datasets of the measured variables (Fan et al., 

2016). The latent variables will be determined from the data that will be collected 

through the survey questions, which will contain variables that describe the latent 

variables described by the SEM indicated in Figure 6.  Through the CFA the 

weighting of the benefit and cost element will be determined. The CFA will also 



 

29 

 

ensure accuracy in the development of the SEM by including error co-efficient of 

the description of the latent variables with the measured variables. 

 

Figure 6: Electric Smart Meter Customer Value Structural Equation Model 

CFA 

3.7 Data analysis and interpretation 

In a study to determine the Customer Perception Value of tablet PCs, the 

Structural Equation Model (SEM) was adopted to develop an equation with latent 

variables and measured variables to describe the Customer Perception Value. 

Post developing the model, polar diagrams were utilised providing a graphical 

view of the elements that contribute more to the customer perception value and 

the elements that contribute less to customer value were also represented on the 

polar diagrams (Amini et al., 2014). 

In this research, the same approach will be undertaken, that is, adopting the 

Structural Equation Model to develop a model that represents the relationship 

between Customer Value and the other latent variables (5 performance 

objectives, and data and electricity cost) and the associated measured variables. 

The data analysis process of the SEM will be undertaken through “lavaan”, a 



 

30 

 

statistical software package running on R. Through lavaan both the Path analysis 

and Confirmatory Factor Analysis (CFA) will be determined. R as well as 

Microsoft Excel Statistical tools are used to answer the other research questions 

which require the application of descriptive statistics. Furthermore, polar 

diagrams are used to provide a graphical representation of the different 

categories of customer value answering the research questions.  

3.8 Validity and reliability 

This section describes the validity and reliability of the research instrument. 

3.8.1 Validity 

To ensure the external validity of the research instrument, a literature review 

composed of peer reviewed articles is conducted providing groundwork and 

management theories, used to structure the research questions to appropriately 

answer the problem at hand. To ensure internal validity, which is described as 

the relationship between at least two variables, the SEM is developed covering 

the relationship between the Customer Value, the latent variables and the 

measured variables. 

3.8.2 Reliability 

In a study to determine Customer Perception Value (CPV) of tablet PCs, a 

Cronbach Alpha and the composite reliability (CR) of each of the cost and benefit 

elements was considered to validate the reliability of the relationship. This 

resulted in within range values, providing confidence in the reliability of the 

instrument (Amini et al., 2014). In this study the same approach is undertaken 

which entails determining the Cronbach Alpha and composite reliability of the 

measured variables making up the latent variables that describe the customer 

value variable. 



 

31 

 

3.9 Ethical considerations 

3.9.1 Anonymity of participants 

The identity of all the participants were not requested when they responded to 

the study. The participants responded to questions through an online survey 

which did not record the identity details of the respondents including the IP 

address of the computer or mobile device through which participants responded 

to questions (Scherbaum & Shockley, 2015). 

3.9.2 Permission to conduct the study 

An ethics clearance form was completed with details of the research study to 

provide context to the ethics committee to determine whether or not the research 

violates any ethical restrictions. Data collection for this study commenced only 

once the ethics clearance certificate was granted. 

 

4. PRESENTATION OF RESULTS 

4.1 Introduction 

The purpose of this study is to identify electricity smart metering functions that 

create value for residential customers in South Africa. In this section the results 

obtained from the statistical analysis of the data collected through the online 

survey is presented. The results will be presented in a fashion that follows the 

research question identified, described as follows: 

● What is the customer value associated with electric smart metering 

functions in South Africa? 

● What is the dominant smart metering technologies functions contributing 

to customer value? 



 

32 

 

● Is there is a difference in customer values of smart metering technologies 

by the different customer types?  

The results of the research stem from the research questions obtained from the 

literature review which was broken down into a set of survey questions developed 

to elicit information to address the problem statement. The survey questions were 

composed of the following categories:  

1. Demographic data – Gender and Age (Categorical data) 

2. Values of smart metering functions (Likert scale) 

The data collected was composed of 365 responses, 12 responses were rejected 

due to incorrect completion by the respondents, and this resulted in a sample 

comprising 353 participants for this research. The participants in the survey 

consisted of 44.7% female and 55.3% male, with an age group consisting of 

27.9%, 29.9%, 33,4% and 8.8% for the ages 35-55, 25-34, 18-24 and above 55 

years, respectively.  Figure 7 summarises the distribution of the different 

customer types that responded to the survey. 

Gender: 

 

Age (Years): 

 

Figure 7: Electric smart meter customer value Respondent distribution 

according to gender and age. 



 

33 

 

The responses to the questions were based on a Likert scale described as 

follows: 

1. Strongly Agree = 5 

2. Agree = 4 

3. Neutral = 3 

4. Disagree = 2 

5. Strongly Disagree = 1 

On Table 3 below is the list of questions categorised according to the latent 

variables they describe. 

Table 3: List of Survey Questions describing latent variables for the study. 

Latent Variables Questions 

Speed 

1.1 If it were possible, it will be important for you to know how much 
electricity you are using on daily basis. 

1.2 If it were possible, it will be important for you to know the electricity 
charges applied on your electricity usage on daily basis. 

Quality 

2.1 If it were possible, it will be important for you to get invoices with 
up to date information on your electricity usage any time you request 
them. 

2.2 If it were possible, it will be important for you get bills with accurate 
electricity usage information from the electricity provider. 

2.3 If it were possible, it will be important for you to know your 
contributions to Carbon Emissions, and meeting Climate change 
targets based on electricity usage? 

2.4 It is important for you to know how you are billed for your electricity 
(You want visibility of your tariff charges). 

