DEG R E E P RO JE C T

An overview of the cloud migration process:

a case study of a migration from legacy application
to cloud based application

Denise Hirsch

Bachelor programme in Digital Service Innovation

2023

Luleå University of Technology

Department of Computer Science, Electrical and Space Engineering
[This page intentionally left blank]
Abstract
Switching from traditional information technology to cloud based technology is becoming
more common for organizations looking to obtain benefits such as scalability, pay-per-use,
and increased availability. A cloud migration is a project of high-level complexity, and having
a methodological step-by-step approach can assist an organization through the process.
Although there is a substantial amount of research on cloud migration, there is a lack of
step-by-step descriptions to guide organizations looking to switch to cloud technology.

This is a case study evaluation of a cloud migration, where the cloud migration process is
explored and compared to established frameworks for cloud migration. The case study
organization is a large Swedish retail company that performed a cloud migration between
2020-2022, when switching from a legacy point-of-sale system to a cloud based application
for commerce. Data collection consisted of a document review, an archival data review, and a
survey that was sent out to end users of the cloud application. These three methods were
selected to obtain triangulation of data and to provide an accurate description of the cloud
migration process of the case study organization.

The results of the study demonstrate that the evaluated cloud migration process had many
similarities to the frameworks for cloud migration, although with differences in order of
execution and emphasis on activities. Challenges experienced during the cloud migration
process were related to interoperability, project complexity, and having a third-party service
provider. By contrast, success factors of the cloud migration were proper planning for support
flow and deployment, having substantial user support, and sufficient training for end users.

This study provides a description of an executed cloud migration with an overview of the
process from start to finish, along with insights into challenges and lessons learned regarding
project execution. This could be of value for organizations looking to migrate from legacy
applications to cloud based applications.
Key terminologies

Central Processing Unit (CPU) - The unit of a device that does the main processing and tells
the other components what to do (Microsoft Support, 2023)

Cloud Service Provider (CSP) - A provider of cloud services (AWS, 2023; Microsoft Learn,
2023)

Cloud unit, or Cloud Service Unit (CSU) - Another term used for describing a unit in an
availability zone, which is an area where the data centers, network connection and servers
used for cloud computing are localized (Microsoft Learn, 2023).

Customer Relationship Management (CRM) - data-driven software tools for management

of an organization’s customer relationships (Microsoft, 2023).

Digital payment platform (DPP) - term used, in this case, as a description for a payment
solution that allows for digital payment options.

Distributed Denial of Service (DDoS) - An attack where users from multiple instances flood
the traffic of a single instance with the aim to degrade performance of a service (Microsoft
Learn, 2023)

Enterprise Resource Planning (ERP) system - a system that enables managing and
facilitating all business processes and operations of an organization (Microsoft, 2023).

Identity and Access Management (IAM) - Management of users and entities to make sure
they have the right level of access to a system (Cisco, 2023).

Infrastructure-as-a-Service (IaaS) - A cloud service where the basic structure for cloud IT
infrastructure in terms of hardware and software, such as virtual machines, storage and
databases is provided (AWS, 2023).

Legacy application - A software application for business operations that have been used for a
long time in an organization (Gholami et al., 2016).

Platform-as-a-Service (PaaS) - A cloud service where the infrastructure, middleware,

operating systems, and development tools are provided to the consumer, which enables the
development and deployment in the cloud environment (AWS, 2023; Microsoft Learn, 2023).

Point-of-sale (POS) system - A system that enables purchases for sales of goods (Broberg et
a., 1976).
Service Level Agreement (SLA) - An established contract between a service provider and a
customer, that describes factors such as services, guarantees, responsibilities etc. (CSA,
2023).

Software-as-a-Service (SaaS) - An application offered as a service that is running on cloud

infrastructure (CSA, 2023).
Table of contents
1 Introduction................................................................................................................ 1
1.1 Aim................................................................................................................................. 1
2 Theory..........................................................................................................................3
2.1 What is cloud computing?...............................................................................................3
2.1.1 Cloud service models............................................................................................. 3
2.2 Challenges and recommendations for cloud migration projects.....................................4
2.2.1 Security and privacy...............................................................................................5
2.2.2 Availability.............................................................................................................6
2.2.3 Interoperability.......................................................................................................6
2.2.4 Cloud expertise and project planning.....................................................................7
2.2.5 Responsibility and control..................................................................................... 7
2.2.6 Cloud acceptance................................................................................................... 8
2.3 Frameworks for cloud migration.....................................................................................8
3 Method.......................................................................................................................12
3.1 Case introduction.......................................................................................................... 12
3.2 Source criticism.............................................................................................................12
3.3 Data collection.............................................................................................................. 13
3.3.1 Document review and archival data review.........................................................14
3.3.2 Survey.................................................................................................................. 14
3.4 Data analysis................................................................................................................. 15
3.5 Ethical considerations................................................................................................... 17
3.6 Method discussion........................................................................................................ 17
4 Results........................................................................................................................20
4.1 Document review and archival data review..................................................................20
4.1.1 Current situation (as is)........................................................................................ 20
4.1.2 Providers analysis................................................................................................ 21
4.1.3 Requirements specification.................................................................................. 21
4.1.4 Assessment of migration......................................................................................22
4.1.5 Migration phase....................................................................................................24
4.1.6 Maintenance......................................................................................................... 25
4.1.7 Challenges............................................................................................................26
4.2 Survey........................................................................................................................... 27
4.2.1 Introduction and preparation................................................................................28
4.2.2 Support................................................................................................................. 30
4.2.3 User Experience................................................................................................... 32
4.2.4 Information and communication channels...........................................................34
5 Analysis & Discussion.............................................................................................. 36
5.1 Triangulation of data..................................................................................................... 36
 5.2 R1 How did the cloud migration process compare to the process described in
established cloud migration frameworks?...........................................................................36
5.3 R2 What challenges did the case study organization encounter during the cloud
migration process?.............................................................................................................. 40
5.4 R3 What success factors and learnings can be extracted from the evaluated cloud
migration process?.............................................................................................................. 42
6 Conclusion.................................................................................................................44
6.1 Further studies...............................................................................................................45
7 Ethical reflection.......................................................................................................46
1 Introduction
Cloud computing has become a key technology for IT infrastructure and applications across
different businesses and industries (Kankia et al., 2022). Many organizations are making a
strategic shift from traditional information technology to cloud technology because of
advantages such as higher flexibility, efficiency, security and availability of the applications
(Tatić et al., 2020). Another motivational factor is cost reduction, since cloud providers can
offer low monthly fees (due to economies of scale), billing for solely the amount of data being
used, and scalability of workloads (AWS, 2023; Microsoft Learn, 2023). However, the
migration to cloud is a highly complex project requiring a plan with clear goals and strategies,
strong leadership, and competence in order for project success (Tatić et al., 2020).
Additionally, cloud migrations often involve combining internal and external processes, and
integrating cloud services with the existing architecture (hardware and software), which adds
to the complexity. On behalf of these intricate factors, Tatić et al. (2020) emphasize on the
importance of having a symbiotic view of business processes, technology, and people to
ensure that the cloud technology in the end is accepted by the users.