2.5 If your electricity supplier were to send its employees to your 
house to collect electricity usage readings from your meter, you would 
be comfortable with the reads collected manually (recorded on a 
notebook). 

Flexibility 

3.1 It is important for you to accurately budget for your electricity 
usage. 

3.2 If it were possible, it will be important for you to have the option to 
choose between post-paid (use first and pay after) and prepaid (pay 
first and use after). 



 

34 

 

3.3 If it were possible, it will be important for you to know how other 
customers are using electricity. 

3.4 You would be comfortable if your electricity provider were to send 
its employees to your house to take meter readings every month. 

3.5 Will you change your electricity usage if you know it will prevent 
load shedding in your area and other areas? 

3.6 Will your need to install a Solar PV (photovoltaic) increase if you 
would have knowledge of your excess electricity exported to the grid 
and be compensated for the exported electricity? 

Dependability 

4.1 If it were possible, it will be important for you, for customers 
stealing electricity (illegally tampering with the electricity meter) to be 
disconnected? 

4.2 If it were possible, it will be important for you to know when there 
is and will be electricity outage in your area? 

4.3 If it were possible, it will be important for you to be alerted when 
your electricity credit is low? 

4.4 You are okay with having load shedding as long as it does not 
impact important events in your life. 

Cost 

5.1 You will change how you use electricity if there is chance to save 
on electricity. 

5.2 You will change your daily electricity usage routine, if there is a 
chance to save on electricity costs. 

Costs 

6.1 Your electricity bill is high. 

6.2 If you were to get regular updates on your mobile device on your 
electricity usage and status of electricity usage in your area, you 
would be concerned about the cost of data to access this information. 

In the next sections is the statistical results of the data, categorised as per the 

research questions identified.  

4.2 Research Question 1 

The first research question is identified as follows: 

● What is the customer value associated with electric smart metering 

functions in South Africa? 



 

35 

 

To answer this question statistical analysis of the data collected is undertaken. 

To answer this question a Structural Equation Model described by the data 

collected was developed. The developed model is described by the below 

formula. 

𝐶𝑉 =
1

𝑛
 ∑ 𝑥𝑖𝐵𝐸𝑖

𝑛
𝑖=1 −

1

𝑚
∑ 𝑦𝑗𝐶𝐸𝑗

𝑚
𝑗=1       

Where BE is the benefit element and CE is the Cost Element (Amini et al., 2014). 

4.2.1 Structural Equation Modelling 

The Structural Equation Model is developed to determine the Latent variables 

(Speed, Cost, Quality, Flexibility, Dependability and Costs) described by the 

appreciation and/or dislikes of the different smart metering functions and 

attributes detailed on the survey questions, where participants responded to, 

based on a 5-point Likert scale.  

The developed SEM to describe the latent variables and corresponding 

weightings is described below on Figure 8. This model was generated using 

Lavaan, Structural Equation Modelling tool on R.  

 

Figure 8: Structural Equation Model of the electric smart meter Customer 

Value Construct. 



 

36 

 

Initially the cost element of the customer value model was described by two 

variables, the Data Costs and Electricity Costs. During data analysis, the 

electricity cost element was eliminated as it is not necessarily related to the 

electric smart metering capabilities, and it did not add value in the resulting SEM. 

The electricity cost here was the per Kilowatt hour cost and not necessarily the 

total bill that customers receive from the utility.  

In Table 2 below is the resulting model evaluation which indicates an acceptable 

fit of the developed model for customer value, with the obtained values 

corresponding to the guided values indicated in the parameters column for a 

model fit. 

Table 4: Electric smart meter Customer Value Structural Equation Model 

Evaluation 

Measurement Parameter Value 

Badness of fit 

Root mean square error of approximation 

(RMSEA) < 0.06 (0 perfect fit) 0.059 

Chi-square test 0 

Goodness of fit 

Comparative it Index (CFI) - Close to 0.95 0.924 

Tucker Lewis Index (TLI) >0.90 0.895 

4.2.1.1 Reliability Analysis 

The exercise of developing the Structural Equation Model or the measuring 

instrument requires the analysis of the reliability of the model. To do this the 

Cronbach Alpha and Composite Reliability of the each of the latent variables is 

considered. The Cronbach Alpha will determine if the different constructs 



 

37 

 

describing the latent variables produce similar scores. The Composite Reliability 

will determine the overall reliability of a collection of heterogeneous but similar 

items. Table 5 demonstrates the resulting reliability of the developed SEM. 

Table 5: Evaluated reliability of the developed electric smart meter 

Customer Value SEM. 

Latent Variable Cronbach’s Alpha Composite Reliability 

Speed 0.777 0.777 

Quality 0.669 0.668 

Flexibility 0.541 0.528 

Dependability 0.598 0.626 

Cost 0.825 0.828 

Typically, the Composite Reliability and Cronbach Alpha should be 0.7 and 0.8, 

respectively. The results obtained indicate that Quality, Flexibility and 

Dependability have lower values than the required critical level tests. These 

results are after attempts to improve the reliability of latent variable which 

previously were much worse. With regards to Costs, two questions were used to 

describe it, and this resulted in poor reliability due to one of the questions.  The 

improvement involved elimination of the variables or questions that were set to 

describe the latent variables. The below is the list of questions that were 

eliminated for the different latent variables first listed on Table 3. For costs, only 

one item remained which removed the necessity to perform a reliability and 

considering the item as a latent variable. The reliability therefore considered only 

the 5 performance objectives (Speed, Quality, Flexibility, Dependability and 

Cost). 