There has been a lot of research describing the benefits and challenges of cloud migrations, as
well as key motivators for migrating to the cloud (Kankia et al., 2022; Moscoso-Zea et al.,
2018). However, less research has been conducted to describe the cloud migration process in
detail, and the existing research is predominantly theoretical and not very hands-on (Kankia et
al., 2022; Paredes-Gualtor, 2017). Gholami et al. (2016) argue that there is a lack of depiction
on how to carry out a cloud migration, and that a methodological, stepwise approach can
facilitate the process (Gholami et al., 2016). Moscoso-Zea et al. (2018) further emphasize the
need for more step-by-step descriptions of the cloud migration process to support
organizations before and during cloud migrations. The objective for this study is to contribute
with insights from a performed cloud migration project, along with a stepwise description of
the cloud migration process, from start to finish.

1.1 Aim
This is a case study evaluation of a migration from a legacy point-of-sale (POS) system to a
cloud based POS system. The aim is to explore a cloud migration project by evaluating the

1
cloud migration process stepwise from an organizational perspective, i.e. the perspective of
the case study organization. The study intends to answer the following research questions:

- R1 How did the cloud migration process compare to the process described in
established frameworks for cloud migration?

- R2 What challenges did the case study organization encounter during the cloud
migration process?

- R3 What success factors and learnings can be extracted from the evaluated cloud
migration process?

This study focuses on providing a synoptic description of a cloud migration process, to get a
general overview of the different activities and steps included in the process.

2
2 Theory
The following section presents general aspects of cloud computing and research on cloud
migration.

2.1 What is cloud computing?

Cloud computing is the on-demand service delivery in the form of compute power, databases,
storage, servers and applications (AWS, 2023). Key characteristics for cloud computing are
elasticity of workloads, pay-as-you-go pricing, storing of data across multiple data centers,
and the fact that it is all done over the internet. These characteristics enable scalability,
reliability and availability, which are frequently used terms to describe the overall benefits of
cloud computing (AWS, 2023; Microsoft Learn, 2023). Scalability in the cloud refers to being
able to add or remove resources such as storage capacity or system components, which allow
cloud consumers to quickly expand, lower costs, and facilitate disaster recovery (AWS, 2023).
Cloud services offer scalability by built-in-features that enable automatic scaling of resources
based on demand, while simultaneously allowing the cloud consumer to plan for expected
peaks with scheduled auto scale (AWS, 2023; Microsoft Learn, 2023). Reliability and
availability in the cloud refers to making sure that services are up and running with high
accessibility and security (AWS, 2023). With cloud computing, data is encrypted and
scattered through multiple storage units/servers (Hassan et al., 2022) which enables
applications to run with minimal interruption or downtime on secure connections (AWS,
2023). Furthermore, since the services or applications are accessed through internet
connection, they enable the cloud consumer to access applications at any time and from any
location (AWS, 2023; Microsoft Learn, 2023). This is particularly valuable for companies
with offices or outlets distributed across different countries and for companies with the need
for constant activity or monitoring (Tatić et al., 2020).

2.1.1 Cloud service models

With cloud computing there is an emphasis on a shared responsibility model between the
cloud provider and the cloud consumer that varies depending on the cloud service model
(Microsoft Learn, 2023). There are three common service models for cloud computing:
Infrastructure as a Service (IaaS), Platform as a Service (PaaS), and Software as a Service
(SaaS) (AWS, 2023; CSA, 2023).

3
 - IaaS provides the basic structure for cloud IT infrastructure in terms of hardware and
software, such as virtual machines, storage and databases (AWS, 2023). This service
model provides the maximum level of flexibility for the cloud consumer to control
their IT resources. However, IaaS simultaneously places the most amount of
responsibility on the cloud consumer in comparison to the other service models
(Microsoft Learn, 2023). The cloud provider is responsible for maintaining and
securing the hardware and the network connectivity, while the cloud consumer is
responsible for securing and maintaining the infrastructure in terms of installation,
patching, and updating.

- PaaS provides the infrastructure, middleware, operating systems, and development

tools, which enables the cloud consumer to develop and deploy applications in a
flexible way (AWS, 2023; Microsoft Learn, 2023). In terms of the shared
responsibility model, the cloud provider is in this case responsible for the basic
infrastructure as well as maintaining the next level of structure (operating systems,
development tools and databases for example). The cloud consumer simply manages
the applications on the platform and is responsible for keeping it up to date.

- SaaS represents a complete product in the form of an application that is completely

managed and secured by the cloud provider (AWS, 2023; CSA, 2022). Consequently,
this service model places the least amount of responsibility on the cloud consumer,
resulting in the least amount of control or influence of the application. Even though
the security of the application is placed on the cloud provider, the access of users,
information and data, and devices are the responsibility of the cloud consumer (CSA,
2022).

Although there are certain implications of responsibilities for each cloud service model, it is
common to provide a detailed description of obligations and responsibilities for each party in
a Service Level Agreement (SLA) between the cloud consumer and cloud provider (Microsoft
Learn, 2023).

2.2 Challenges and recommendations for cloud migration projects

The decision to migrate to the cloud is often based on key motives such as a need for
scalability to meet market needs, streamlining of business processes, or improving user

4
experience or competitive strategy (Tatić et al., 2020). In contrast, factors that are perceived
as challenges for cloud migration are cost of investment, worries about privacy, lack of
in-house expertise, and project complexity. It is important to note that these challenges are
similar to that of any IT migration or software development project, and having a
methodological step-by-step approach could benefit all IT projects (Gholami et al., 2016).
However, Gholami et al. (2016) point out that there are certain unique features to cloud
technology, such as interoperability, availability, and multi-tenancy, that make a cloud
migration differ from a legacy software implementation. How much these factors influence a
cloud migration often depends on the type of migration, such as if there is a migration of
infrastructure (IaaS) or replacement of components or an entire application (SaaS) (Gholami
et al., 2016; CSA, 2022). Such factors are presented and explained further in the following
sections.

2.2.1 Security and privacy

Shuaib et al. (2019) argue that security and privacy are two main challenges for cloud
migrations, with concerns such as Distributed Denial of Service (DDoS) attacks, data theft,
data loss, and identity and access management (IAM). Migrating to cloud can entail complete
or partial move of organizational data from applications or databases to the cloud, depending
on the service model (Rad et al., 2017). Even though privacy and security of data is important
for all types of applications, it becomes essential for cloud applications due to data being
divided and stored in several locations without direct physical access by the cloud consumer
(Hassan et al., 2022). Different cryptographic and non-cryptographic techniques have been
applied to improve and ensure data protection in the cloud, however, security and privacy
remain two challenges for cloud migrations. Moreover, Hassan et al. (2022) argue that in
certain cases it will be more important for the cloud consumer to establish data privacy than
data security, since the cloud provider will have access to all cloud user’s data.