 

 



 

38 

 

Table 6: Eliminated questions to improve SEM reliability. 

Latent Variables 
Eliminated questions from the SEM to improve 

reliability 

Quality 

2.2 If it were possible, it will be important for you get bills 
with accurate electricity usage information from the 
electricity provider. 

2.5 If your electricity supplier were to send its employees to 
your house to collect electricity usage readings from your 
meter, you would be comfortable with the reads collected 
manually (recorded on a notebook). 

Flexibility 
3.4 You would be comfortable if your electricity provider 
were to send its employees to your house to take meter 
readings every month. 

Dependability 

4.1 If it were possible, it will be important for you, for 
customers stealing electricity (illegally tampering with the 
electricity meter) to be disconnected? 

4.4 You are okay with having load shedding as long as it 
does not impact important events in your life. 

Costs 6.1 Your electricity bill is high. 

To improve further the reliability of the SEM, the above eliminated questions can 

be restructured to enable responded to give reliable feedback.  

4.2.1.2 Validity Analysis 

The exercise of developing the SEM requires the analysis of the validity of the 

model. To do this the Convergent validity is undertaken which compares the 

factor loadings of the latent variables to the value of 0.7. The factor loading should 

be greater than 0.7 to confirm the validity of the developed model. Table 7 

presents the resulting validity analysis of the developed model. Again here the 

Flexibility factor loadings are below 0.7 compromising on the validity of the model, 

this will require further extension and future research. 

 



 

39 

 

Table 7: Evaluated validity of the developed electric smart meter Customer 

Value SEM. 

Latent Variable Convergent Validity 

Factor loadings (>0.7) 

Speed Yes 

Quality Yes 

Flexibility No (0.47;0.57;0.45;0.53) 

Dependability Yes 

Cost Yes 

Costs Data Yes 

4.2.1.2 Descriptive Statistics 

The resulting descriptive statistics are based on the computed Customer Value 

from the developed SEM. From equation 1 the Customer value is expected to 

range from 5 to – 5 depending on how the smart metering technologies are rated. 

From Table 9 the customer value (- 0.2) is based on the aggregate means of the 

obtained latent variables and the cost variable listed below on Table 8.  

Table 8: Means of the obtained variables (including latent) and cost elements. 

Variable Mean 

Costs (Cost of data) 3.32861 

Speed 3.7238 

Quality 2.99336 

Flexibility 1.896 

Dependability 3.157 

Cost 3.797 

 

 



 

40 

 

Table 9: Descriptive statistics of electric smart meter Customer Value 

Customer Value 

Mean - 0.215238149 

Standard Error 0.070890922 

Median -0.3808 

Mode -1.602 

Standard Deviation 1.331919494 

Sample Variance 1.774009537 

Kurtosis -1.100107553 

Skewness 0.384614904 

Range 4.591333333 

Minimum -2.193333333 

Maximum 2.398 

Sum -75.97906667 

Count 353 

Confidence Level (95.0%) 0.139423035 

4.3 Research Question 2 

The second research question is, what is the dominant smart metering 

technologies functions contributing to customer value? The hypothesis of this 

research question is described below. 

H0: Null Hypothesis – Smart metering functions equally contribute 

to Customer Value 

H1: Alternate Hypothesis – Smart metering functions’ 

contributions to customer value vary.  

To answer this question first the descriptive statistics of the different latent 

variables describing the different smart metering functions is determined together 

with the ANOVA which compares the variances of the different latent variables. 

On Error! Reference source not found. are the results of the hypothesis testing 



 

41 

 

based on ANOVA determining whether the different smart metering functions’ 

(based on the SEM latent variables, Speed, Quality, Flexibility, Dependability and 

Cost) contribution to customer value is the same or it varies. The Null Hypothesis 

suggests that the different functions contribute equally towards the customer 

value, and the Alternate Hypothesis is that there is a difference. A Single Factor 

ANOVA is considered here due to multiple groups (>2) that need to be compared. 

This resulted in a P-Value of 0 < 0.05 which suggests that there is a significant 

difference at 5% confidence. This means there is a difference in the contribution 

of the different smart metering functions towards customer value, so the null 

hypothesis is rejected. 

Table 10: Hypothesis testing of different smart metering functions on Customer 

Value 

SUMMARY     

Groups Count Sum Average Variance 

Speed 353 1314.5 3.723796 0.198659 

Quality 353 1056.6567 2.9933617 0.084343 

Flexibility 353 669.188 1.8957167 0.057698 

Dependability 353 1114.56 3.1573938 0.088795 

Cost 353 1340.2 3.7966006 0.278943 

 

ANOVA       

Source of 

Variation SS df MS F P-value F crit 

Between Groups 825.4706 4 206.368 1456.496 0 

2.3769

8 

Within Groups 249.3704 1760 0.14169    

       

Total 1074.841 1764         



 

42 

 

Below is the means of the 5 operations performance objectives determined based 

on the developed SEM. This indicates the dominant element which is the Cost 

followed by Speed. 

Benefit element Mean 

Speed 3.7238 

Quality 2.99336 

Flexibility 1.896 

Dependability 3.157 

Cost 3.797 

4.4 Research Question 3 

The third research question is to determine if there is a difference in customer 

values of smart metering technologies by the different customer types?  

H0: Null Hypothesis – There is no difference in the customer value 

associated with electric smart meters by different customer types. 

H1: Alternate Hypothesis – There is a difference in the customer 

value associated with electric smart meters by different customer 

types. 