It is perhaps easy to assume that a cloud application of the type SaaS would signify less
security risks than that of an IaaS which involves change of infrastructure (CSA, 2022).
However, the Cloud Security Alliance (2022) states that not much guidance exists regarding
security of SaaS implementations, despite many organizations using such applications to run
critical business operations and storing sensitive data within the application. Thus, it is
important for the cloud consumer to understand what type of data will be stored in the cloud
and what parties will have access to it (CSA, 2022).

5
2.2.2 Availability
Availability is one of the key benefits to cloud technology since cloud applications allow
users access regardless of location, time or device (AWS, 2023; Microsoft Learn, 2023; Tatić
et al., 2020). On the other hand, availability can also pose a challenge for cloud providers,
since cloud applications require broad bandwidth network connection and are based on
multi-tenancy (Gholami et al., 2016). Multi-tenancy in cloud computing is the ability for
multiple cloud consumers (e.g. tenants) to use the same computing resources by a single cloud
provider, which maximizes the utilization of resources. This principle also signifies that one
tenant could consume more CPU or bandwidth than others and thus cause degraded quality of
service to other tenants. This effect is referred to as the ‘noisy neighbor problem’ and is most
often an unintended effect by the cloud consumer (Microsoft Learn, 2023), and a potential
risk for SaaS applications in particular (Gholami et al., 2016). It can, however, also be caused
intentionally by cloud consumers trying to exploit vulnerabilities of the system through DoS
or DDoS attacks (Microsoft Learn, 2023).

Additional threats to availability are concerns such as cloud unit outages, network failure and
temporary service unavailability (Gholami et al., 2016). Even though cloud providers work
with countermeasures for such factors, including problems with multi-tenancy, some of these
factors are out of the cloud provider’s control (Microsoft Learn, 2023; Gholami et al., 2016).

2.2.3 Interoperability
Interoperability is the working communication between systems, which can pose a challenge
for cloud migrations due to differences in technologies between cloud providers or systems
(Gholami et al., 2016). Because of incompatibilities between cloud technologies, it can be
difficult to move applications or switch technologies from one provider to another, resulting
in a vendor lock-in. ISO/IEC 19941:2017 came as a standard to ensure that there was a
common understanding between cloud providers on interoperability and portability
(International Organization for Standardization [ISO], 2017). Even so, there are still
challenges of compatibility when integrating existing hardware and software with cloud
technology that make a cloud migration complex (Tatić et al., 2020). For instance, the specific
attributes of cloud such as elasticity and multi-tenancy might present problems when
migrating a legacy application to run on cloud technology (Gholami et al., 2016).

6
2.2.4 Cloud expertise and project planning
There are also a few non-technical hinders associated with cloud migrations, such as lack of a
well-defined cloud strategy, uncertainty about when to initiate cloud migration, little
knowledge of cloud technology, and project complexity (Tatić et al., 2020; Shuaib et al.,
2019). When the expertise on the cloud application is mainly provided by the cloud provider,
support managers also risk the need for adding more resources to compensate for lack of
in-house expertise (Khajeh-Hosseini et al., 2010). Having staff with competence in cloud
computing in the development team and project management, is therefore a success factor to
cloud migration projects (Tatić et al., 2020). Cloud expertise can, for instance, assist the
organization when navigating through complex factors such as regulations, SLAs,
interoperability, scalability and feasibility (Shuaib et al., 2019). Moreover, Gholami et al.
(2016) suggest that establishing distinct roles, responsibilities and required expertise can
facilitate the cloud migration process, particularly for developers with less experience in
cloud technology.

2.2.5 Responsibility and control

The cloud provider assists a cloud consumer with the development and maintenance of a
cloud application to various degrees depending on the service model (AWS, 2023; Microsoft
Learn, 2023). However, Khajeh-Hosseini et al. (2010) concluded in their study that there are
also certain challenges with cloud migrations that come as a result of having a third-party
service provider (cloud provider). One of the identified challenges were for support
technicians, sales and marketing staff, and customer care staff to get less control and influence
over technical aspects or arisen issues. A reason for this is the need of having to report and
chase issues with the cloud provider and becoming dependent on their responsiveness. This
could, in turn, affect quality of service, user support or customer service if issues or tasks take
longer time to get resolved during and after migration (Khajeh-Hosseini et al., 2010).

A SLA can help clarify the services, obligations, and responsibilities of the cloud provider by
specifying aspects such as performance, billing, legal documents, and data backup (Shuaib et
al., 2019). This is valuable for the cloud consumer, since there is less confusion and more
clear expectations of the services provided by the cloud provider. Shuaib et al. (2019) argue
that SLAs are often an overlooked topic with cloud migrations, despite the fact that the
execution and design varies greatly depending on the cloud provider. Having a well

7
formulated SLA can thereby be an important success factor for organizations performing a
cloud migration (Shuaib et al., 2019).

2.2.6 Cloud acceptance

Morgan & Conboy (2013) discovered in their research that the majority of challenges related
to cloud migrations are user related and not technical. For instance, end users may initially
need more time to perform tasks in a new cloud environment, and if there is little in-house
expertise or training during the cloud migration, the quality of service could be affected
negatively (Khajeh-Hosseini et al., 2010). Additionally, any annoyance or hardship that a user
experiences with a cloud application can potentially discourage the usage and acceptance of a
cloud application (Amron et al., 2022). The success of the cloud application implementation
is therefore dependent on the end users’ readiness, willingness and acceptance. Factors that
positively affect the readiness and acceptance are an optimistic view of the cloud application
that is to be implemented, as well as confidence and knowledge in the technology. To promote
project success, it is therefore important to work towards establishing cloud acceptance by
having sufficient training, support, and information for the end users (Amron et al., 2022;
Morgan & Conboy, 2013).

Even though user acceptance is an important success factor to any IT-project, it is emphasized
in this case because of it being highlighted by several studies as an important factor for cloud
migration projects (Morgan & Conboy, 2013; Amron et al., 2022).

2.3 Frameworks for cloud migration

This section presents three studies with proposed frameworks for cloud migration that
describe the cloud migration process in a step-by-step approach.

Khan & Al-Yasiri (2016)

Khan & Al-Yasiri (2016) developed a framework for cloud computing adoption for SMEs,
that describes the migration process in three stages: cloud requirement stage, cloud
preparation stage, and cloud migration stage. These three stages are also described as the:
understanding stage (to-be), essential stage (as-is), and deployment stage (transformation).
The framework distinguishes the migration process according to roles and responsibilities,
and the level and size of the organization.

8
 - The cloud requirement stage signifies establishing the choice of deployment model
(public, private or hybrid) and service model (SaaS, IaaS or PaaS), according to the
size and level of the organization (Khan & Al-Yasiri, 2016).

- The cloud preparation stage is the stage of trying to understand current business
processes and conditions of the organization to assess the appropriate planning and
measures for a cloud migration (Khan & Al-Yasiri, 2016). This is done by performing
a feasibility study, identifying stakeholder needs (cloud provider and enterprise user),
assessing organization readiness, and using a tool called 4 steps to cloud. In several of
these steps, risks and governance for the migration process are analyzed and planned
for.

- The cloud migration stage is the process of going live with the cloud application,
which entails selecting a provider and establishing a contract, testing and deploying
the cloud application, and lastly, monitoring and performing maintenance of the
application (Khan & Al-Yasiri, 2016).

Moscoso-Zea et al. (2018)

Moscoso-Zea et al. (2018) argued that the previous research on cloud migration had not
provided any sufficient frameworks or guides on how to successfully go through the process
of cloud migration step-by-step. Consequently, Moscoso-Zea et al. (2018) proposed a seven
step framework for the cloud migration process.

Step 1, current situation definition, refers to establishing the internal and external
characteristics of the organization that should be taken into account in the decision-making of
cloud migration. These characteristics are: organization (size and age), product/service (the
value offer), infrastructure (self-hosted or third-party implementation), customers (customer
distribution and customer role), and environment (competing companies and target market
size).

Step 2, requirements definition, refers to establishing the requirements of a cloud

application/implementation according to different aspects such as: performance, reliability,
capacity, availability, security, regulations and operativity.

9
Step 3, providers analysis, refers to reviewing the available cloud providers and their
respective alternatives for implementation.

Step 4, providers assessment, refers to setting the business objectives and evaluating the
alternatives provided by a certain cloud provider that satisfy these objectives according to
technical, legal and economical feasibility.

Step 5, decision, refers to the process of making a decision in terms of cloud provider as well
as cloud computing model.

Step 6, migration, refers to the process of going live with the cloud application. For this step,
Moscoso-Zea et al. (2018) adopted a seven step process proposed by Amazon Web Services:
● Get acquainted with the specific cloud environment
● Build a support network for the cloud application within the organization
● Test existing software in the new cloud environment
● Migrate company data
● Migrate company specific applications
● Set up options for scalability, security and maintenance
● Deploy application

Step 7, follow-up, refers to monitoring and logging performance. This step was added to the
earlier proposed framework by Paredes-Gualtor et al. (2017), because of a need to review and
assure performance after the cloud application has been deployed.

Zbořil & Svatá (2022)

Zbořil & Svatá (2022) created a cloud adoption framework based on the guidelines and
checklists for cloud migration by the cloud providers Microsoft and Amazon by describing
the cloud migration process step-by-step. A phase that was not included by either cloud
provider was a vendor negotiation, which Zbořil & Svatá (2022) added as the second phase of
their framework. Their cloud adoption framework consists of 8 phases according to the
perspectives: business, people, governance, platform, security, and operations. The phases are
presented below in a more detailed view.

10
Step 1, strategy, refers to establishing cloud migration drivers, developing a business case,
formulating a cloud adoption plan, and analyzing stakeholders involvement.

Step 2, vendor negotiate, refers to specifying vendor requirements, assessing and choosing a
vendor, defining the shared responsibility model, and establishing a contract.

Step 3, plan, refers to creating a plan for data migration, planning of roles and responsibilities,
existing skills and organizational readiness.

Step 4, ready, refers to preparing the environment, configuration of platform, specifying steps
of procedures, and establishing IAM and proof of concept. Training of administrators and
users is also performed at this stage.

Step 5, adopt-migrate, refers to the migration process with resource migration to the cloud and
solution is implemented.

Step 6, adopt-innovate, refers to refining and improving current cloud implementation by

looking at application development, threat protection and technology innovation.

Step 7, govern, refers to establishing a governance methodology, and defining cloud

organizational policies and required expertise in cloud technology.

Step 8, manage, refers to managing the cloud application with monitoring through operations
management, cloud service lifecycle management, and license management.

Theoretical framework
The three frameworks for cloud migration presented in this section were used as a basis for
argumentation and comparison to the cloud migration process in the case study. The
frameworks by Moscoso-Zea et al. (2018) and Zbořil & Svatá (2022) were considered the
most relevant for this study, because of a more extensive step-by-step depiction than that of
the framework by Khan & Al-Yasiri (2016).

11
3 Method
This study was an explanatory study based on interpretivist research theory, which allows for
a deeper understanding of a subject and answering research questions of the type how or why
(Fischer & Ziviani, 2004). The research design was a case study, which is a suitable research
method for performing scientific evaluations (Yin, 2012) to explain casual relationships
(Fischer & Ziviani, 2004). This was an embedded single-case study, signifying that the case
study focuses on a single case with multiple units of data (Yin, 2012). The research approach
was abductive, due to the need for switching between theory and data (Jacobsen, 2017) when
comparing previous studies on the cloud migration process with the case in question. Basing
the case study on theory can help organize the initial data analysis and generalize findings
(Yin, 2012), which is why this comparative analysis was conducted.

3.1 Case introduction

The case study organization is a large sized Swedish, international retail company with stores
in multiple European countries. The organization executed a cloud migration project between
2020-2022 for their retail stores in Sweden, Norway and Finland. The background for the
project was a need for an Enterprise Resource Planning system (ERP) that would enable the
company to expand, facilitate e-commerce, and streamline resources. The point-of-sale
system (POS) at the time, was facing replacement in the near future, which led to the
conclusion to include the replacement of the POS system in the overall project. This would be
an implementation of a SaaS that, because of the large venture, would be planned and
executed as a separate cloud migration project. The project started around 2020 in terms of
planning investment and execution, and went live during spring of 2022.

3.2 Source criticism

The sources used for the theory section were a combination of research articles, reports and
internet sources. Bouali & Kolinsky (2023) emphasize the importance of evaluating the
reliability and credibility of a source by looking at factors such as when the information was
published, by whom and to what purpose. Tardiff (2022) claims that source criticism should
not be treated as a checklist for reliable sources, and stresses the importance of critical
reflection when evaluation of the authority and accuracy of sources. These aspects were taken
into account when evaluating the credibility and relevance of the sources used in this study.

12
For sections 2.1 and 2.2, which describe cloud computing and challenges with cloud
migration, there were several internet sources from Amazon, Microsoft, and the Cloud
Security Alliance (CSA). This was due to a lack of description of the basic concepts of cloud
from other sources, and because these organizations have a substantial amount of technical
documentation on cloud computing. To prevent problems with accuracy and purpose, these
sources were therefore often used in combination as a way to promote objectivity.

Research articles used for sections 2.1 and 2.2 were found through the databases available
through the university library of Luleå Technical University. The search terms used were
‘cloud migration’, ‘cloud migration process’, ‘cloud adoption’, and ‘cloud computing’. To
account for currency and authority of the sources, the selected research articles were
peer-reviewed and published from 2010 and after.

For section 2.3, which presents previously established frameworks, research articles
describing frameworks for cloud adoption or cloud migration were reviewed in several steps.
Due to a differing use of terminology in the topic area, ‘cloud adoption’ was considered an
alternative term to ‘cloud migration’ when reviewing articles. Inclusion criterias were that the
articles needed to be peer-reviewed articles, published from 2010 and later, and provide a
step-by-step depiction of a cloud migration. Five research articles describing frameworks for
cloud migration were selected after an initial review that consisted of reading the abstract,
results, and conclusion. Two frameworks were thereafter excluded after further review
because of a lack of a methodological step-by-step description of a cloud migration. However,
the frameworks by Khan & Al-Yasiri (2016), Moscoso-Zea et al. (2018), and Zbořil & Svatá
(2022) provide a step-by-step depiction of a cloud migration process from start to finish, and
were therefore regarded as more relevant for this study.