The hypothesis considers two categories of customer types, first based on 

gender, and then on age. To answer the question that considers gender, a 

Hypothesis testing using the t-Test was undertaken since it is a comparison of 

two variables. To answer the question that considers age, a single factor ANOVA 

was used to do the hypothesis testing. Both hypotheses testing were done on the 

obtained Customer Value which resulted from the model developed by the SEM.  

On Table 11 are the results of the hypothesis testing based on the t-Test to 

determine whether there is a difference in customer value of smart metering 

technologies between male and female. The Null Hypothesis suggests that there 

is no difference, and the Alternate Hypothesis is that there is a difference. A two-

tail test was applied resulting in a P-Value of 0.61 > 0.05 which suggests that 



 

43 

 

there is no significant difference at 5% confidence. This means there is no 

difference in customer value between male and female, so the null hypothesis is 

accepted. 

Table 11: Customer Value Hypothesis testing based on gender 

t-Test: Two-Sample Assuming Equal Variances 

  Male Female 

Mean -0.182036 -0.254833126 

Variance 1.6982266 1.87266262 

Observations 192 161 

Pooled Variance 1.7777416  

Hypothesised Mean Difference 0  

df 351  

t Stat 0.510923  

P(T<=t) one-tail 0.3048631  

t Critical one-tail 1.6492064  

P(T<=t) two-tail 0.6097262  

t Critical two-tail 1.9667456   

Table 12 presents the results of the hypothesis testing based on single factor 

ANOVA to determine whether there is a difference in customer value of smart 

metering technologies between different age groups (18-24, 25-34, 35-55 and  

above 55 years). The Null Hypothesis suggests that there is no difference, and 

the Alternate Hypothesis is that there is a difference. A Sigle Factor ANOVA was 

applied here due to multiple groups (>2) that had to be compared. This resulted 

in a P-Value of 0.537 > 0.05 which suggests that there is no significant difference 

at 5% confidence. This means there is no difference in customer value between 

the different age groups, so the null hypothesis is accepted. 



 

44 

 

Table 12: Customer Value Hypothesis testing based on age group 

Anova: Single Factor    

SUMMARY     

Groups Count Sum Average Variance 

Age 18-24 119 -40.4889333 -0.3402431 1.667469815 

Age 25-34 105 -8.6592 -0.0824686 1.684776935 

Age 35-55 99 -18.7178667 -0.1890694 1.947471712 

Above 55 30 -8.11306667 -0.2704356 1.991422526 

ANOVA       

Source of 

Variation SS df MS F P-value F crit 

Between Groups 3.8696 3 1.2898789 0.7254 0.53738 2.63049 

Within Groups 620.58 349 1.7781711    

       

Total 624.45 352         

4.5 Summary of Results 

The statistical results of the 3 research questions identified to understand the 

customer value associated with electric smart meters are presented. First a 

model to describe the customer value was developed using the Structural 

Equation Model (SEM), this resulted in an acceptable fit. The obtained model 

lacks in reliability due to the latent variables “Flexibility, Quality and 

Dependability”. The validity of the model is also compromised by the “Flexibility 

latent variable”. Due to this limitation a further enhancement is recommended on 

the model in future research.  

With these limitations the first research question resulted in customer value of 

smart electric meters obtained as a mean of -0.21 on a scale of 5 to -5, where 5 

indicates maximum positive value and -5 being maximum cost.   



 

45 

 

The second research question resulted in the rejection of the Null Hypothesis , 

which suggested that the different latent variables describing different smart 

metering functions contribute equally to customer value. This was tested using a 

Single Factor ANOVA which resulted in a P-value of 0 (<0.05) suggesting that 

there is a significant difference between the contributions of the latent variables 

towards the customer value at 5% confidence. Furthermore its was identified 

based on the means of 5 performance objectives deduced from the SEM that 

Speed and Cost are the dominant functions contributing to customer value 

The third question resulted in the Null Hypothesis being accepted. The null 

hypothesis stated that there is no difference in customer value for different 

customer types. The test first considered the aggregate customer value for 

different customer types based on gender, and the t-Test was performed resulting 

in a P-Value of 0.61 > 0.05 which suggests that there is no significant difference 

at 5% confidence between male and female customer groups. A second test was 

performed considering different customer types grouped according to age (18-

24, 25-34, 35-55 and above 55 years). A Single Factor ANOVA resulted in a P-

Value of 0.537 > 0.05 which suggests that there is no significant difference at 5% 

confidence between the different age groups. 

5. DISCUSSION OF RESULTS 

5.1 Introduction 

The aim of this section is to provide a discussion of the results presented in 

Chapter 4 and demonstrate how these results provide answers to the research 

questions presented. The discussion of these results also undergoes a 

comparison with literature to identify any similarities and deviations. 

The results obtained from the survey responses with an almost equal distribution 

between male and female, the age group distribution is almost equal except for 



 

46 

 

the older generation (above 55 years) accounting for less than 10% and the other 

age groups (18-24, 25-34 and 35-55) accounting for approximately one third 

each. The results were based on a sample size of 353. 

5.2 Results 

5.2.1 Research Question 1 

The first research question identified for the study was to determine the customer 

value associated with electric smart metering functions in South Africa. This is 

the main question of the study which seeks to determine whether the perceived 

costs of a smart meter outweigh the benefits as described by the formula below. 

Customer Value = Sum of benefits – Sum of Costs incurred 

The benefits of having a smart meter were determined by posing numerous 

questions to the respondents to rate their like or dislike of the different features 

of smart meters based on five external operations performance objectives 

(Setijono & Sandberg, 2005). Due to the inability to collect data from respondents 

directly for the five performance objectives, the Structural Equation Model (SEM) 

was employed with the five operations performance objectives as the latent 

variables. The developed SEM resulted in a fitting model ꟷ validation and 

reliability tests resulted in positive results with a compromise due to the Flexibility, 

Dependability and Quality latent variables. Even with this compromise, the 

resulting model is sufficient to provide an indication of the perceived customer 

value of electric smart meters in South Africa. 