3.3 Data collection

Data collection consisted of a combination of methods that collected both qualitative and
quantitative data, which is a common approach for case studies (SAGE, 2007; Leavy, 2014;
Yin, 2012). The data collection included a survey, a document review, and an archival data
review with the aim to obtain triangulation by having multiple sources of data pointing to the

13
same results (Yin, 2012). Fischer & Ziviani (2004) argue that quantitative data is an important
addition to the qualitative data to case studies for this very reason.

3.3.1 Document review and archival data review

The first stage of data collection was performing the archival data review and document
review, which was to be the basis for describing the overall cloud migration process of the
case study organization. The document review consisted of gathering and analyzing stored
documents regarding the project, such as requirement specification, planning and schedules.
The archival data review consisted of data collected through the organization’s incident report
system and messages gathered through the company’s communication platform.

3.3.2 Survey
The purpose of the survey was therefore to gather data directly from the end users, to provide
insights on training/introduction, project management, cloud acceptance, and challenges. An
additional purpose of the survey was to reassert data findings from the document review and
archival data review, by having triangulation of data (Yin, 2012).

The survey was designed as an online questionnaire that was sent out to all of the Swedish
store managers of the retail company, which were representative of the end users of the cloud
application. An email with a link and QR code to the questionnaire was sent out to a total of
145 Swedish store managers within the organization. The respondents had one week of
response time and were reminded by email before the final response day. The aim was to
achieve a response rate of 60% or higher, which is considered a sufficient response rate for
high validity and reliability (Fincham, 2008).

The questionnaire was constructed in Microsoft Forms and consisted of 14 questions that
were close-ended and multiple choice, with two open-ended questions at the end. The
response options of the multiple response questions were shuffled for each respondent to
avoid a connection between placement of response options, and thus promote higher validity
(Wolf et al., 2016). The questions were organized into five groups to evaluate the user profile,
introduction and training, support, cloud acceptance and communication with the project
management. This questionnaire design was created based on findings in the initial document
review and archival data review as well as research on important aspects for cloud acceptance
by Khajeh-Hosseini et al. (2010), Morgan & Conboy (2013), and Amron et al. (2022). The

14
first group consisted of three demographic questions to gather a sense of the user profile. The
second group consisted of three questions regarding the training and introduction to the cloud
application. The third group consisted of three questions regarding the contact between the
end user and support services before and during rollout. The fourth group was regarding User
Experience, to evaluate the opinion, knowledge and confidence in the cloud application. For
this group there were two Likert scales, which is a common psychometric tool to evaluate an
attitude or opinion on a subject (Joshi et al., 2015). In this case, the respondents were asked to
rate general aspects and specific functions of the cloud application. The fifth, and final group,
was regarding how the respondents perceived updates/changes as well as the information on
updates during the project. The group consisted of one Likert scale where the respondents had
to rate how much they agreed to three different statements on the matter.

Due to Swedish being the native language of the respondents, the questionnaire was
constructed in Swedish to motivate for higher participation and validity of results (Wolf et al.,
2016). However, in the presentation of results, the questions were translated to English for the
purpose of facilitating interpretation and presentation for non-swedish speakers. Internal
terms and phrasing was controlled by having the system owner and project manager review
the questions of the questionnaire. As an example, the cloud based POS system was referred
to as ‘the new checkout’ (‘nya kassan’) in the questionnaire, since this was a more common
phrasing amongst the users. However, the wording in the translated version of the questions
was ‘the new system’ when referring to the same application, to facilitate the interpretation of
questions and responses. The original, Swedish, formulation of questions can be found in
appendix A, although with the company name and specific internal terms excluded from the
text.

3.4 Data analysis

The document review and the archival data review were the basis for describing the cloud
migration process of the case study organization in results, section 1. Documents such as
requirements specification, planning schedules and presentations were screened for relevance,
analyzed and thereafter described in text form. Incidents and problems described in the
incident report system, in association to the cloud migration project, were compiled into an
Excel file. A thematic analysis was thereafter made to pin-point patterns that could be used to
describe the cloud migration process in steps, as well as the challenges encountered during the

15
project. A thematic analysis is a method where patterns and themes in qualitative data are
identified and thereafter used to describe an issue (Braun & Clarke, 2006). Based on the
identified patterns and themes, the cloud migration process was divided into six steps that
were used as a baseline for describing the results:
1. Current situation (as is)
2. Requirements specification
3. Providers analysis
4. Assessment of migration
5. Migration phase
6. Maintenance

Factors or aspects that could be considered challenges to the cloud migration process were
clustered into two categories: support & technical challenges and organizational & project
management challenges.

Survey
The survey responses were collected through Microsoft Forms and exported to an Excel file.
The data was compiled according to percentage and number of responses, with an exception
for the Likert scales. As recommended for this type of question, weights were added to the
response scale according to a 5 point scale (Joshi et al., 2015), see below in table 1.

Table 1. Response scale weight

Likert response Weight Likert response Weight

Strongly disagree 1 Very poor 1

Disagree 2 Poor 2

Undecided 3 Undecided 3

Agree 4 Good 4

Strongly agree 5 Very good 5

These weights were thereafter used for calculating the weighted mean or median of each
Likert scale item response, depending on if the statements for each question should be treated
as ordinal or interval data (Joshi et al., 2015). If the items are interrelated and together form a
basis for measuring the attitude, the item response could be treated as an interval scale by

16
looking at central tendency and the mean (Joshi et al., 2015; Sullivan & Artino, 2013). In this
study, two of the Likert scales were treated as such, with calculations of frequency rate and
means for each item statement and an overall mean for the question. The final Likert scale
had independent statements and was therefore presented in frequency rate and median for
each statement.

The open-ended questions at the end of the survey were analyzed and presented in text form
as generalizations of statements, including a few quotations of the respondents.

Theoretical framework as basis for argumentation

The three frameworks for cloud migration, presented in section 2.3, were used as a basis for
comparison between the evaluated cloud migration process and theory, which was performed
in the analysis and discussion section (section 5).

3.5 Ethical considerations

This study took several measures to account for ethical concerns such as anonymity and
integrity. One measure was to keep the company anonymous in the presentation and
evaluation of the case which allowed deeper exploration and depiction of the case. For the
survey, the respondents did not share any demographic information, such as age or gender,
since this was considered irrelevant for the data collection. The respondents were also
informed regarding the use of data for the study and that the participation was voluntary and
anonymous, to ensure good survey ethics such as consent, anonymity and integrity (Wolf et
al., 2016).

3.6 Method discussion

Validity is to ensure the method is measuring what it is supposed to, while reliability is to
ensure consistency of the results from a method (Wolf et al., 2016). The validity and
reliability of case studies are topics that have been vigorously discussed due to complexities
with the case study method (Yin, 2012). Transparency and clear descriptions of methodology,
a mix between qualitative and quantitative data, and separating data from interpretation are
thus important factors to increase validity and reliability of a case study (Fincher & Ziviani,
2004). Concerns and tradeoffs in methodology that could have a negative impact on validity
and reliability of the study are discussed further in this section.