On a scale of 5 to -5, the obtained aggregate perceived customer value is slightly 

below average at -0.2, this indicates that South African electricity residential 

customers do not necessarily have a complete dislike of smart meters. This 

suggests that communication of smart meters to South African residents would 

not result in mass rejection or scepticism as observed in other similar economies 



 

47 

 

such as Indonesia and USA (Chou & Gusti Ayu Novi Yutami, 2014; Lineweber, 

2011) if the available features are appropriately communicated and made 

available to customers. Furthermore, the resulting Customer Value is as a result 

of a combination of five benefits, therefore an average benefit also suggests that 

there are some smart metering features which are not important specifically for 

the South African customers and yet others are important for them.  

In a study by (Muchenje & Botha, 2021) on determining consumer centric factors 

for the implementation of smart meters in South Africa, it was observed that not 

all smart metering features presented to customers were appealing, but among 

the features presented elements such as cost, ease of use, social norms, attitude, 

trust in technology played a significant role. This study indicated that South 

African consumers showed differing interests on smart metering technologies 

compared to the European and American economies (Muchenje & Botha, 2021). 

In the next section is a research question which seeks to understand the smart 

metering features that are particularly important to South Africa customers. 

Furthermore, the resulting customer value is also influenced by the cost of having 

a smart meter and accessing the features of the smart meter. One of the main 

costs identified is the need for customers to have mobile data or internet, to 

access the features, such as viewing historical consumption data, receiving alerts 

and notifications. Respondents rated the cost of data, on average, between 

neutral (3) and agree (4) on a scale of 1 to 5 which indicates that data costs are 

a concern to South African customers. The response on data costs is in line with 

the concerns reported in South Africa, that despite the reduction of tariffs which 

reduced the cost of 1GB from R149 to R99 by two of South Africa’s major 

telecommunication companies, the cost of data is still found to be very high 

(Chinembiri, 2020). 

5.2.2 Research Question 2 



 

48 

 

The second research question for this study was to determine the dominant smart 

metering functions that influence customer value. This is mainly on the benefits 

of the smart metering function, as previously described the benefits of electric 

smart metering functions are described by the five external operations 

performance objectives. Since this cannot be directly measured or data on it 

collected through a survey, it is computed by the Structural Equation Model 

described in the first research question. In this case the idea is to understand 

which of the latent variable/s has a dominant contribution towards the customer 

value.   

The statistical analysis utilising the Single Factor ANOVA indicates that the latent 

variables do not equally contribute towards Customer Value. This is 

demonstrated by a P-Value of 0 < 0.05 indicating that there is a significant 

difference between the contributions of the latent variables towards the customer 

value at 5% confidence. This result is in line with the finding by (Muchenje & 

Botha, 2021) which indicates that different factors affect the value of smart meters 

differently.  

 

Figure 9: Latent Variables Contributions to Customer Value. 



 

49 

 

Figure 9 is a polar diagram providing a graphical representation of the 

contribution by each of the latent variables considered in this study. This result is 

based on the means of the different latent variables obtained from the SEM 

computed through the data collected. From the polar diagram it is apparent that 

the cost reduction feature of electric smart meters which allows customers to save 

money is an important factor to South African customers. This result suggests 

that the smart metering features that will allow customers to reduce costs is much 

needed. This cost reduction feature is about the ability for smart meters to provide 

information that will trigger them to change their behaviour with regard to 

electricity usage. This is in line with findings in Scotland where the older 

generation exhibited a higher willingness to pay for smart meters due to the 

meters’ ability to enable the customer to reduce costs in their electricity usage 

(Barnicoat & Danson, 2015). 

The cost feature is followed by the speed feature which is ranked high as well 

thus indicating the need by South African customers to utilise the features of 

smart meters that allow them to receive service speedily.  These features include 

the ability for customers to receive their electricity consumption in a near real-

time. Speed also becomes critical to meet the cost reduction need whereby, 

receiving data in real-time allows the consumer to adjust electricity usage to 

minimise cost ꟷ an opportunity which without the speed factor would be missed. 

The speed feature is followed by the dependability feature in the rankings and 

requirements of the South African consumers. South African consumers rely on 

the ability of the smart meters to provide information that will enable consumers 

to adjust their consumption, change their behaviour to prevent unwanted 

outcomes such as loadshedding or using too much electricity. So, this nudge 

capability introduced by the smart metering capabilities creates a dependency, 

meaning incorrect information will result in customers loosing on the opportunity 

to save on electricity, the opportunity to identify and prevent things going out of 

hand such as load shedding. Furthermore the dependability opportunity arises 

from the ability of electric smart metering technologies to provide customers with 



 

50 

 

outage information, to help customers to plan daily activities and avoid 

interruptions arising from power outages. The other opportunity realised from the 

dependability feature is the ability to provide billing information in a timely manner 

and this could prevent the accumulation of debt. 

From these results, the Cost, Speed and Dependability features, are the 

dominant contributors towards a positive Customer Value. The remaining two 

(Quality and Flexibility), though they are critical for a smart metering operation, 

rank lower on the value system of South African customers. Table 13 summarises 

the smart metering features that are critical and dominant in the contribution of 

electric smart meter customer value. This is in line with what is stated by Sovacool 

et al (2017), with an exception of the ability to switch between prepaid and post-

paid which is covered by the Flexibility feature of electric smart meters. 