17
Validity
One factor that could affect validity of the document review is that the provided data was
mostly obtained by the case study organization, which could result in a potentially biased
picture. The triangulation of data by combining different methods such as archival data,
documentation, and survey, was therefore a critical factor to promote validity of this case
study. These are all factors that were taken into account when designing the research design
for the data collection. The archival data review and survey provided an amount of
quantitative data that was a valuable addition to the qualitative data collected from the
document review and the archival data review.

Validity is also an essential factor to consider when designing a survey, since answers to
questions are dependent on the formulation and construction of questions asked (Wolf et al.,
2016). There were several measures taken to strengthen the validity of the survey, such as
formulating questions with familiar wordings and ensuring that the questions were in the
respondents native language (Wolf et al., 2016). Another measure was to shuffle response
options of the multiple-choice questions to prevent respondents from choosing a response
option based on order (Wolf et al., 2016). There were also measures to minimize loss of
partial data by having the respondents receive a pop-up that urged them to fill out a particular
question before proceeding, in case Microsoft Forms detected partial loss of data.

It is important to note that the questions of the survey were later translated into English when
presented in the result section, to enable interpretation for English speaking readers. The
original, Swedish formulation, is presented in appendix A to provide transparency to the
survey design. Since there was no comparison between data from different forms, and the
translations are stating what the respondents had already reported, the validity of the results
were most likely not affected by the translation on a larger scale.

To improve validity of the study further, interviews would have been an applicable addition,
to provide more qualitative data and insights from members of the project management.

Reliability
Reliability is a potential challenge for case study research due to the uniqueness of cases and
the difficulties of reproducing results when observing people or organizations (Yin, 2012).
There are, however, actions to take to improve the reliability of case studies that were

18
incorporated into this study as well. For instance, the methodology of the survey design was
explained in a manner to provide transparency and insight by describing tools, questions and
phrasing. These measures should allow for repeating the survey in a similar execution and
thereby strengthen the reliability of the study (Wolf et al., 2016). Reliability of the document
review and archival data review is more difficult to evaluate, due to the variation of data
depending on the case study organization. Transparency is an important factor to counteract
this concern (Yin, 2012), which in this case is illustrated by clarifying what type of documents
and archival data was reviewed and to which section of results they were used. One measure
to further improve reliability of these reviews could have been to create a case study database
where all data could be gathered without initial interpretation, which would promote further
transparency of the study.

19
4 Results
This section presents the results of the case study evaluation, according to two subsections:
document review and archival data review, and survey.

4.1 Document review and archival data review

The results of the document review and archival data review are presented by describing the
case study organizations’ cloud migration process in the following steps: current situation - as
is, providers analysis, requirements specification, assessment of migration, migration phase,
and maintenance. There is thereafter a section describing the challenges encountered during
the migration process.

4.1.1 Current situation (as is)

The case study organization is a large Swedish retail company of over 10 years of age. The
organization’s product/service offer is retail with an emphasis on value-creation for the
customer and maintaining a good customer relationship. The organization has stores in
several European countries with their customer base primarily concentrated around
Scandinavia, and business-to-consumer (B2C) as their customer role.

IT-infrastructure
The IT-infrastructure of the Swedish stores, prior to the cloud migration, consisted of a legacy
POS system that was server based. For each store, the POS system consisted of one or several
check-outs (computer, screen and software application) and a separate computer that acted as
an onsite POS server to which the data was transferred. Each check-out station also had a
DPP, a receipt printer, and a scanner for registering items. The organization also had a
Customer Relational Management (CRM) system for which data of customers were stored
and accessed by the POS system.

Business goal and needs

The decision to switch to cloud was based on the need for a cloud based ERP system that
would enable growth and streamlining of the organization, particularly to facilitate startups of
new stores and gather all outlets for sales, products, and price into one master channel. Other
reasons were to enable flexibility and improvement of promotions, as well as personalization

20
of customer relationships. The need for an ERP system greatly determined the following
planning and design of architecture, and resulted in a larger migration project with several
sub-projects.

The decision to implement a cloud based POS system was powered by the fact that the digital
payment platform (DPP) at the time was in need of replacement, and the POS system was
facing replacement in the near future as well. For the purpose of streamlining and gathering
resources, it was considered beneficial for the organization to choose the cloud based POS
system by the same provider as the selected ERP system. The implementation of a cloud
based POS system was treated as a separate project by the project management, although
executed simultaneously as the implementations of a new DPP, CRM, and ERP system. The
business need was thus for an omnichannel cloud application for commerce of the service
model SaaS, that would enable in-store and back-office operations. It was necessary for the
organization to have a hybrid deployment model due to the need for open access while at the
same time maintaining certain parts of the business processes private.

4.1.2 Providers analysis

For the case study organization, the decision to implement a cloud based ERP system was
made before establishing requirements or assessing possible solutions by a cloud provider.
This led to the decision of adopting the cloud application for commerce by the same cloud
provider, in order to integrate systems and projects. In regards to this, no providers analysis
was made to compare solutions of different cloud providers.

4.1.3 Requirements specification

For the case study organization, there were a few summarized business priority requirements
regarding functionality, usability, graphics/design, and project structure that were specified at
the beginning of the project. However, the specification of requirements for the cloud
application was for the most part performed simultaneously as assessing the cloud solution.

The end users of the cloud application for commerce were mainly the store managers and
store employees. The main vision and focus was therefore the need for an efficient and
smooth POS process so that the store personnel could focus on the interaction with the
customers. For functionality, there were certain requirements specific for a retail setting, such
as a system that:

21
 ● was adapted to the country specific laws, rules and regulations
● allows customers to purchase items with payment options such as credit cards, gift
cards, invoice, and cash, depending on the country
● enables discounts and promotions
● allows customers to make refunds or reclaims
● allows for management of membership and bonus checks
● facilitates end-of-day administration
● enables administration of deliveries and yearly inventories

It was also specified that it was important to have a system with high reliability and
availability, easy navigation with few clicks and system windows, and graphics that match the
organizational standard.

Besides the basic functionality requirements, there were also a few requirements for the cloud
migration project in general, such as:
● updated routines and safety manual
● training and instructions based on different roles and tasks
● dedicated support group from IT and business unit during deployment
● information and manuals in native language
● daily access to service desk, as well as evenings and weekends

The more detailed requirement specifications were constructed during the process of
designing the company specific solution, and refined over the course of the migration process
into functional categories. A more detailed requirement specification was created
post-migration.

4.1.4 Assessment of migration

The decision was made to implement the cloud provider's SaaS solution for commerce, which
consisted of an application for POS for the check-outs, an application for daily administration
on the store's workstation computers, and an application for backend administration of
workers and their accounts (see figure 2 below). There was also a User Acceptance Testing
(UAT) environment that would be used for the testing of the cloud application, and as a
training environment for the end users.