Table 13: Dominant smart electric metering functions on Customer Value. 

Cost Speed Dependability 

Ability to get alerts with 

information to change electricity 

usage behaviour to save 

electricity. 

Ability to know how 

much electricity a 

customer is using on 

daily/real-time basis. 

Ability to know when 

there is and will be 

electricity outage in 

customer's area. 

 

Ability to know the 

electricity charges 

applied on daily basis. 

Ability to be alerted 

when electricity credit is 

low. 

  

Ability to participate in 

minimising or prevent 

load shedding 

(minimising electricity 

usage) 

It is also worth noting that South Africa customers considered, Flexibility as a 

factor that is not important to them as an electric smart meter. In the case of 

quality, respondents indicated a neutral response in general. Table 14 shows the 

Flexibility and Quality features that respondents were presented with. 



 

51 

 

Table 14: Non-critical smart electric metering functions on Customer Value. 

Flexibility Quality 

Ability to accurately budget for electricity 

usage 

Ability to get invoices with up-to-date 

information on electricity usage any time it 

is requested 

Ability to choose between post-paid (use 

first and pay after) and prepaid (pay first 

and use after) 

Ability to get bills with accurate electricity 

usage information from the electricity 

provider 

Ability to know how other customers are 

using electricity 

Ability to know and see how customers are 

billed for their electricity (Visibility of tariff 

charges to customers) 

Ability to know how much excess electricity 

is exported to the grid for customers with 

renewable energy sources (e.g. PV)  

 

 

5.2.3 Research Question 3 

The third research question is to determine if there is a difference in customer 

value of electric smart meters by different customer types where age and gender 

are the differentiators. From the statistical results it was established that there is 

no difference in the perceived customer value of electric smart meters by the 

different customer groups that are differentiated by age and gender as the groups 

have similar customer value for electric smart meters.  

This result differs with the findings by (Barnicoat & Danson, 2015) which indicated 

that the older generation in Scotland in the  low income band had different 

customer value of smart meters with a strong willingness to pay for smart meters 

due to the ability for smart meters to provide feedback on consumption triggering 

change is consumption behaviour.  

As a result of the award winning electrification programme, the Integrated 

National Electrification Programme (INEP), 87% of South African households 

have been provided with access to electricity, however affordability still remains 

an issue due to the high levels of poverty with close to 50% of South African 



 

52 

 

households using electricity with a combination of other energy sources such 

firewood, paraffin and gas for cooking (Simisha, Reddy, Wolpe & Radebe , 2017).  

Based on this finding, it is apparent that South African customers are mainly on 

the low-income side. Because majority of South African households are on the 

low income side, their response to opportunities brought by technologies such as 

electric smart meters are rather influenced by this low income status and this is 

consistent across all customer groups (male and female; young, mid and old 

age). This also gives rise to opportunities for utilities as it will not encounter 

challenges of varying customer needs that need to be taken care of differently. 

One solution can be adopted based on the findings for research question 2 and 

meet the needs of residential South African customers. 

5.3 Summary of Results Discussion 

From the results obtained, a slightly below average customer value is observed. 

Initially, this may be interpreted as a similar view to  (Ball, 2019) that states that 

smart meters do not add value to customers but only to companies and utilities. 

Based on the analysis performed, it has been established that customers find 

value in the smart meter functions described by the latent variables Speed, Cost 

and Dependability. Customers are neutral with regard to the quality component 

that smart metering functions brings on the visibility of electricity consumption. 

Lastly customers are neutral on the flexibility component that smart metering 

functions bring. 

The other component that negatively influences the customer value for electric 

smart meters is the cost associated with internet connectivity which is required to 

access the electric smart meters functions owing to the high cost of data levied  

by the South African telecommunication providers (Chinembiri, 2020). Based on 

this result for South Africa, in particular, cost reduction is the main driver in the 

value that is associated to the electric smart meters, and this is visible in the value 

that South African customers put on the latent variables that have an influence 



 

53 

 

on cost, such as Speed, Dependability and Cost itself. This response is in line 

with what is observed on the smart metering deployments that have been taking 

place in South Africa where it is stated that smart meters are gaining popularity 

(Nhede, 2018; Zyl, 2018). Based on this, it can be observed that there indeed 

value in the smart metering technologies to South African customers, what 

remains a challenge is the cost of the supporting technologies that enable 

customers to fully access the electric smart metering benefits. 

Further analysis also indicated that there is no difference in the perceived value 

of smart electric meters by the different customer groups (based on gender and 

age). As opposed to other countries such as Scotland where the older low-income 

generation found the cost saving feature as a point of trigger for willingness to 

pay for smart meters (Barnicoat & Danson, 2015).  

6. CONCLUSION AND RECOMMENDATIONS 

6.1 Conclusion 

The purpose of this study was to determine the customer value of electricity smart 

meters in South Africa for residential customers. The benefit elements of the 

customer value were described by five external operations’ performance 

objectives (Speed, Quality, Dependability, Quality and Flexibility) with the cost 

element described as the cost of accessing smart metering functions by the 

consumer. Due to the inability to directly measure the five operations’ 

performance objectives through survey, a Structural Equation Model (SEM) 

analysis was adopted.  