22
The end users of the cloud application for commerce would mainly be the store personnel and
store managers, particularly for the application for POS and daily administration. The
application for backend administration would occasionally be used by store managers, yet
more frequently by support staff, technical managers, and the finance department.

Figure 2. Architecture of cloud application for commerce

The organization decided to partner up with a third-party deployment partner that specialized
in cloud implementations for retail, and had previous experience of the cloud application for
commerce. The deployment partner was to assist the organization with deployment,
development, trouble-shooting, and contacting the cloud provider. It was also decided that the
installations would be outsourced to a third-party service provider that would set up and
install the equipment at each store according to an installation protocol. The new
IT-infrastructure for the stores would entail: new hardware (computer, screen, customer
screen, and DPP), the cloud application as software, existing peripheral equipment (receipt
printer, scanner, keyboard and mouse), a new CRM system by another provider, and a
fiber-optic network connection.

The implementation of an ERP system corresponds to an IaaS, which places responsibility on

the cloud consumer regarding securing and maintaining the infrastructure with installation,
updates and patches (Microsoft Learn, 2023). Although for the cloud application for
commerce, which was an implementation of a SaaS, the majority of the responsibility was
therefore placed on the cloud provider according to the shared responsibility model. In this
sense, the responsibility for the case study organization regarding security of the cloud
application for commerce, was primarily to work with Identity and Access Management
(IAM) and to apply the least privilege principle by regulating the access of users. This would

23
be orchestrated through the application for backend administration and operations by
technical managers, support managers and project management. There was also a SLA
established between the cloud provider and the organization to clarify responsibilities and
obligations.

4.1.5 Migration phase

The migration phase started with the creation of a requirement specification and design
solution for each country of deployment before going live. There were five weeks of initial
testing in three phases that involved system testing, system integration testing (SIT), user
acceptance testing (UAT), penetration tests, and bug fixes. This was repeated in iterations
over a course of 11 weeks alongside the preparations for rollout.

A support system was created for the deployment due to the expectancy of increased support
calls during the going live phase. The system was designed so that the stores contacted the
service desk for support with technical issues, and super users (store managers that were
familiar with the cloud application) for support with handling or questions. An extra
deployment team was organized to handle all incoming incidents regarding the cloud
application for commerce during the first few weeks of going live. The new support flow
signified that the service desk or system owner forwarded incidents to the deployment team
by creating tickets in the incident report system that were placed in a specific group handled
by the deployment team (see figure 3). The opening hours for the service desk retail were also
extended on evenings and weekends, and the support staff was reinforced with an additional
staff member during most of the day shifts.

Figure 3. Support flow during critical migration phase

24
Manuals and training material were created and presented to the users (store managers and
store personnel) on the intranet and a shared online storage. The store managers received a
walk-through of the cloud application for commerce by the system owner and were
responsible for the training of their own staff. The training material consisted of
demonstration videos of the most common functions and several test cases that were to be
performed in the UAT environment by all of the users. The users performed their training
around one week prior to their respective rollout, by gaining access to the UAT environment
around that time. The general system manual was posted on the intranet as a digital manual
with division on function and processes, to which the users had access through the cloud
application.

The first step of going live was to install the cloud application on two test-pilot stores on each
market, which was executed in March of 2022. The rollouts for remaining Swedish stores
followed after around four weeks of monitoring the test-pilots and four other pilot stores. The
remaining rollout consisted of around 15-20 store installations per week up until June of
2022.

The rollout protocol was designed so that the stores were contacted by an installation
technician the day before installation to go over the procedure for the following day. The
technician would thereafter arrive in the morning and have the first check-out installed with
the cloud application before the store opened. After each installation of a check-out, a test
protocol was performed by the store manager and technician that ended with sending an email
of approval to the service desk. The two other applications for backend administration and
daily administration had already been installed on the workstation computer of the store
offices prior to the installation day.

4.1.6 Maintenance
A stage of maintenance followed after the migration phase, which consisted of monitoring
and logging of performance. Updates due to bug fixes or new releases were executed by the
deployment team in close collaboration with the deployment partner. The support flow also
shifted from a mode of high attention to maintenance by escalating incidents to the
deployment partner directly, instead of the in-house deployment team. This was executed by
using the organization’s incident report system to create tickets with descriptions of the
incidents/problems, along with event logs for the third-party service provider to review.

25
4.1.7 Challenges
This section presents a summary of identified challenges during the cloud migration process.

Support & technical challenges

The amount of support calls and tickets created in the incident report system increased
significantly during the migration phase, as was expected by the project management. The
number of tickets created by the support staff were 1766 for March - April in 2022 in
comparison to 1387, for the same months the year prior. Commonly reported problems during
the cloud migration process were that the application for POS was slow, or freezing when
performing refunds or dealing with campaigns/discounts. These performance issues were
investigated by the deployment partner and either solved or escalated to the cloud provider.
Another commonly reported problem was that the application for backend administration was
slow or freezing, which resulted in the store managers needing assistance from the service
desk when performing these tasks. This performance issue was also investigated by the
deployment partner and several new versions of the application were released to deal with this
issue.

There were 12 incidents between June of 2022 - January of 2023 that involved quality of
service of the cloud application for commerce, in the Swedish stores. The majority of these
were related to outages in local cloud units which caused the application for POS to be
unusable for around 10-15 minutes. These incidents were all escalated to the deployment
partner for problem solving, who in turn escalated the issues to the cloud provider. Another
incident was caused by the auto-scale logic of the application for POS, due to cloud service
units being automatically scaled down during the night when the stores were closed. The issue
started when the units auto scaled up and CPU utilization drastically changed upon opening
the stores, due to sales on Black Friday. This caused the deployment partner and deployment
team to go over the settings for auto-scale logic with the cloud provider, to prevent it from
reoccurring on days where there might be a higher increase in sales.

Lastly, an overall technical challenge for the support staff, development team and project
managers during the cloud migration was the integration of existing hardware with new
hardware and software. Since there were several new systems and hardware in the context,
there were times when troubleshooting involved investigating if the problems were related to

26
the cloud application or not. This was for instance the case regarding reported problems that
involved the receipt printers, scanners, DPP, and CRM system. In those cases the deployment
partner would not investigate further due to the problems being out of their scope, leaving the
task for the support managers and development team.

Organizational & project management challenges

There were also a few challenges during the cloud migration process that relate to
organizational factors or project management. An organizational challenge was changes in
project management staff and expertise due to a long hiatus during the covid-19 pandemic.
There was also a decision to execute the cloud migration project using agile methodology,
which was the first time this methodology was used by the organization in a project.
Additionally, there was a high level of complexity due to several large projects of dependence
being executed parallelly, such as implementation of an ERP system, new CRM system, new
DPP and new application for commerce.

All of the aforementioned challenges were recognized by the project management as

contributing factors for delay of the migration phase of the cloud migration process. It was
also recognized by the project management that more emphasis should have been on the as-is
phase and that specification of requirements should have started earlier in the process and not
simultaneously as designing the solution.