Employing SEM involved setting the five operations’ performance objectives as 

latent variables, described as a set of known variables collected through survey 

questions. The survey resulted in 365 responses; 12 responses were rejected 

due to incorrect completion by the respondents. A total 353 responses were used 



 

54 

 

for this research. This was sufficient for the population size of South African 

residential customers. The resulting SEM from the sample data resulted in a 

satisfactory model fit. For reliability, the results obtained indicated that Quality, 

Flexibility and Dependability have lower values than the critical level tests. The 

reliability of the model required was compromised due to the Quality, Flexibility 

and Dependability elements which indicate the need for further extension to 

improve the reliability of the model. The validity of the model also resulted in 

acceptable levels. 

Based on the developed model, data collected resulted in a mean customer value 

that is slightly below average. This result, however, does not necessarily suggest 

that South African customers do not value electric smart meters, rather it was 

observed that the customers are sensitive to the internet costs associated with 

accessing data on their smart meters to realise its benefits. This sensitivity is due 

to high telecommunications costs in South Africa.  

Further analysis of the components of customer value, addressed by the second 

research question indicated that the contribution of the different latent variables 

(Speed, Quality, Cost, Flexibility and Dependability) differs. It was established 

that Speed, Cost and Dependability rank higher as dominant contributors to 

customer value. Further analysis to address the third research question indicated 

that there is no difference in the customer value derived by different customer 

types with regard to gender or age differences. 

6.2 Recommendations 

Based on the analysis and results of the study, it is observed that even though 

the obtained customer value is slightly below average, electric smart meters are 

valuable to South African customers. This was observed in the analysis of the 

different benefit elements of the customer value, where elements such Speed, 

Cost and Dependability were rated high by customers indicating their need for 

those features. It was also observed that customer value was compromised by 



 

55 

 

the related internet costs for accessing the electric smart meter functions by 

customers.  

Based on these findings it is recommended for utilities, municipalities and 

organisations planning to introduce or deploy electric smart metering 

infrastructure to focus on implementing smart metering functions relating to the 

Speed, Cost and Dependability elements. Furthermore, effective communication 

and guidelines on how to access and utilise these functions should be developed 

and shared with the customers. The results obtained from the three research 

questions indicated that there is no difference in the customer value within 

different customer groups. This factor eliminates the need to customise the 

service to meet varying customer needs. Utilities are advised to take advantage 

of this factor.  

Further recommendations regarding the improvement or validation of the results 

should be undertaken, especially the enhancement of the research questions 

describing the latent variables Flexibility and Dependability and Quality to obtain 

optimal reliability results. Additionally further cost elements contributing to the 

customer value can be investigated to ensure that it model does not only rely on 

one variable being the data costs, doing this can reveal more insight in the 

customer value when more cost element are put forth to the respondents, 

REFERENCES 

Albani, A., Domigall, Y., & Winter, R. (2017). Implications of customer value 

perceptions for the design of electricity efficiency services in times of smart 

metering. Information Systems and E-Business Management, 15(4), 825–

844. https://doi.org/10.1007/s10257-016-0332-9 



 

56 

 

Amini, P., Falk, B., & Schmitt, R. (2014). Quantitative Analysis of the Consumer 

Perceived Value Deviation. Procedia CIRP, 21, 391–396. 

https://doi.org/10.1016/j.procir.2014.02.059 

Ayhan, H. Ö. (2011). Non-probability Sampling Survey Methods. In M. Lovric 

(Ed.), International Encyclopedia of Statistical Science (pp. 979–982). 

Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-642-04898-

2_41 

Ball, J. (2019). Companies, not consumers, are benefiting from smart meters, say 

engineers. NewStatesman. 

https://www.newstatesman.com/spotlight/energy/2019/06/companies-

not-consumers-are-benefiting-smart-meters-say-engineers 

Barnicoat, G., & Danson, M. (2015). The ageing population and smart metering: 

A field study of householders’ attitudes and behaviours towards energy 

use in Scotland. Energy Research & Social Science, 9, 107–115. 

https://doi.org/10.1016/j.erss.2015.08.020 

Chawla, Y., Kowalska-Pyzalska, A., & Skowrońska-Szmer, A. (2020). 

Perspectives of smart meters’ roll-out in India: An empirical analysis of 

consumers’ awareness and preferences. Energy Policy, 146, 111798. 

https://doi.org/10.1016/j.enpol.2020.111798 

Chinembiri, T. (2020). Mobile-Data-Pricing-Policy-Brief 2020 (South Africa). 

Research ICT Africa. https://researchictafrica.net/wp/wp-



 

57 

 

content/uploads/2020/06/Tapiwa-Chinembiri-Mobile-Data-Pricing-Policy-

Brief2-2020-FINAL.pdf 

Chou, J.-S., & Gusti Ayu Novi Yutami, I. (2014). Smart meter adoption and 

deployment strategy for residential buildings in Indonesia. Applied Energy, 

128, 336–349. https://doi.org/10.1016/j.apenergy.2014.04.083 

Echeverri, R., Echeverry, D., & Lozano, C. A. (2017). Metodología para el análisis 

de costos y beneficios de proyectos Smart Grid y su aplicación en un piloto 

de medición inteligente de energía (AMI). 13. 

Fan, Y., Chen, J., Shirkey, G., John, R., Wu, S. R., Park, H., & Shao, C. (2016). 

Applications of structural equation modeling (SEM) in ecological studies: 

An updated review. Ecological Processes, 5(1), 19. 

https://doi.org/10.1186/s13717-016-0063-3 

Gerpott, T. J., & Paukert, M. (2013). Determinants of willingness to pay for smart 

meters: An empirical analysis of household customers in Germany. 

Energy Policy, 61, 483–495. https://doi.org/10.1016/j.enpol.2013.06.012 

Gill, O., & Judith, W. (2008). The consumer implications of smart meters—Report 

for NCC by Sustainability First.pdf. National Consumer Council. 

Jehran Naidoo. (2021). Tired of Eskom? Here is what the solar alternatives will 

cost. IOL. https://www.iol.co.za/business-report/energy/tired-of-eskom-

here-is-what-the-solar-alternatives-will-cost-f650a39d-0072-5118-9161-

961beb9e1c7c 



 

58 

 

Kaufmann, S., Künzel, K., & Loock, M. (2013). Customer value of smart metering: 

Explorative evidence from a choice-based conjoint study in Switzerland. 

Energy Policy, 53, 229–239. https://doi.org/10.1016/j.enpol.2012.10.072 

Lineweber, D. C. (2011). Understanding Residential Customer Support for – and 

Opposition to – Smart Grid Investments. The Electricity Journal, 24(8), 92–

100. https://doi.org/10.1016/j.tej.2011.09.005 

McFarlane, D. A. (2013). The Strategic Importance of Customer Value. Atlantic 

Marketing Journal, 2(1), 62–75. 

Miller, D., & Salkind, N. (2002). Handbook of Research Design & Social 

Measurement. SAGE Publications, Inc. 

https://doi.org/10.4135/9781412984386 

Mogles, N., Walker, I., Ramallo-González, A. P., Lee, J., Natarajan, S., Padget, 

J., Gabe-Thomas, E., Lovett, T., Ren, G., Hyniewska, S., O’Neill, E., 

Hourizi, R., & Coley, D. (2017). How smart do smart meters need to be? 

Building and Environment, 125, 439–450. 

https://doi.org/10.1016/j.buildenv.2017.09.008 

Muchenje, T., & Botha, R. (2021). Consumer-centric factors for the 

implementation of smart meters in South Africa. South African Computer 

Journal, 33(2). https://doi.org/10.18489/sacj.v33i2.909 

Nhede, N. (2018, August 7). Smart electricity meters: Energy conservation 

features to aid in gaining popularity. Smart Energy International. 



 

59 

 

https://www.smart-energy.com/industry-sectors/smart-meters/smart-

electricity-meters-grandview-research-south-africa/ 

Scherbaum, C. A., & Shockley, K. M. (2015). Analysing Quantitative Data for 

Business and Management Students. SAGE Publications, Inc. 

https://doi.org/10.4135/9781529716719 

SEA. (2015). Smart Metering: Overview and Considerations for South African 

Municipalities. Supporting Sub-Saharan Africa’s Municipalities with 

Sustainable Energy Transitions. 

https://africancityenergy.org/uploads/resource_127.pdf 

Setijono, D., & Sandberg, D. (2005). Customer Value Strategy and Customer 

Value Management for Competitive Wood-Flooring Manufacturers 

(Literature Review No. 17; p. 64). School of Technology and Design, Växjö 

University. 

Simisha, P.-E., Reddy, Y., Wolpe, P., & Radebe, H. (2017). Energy Poverty and 

Gender in Urban South Africa. Sustainable Energy Africa. 

https://www.cityenergy.org.za/uploads/resource_432.pdf 

Slack, N., Alistair, B.-J., Robert, J., Hemmanth, S., & Khomotso, P. (2017). 

Operations Management: Global and Southern African Perspectives (3rd 

ed.). Pearson. 



 

60 

 

Sovacool, B. K., Kivimaa, P., Hielscher, S., & Jenkins, K. (2017). Vulnerability 

and resistance in the United Kingdom’s smart meter transition. Energy 

Policy, 109, 767–781. https://doi.org/10.1016/j.enpol.2017.07.037 

Stuart, K. R., Mednick, M., & Bockman, J. (2000). Structural Equation Model of 

Customer Satisfaction for the New York City Subway System. 

Transportation Research Record: Journal of the Transportation Research 

Board, 1735(1), 133–137. https://doi.org/10.3141/1735-16 

Taherdoost, H. (2016). Sampling Methods in Research Methodology; How to 

Choose a Sampling Technique for Research. SSRN Electronic Journal. 

https://doi.org/10.2139/ssrn.3205035 

Taherdoost, H. (2017). Determining Sample Size; How to Calculate Survey 

Sample Size. 2, 4. 

US Department of Energy. (2016). Advanced Metering Infrastructure and 

customer systems. Results from the Smart Grid Investment Grant 

Program. 

https://www.energy.gov/sites/prod/files/2016/12/f34/AMI%20Summary%2

0Report_09-26-16.pdf 

Völker, B., Reinhardt, A., Faustine, A., & Pereira, L. (2021a). Watt’s up at Home? 

Smart Meter Data Analytics from a Consumer-Centric Perspective. 

Energies, 14(3), 719. https://doi.org/10.3390/en14030719 



 

61 

 

Völker, B., Reinhardt, A., Faustine, A., & Pereira, L. (2021b). Watt’s up at Home? 

Smart Meter Data Analytics from a Consumer-Centric Perspective. 

Energies, 14(3), 719. 

Woodside, A. G., Golfetto, F., & Gibbert, M. (2008). Customer value: Theory, 

research, and practice. In Advances in Business Marketing and 

Purchasing (Vol. 14, pp. 3–25). Emerald (MCB UP ). 

https://doi.org/10.1016/S1069-0964(08)14001-7 

World Bank. (2019). Access to electricity (% of population)—South Africa | Data. 

The World Bank. 

https://data.worldbank.org/indicator/EG.ELC.ACCS.ZS?locations=ZA 

Zyl, N. P. (2018, July 26). The rise of smart meters in South Africa. ESI-

Africa.Com. https://www.esi-africa.com/industry-sectors/metering/the-

rise-of-smart-meters-in-south-africa/