4.2 Survey
A total of 89 store managers responded to the questionnaire, which resulted in a response rate
of 61.4%. Due to the survey design with detection of incomplete responses in Microsoft
Forms, there was no partial loss of data.

The results of the first three questions were used to gather a user profile, which is presented
below in table 2. The results of the remaining questions are presented in 4 sections:
introduction and preparation, support, User Experience, and information and communication
channels.

27
Table 2. User profile

Specification Category Frequency Percentage

Work role Store manager 85 96%

Super User 4 4%

Working experience < 1 year 5 6%

at the company
1 - 2 years 9 10%

3 - 4 years 4 4%

5 - 6 years 8 9%

> 6 years 63 71%

Duration of having 3 - 6 months 7 8%

the cloud application
for commerce > 6 months 82 92%

4.2.1 Introduction and preparation

Ⅰ. What information did you receive during the introduction to the new system?

Figure 4. Information received during introduction

The question of received introduction was a multiple response question that allowed the
participants to choose as many options as they would like. The calculations were therefore
displayed as a percentage of respondents who chose a particular response option (see figure
4). The results show that 90% of respondents watched the demonstration videos and 84%
performed the test cases in the system training environment that was provided as an

28
introduction to the cloud application. 24% of the respondents stated that they received a
demonstration by the system owner or a manager and 9% chose the option ‘other’.

Ⅱ. Do you feel that you received sufficient introduction to the new system?

Figure 5. Perception of introduction

A majority of the respondents answered that they felt that the introduction was sufficient, with
a highest percentage of 87% for ‘yes, it was sufficient’ (see figure 5). 9% stated ‘neither/nor’
and 4% ‘no, I would have liked more’.

Ⅲ. How prepared did you feel before the installation of the new system?

Figure 6. Perception of how prepared the respondent felt

The highest percentage for this question was 48% for ‘well prepared’ and thereafter 44% for
‘quite prepared’ (see figure 6). 2% stated ‘poorly prepared’ and 6% stated ‘don’t know’.

29
Overall this points to a majority of the respondents stating that they felt quite prepared or well
prepared prior to installation.

4.2.2 Support
Ⅰ. Was your store contacted by a technician prior to the installation?

Figure 7. Contact with installation technician

The highest percentage for this question was 87% for ‘Yes’ (see figure 7). 4% stated that they
needed to contact the service desk first to be called back and 9% stated ‘don’t know’.

Ⅱ. How often did you contact Service Desk/Super Users regarding the new system during the
first three months?

Figure 8. Contact with service desk or super user

30
The highest percentage for this question was 49% for ‘2-4 times per month’ and thereafter
28% for ‘once a month or less’ (see figure 8).

Ⅲ. What were the most common reasons for contact?

Figure 9. Most common reasons for contact

This question was a multiple response question where the respondents could choose a
maximum of three response options. Two of the response options were provided with
examples of what the category could signify (see figure 9). For peripheral equipment, there
was a clarification that it entailed the scanner and receipt printer. For backend administration,
there was a clarification that it meant the administration of workers such as adding and editing
worker profiles. The results showed that there were four response options that were close in
frequency/percentage. The percentage was 50% for ‘backend administration’, thereafter 45%
for ‘system freezes/slow performance’ and 41% for ‘problems with campaigns/discounts’.
40% of respondents also stated ‘questions regarding handling/function’ as one of the most
common reasons for contact.

31
4.2.3 User Experience
Ⅰ. What is your opinion on the new system regarding the following?

Figure 10. General User Experience of the new system

Table 3. Calculated mean general User Experience

Likert item Mean Category mean

Useful functions 4.06

Reliability and stability 3.88

Learnability 4.13 4.08

Layout and design of menus/buttons 4.29

Simplicity and ease of use 4.03

Regarding general User Experience, the highest percentage was for the response ‘good’ for
each Likert item (see figure 10). The mean also pointed to the same outcome except for the
item regarding reliability and stability, where the mean was slightly below 4 due to the
distribution of answers for this particular item (see table 3). However, the overall category
mean for this question was 4.08, which points to an overall positive attitude for the general
User Experience of the cloud application.

32
Ⅱ. What is your experience of the new system regarding the following functions?

Figure 11. User Experience of different functions

Table 4. Calculated mean function User Experience

Likert item Mean Category mean

Searching for receipts 3.80

Handling of refunds 3.88

Handling of discounts/campaigns 3.27 4.02

Management of members 3.66

Performing purchases 4.60

Opening/closing shifts 4.90

Regarding User Experience of different functions, there was a differentiating distribution of

rating. For opening/closing of shift and performing purchases, the highest percentage was
90% and 61% respectively for ‘very good’ (see figure 11). For the remaining functions, the
most frequent response option was ‘good’ according to percentage of responses. The
distribution of answers was also more evenly spread out for these items, particularly for the

33
items handling of discounts/campaigns and management of members, resulting in a mean
below 4 (see table 4). However, the total category mean was 4.02 for User Experience of
different functions, which points to an overall positive attitude as well.

4.2.4 Information and communication channels

Ⅰ. Below are statements regarding changes or updates to the new system. Mark the answer
closest to your experience.

Figure 12. Experience of information on changes

Table 5. Calculated median communication channels

Likert item statement Median

I would have liked to receive information on changes 4

earlier

I have been informed prior to the change 4

There have been changes too often 2

For the Likert item statement ‘I would have liked to receive information on changes earlier’,
the highest percentage was 40% for ‘agree’ which matches the median of 4 (see figure 12 and
table 5). For the statement ‘I have been informed prior to the change’, the highest percentage
was 46% for ‘agree’ which also matches the median of 4. The final statement ‘there have
been changes too often’, had the highest percentage for ‘agree’ with 34%. However, the

34
median for this statement was 2 due to 22% of respondents choosing ‘disagree’ and 29% of
respondents choosing ‘strongly disagree’.

4.2.5 Additional feedback

At the end of the questionnaire the respondents were given the opportunity to give additional
feedback with two open-ended questions:
● Are there any functions you feel are missing from the new system?
● Do you have any additional feedback for future projects?

The open-ended question regarding functions received 42 responses, where a few respondents
simply stated that they did not feel that any function was missing. The most frequently
mentioned function was the ability to view customer bonus checks in the system, which was
mentioned by 15 respondents as a function they felt was missing from the current system.
There were also several mentions of the handling of members and refunds, which were two
aspects that also received a lower rating in the Likert scale regarding User Experience on
functions. For instance, one respondent stated: “jag och mina kollegor upplever det ganska
"segt" när man ska lägga till medlem” [“me and my colleagues feel that it is pretty ‘slow’
when adding a member…”].

As for additional feedback for future projects, there were a total of 19 responses that did not
indicate any particular theme or pattern. Some examples of responses as feedback for future
projects were “gärna fysisk person i butik, som introducerar stora nya förändringar”
[“preferably a physical person in the stores when introducing new big changes”], and “varit
lite rörigt och frustrerande med rabatthanteringen i kassan” [“has been a little confusing and
frustrating with the handling of discounts in the system”]. There were also several
respondents that simply described that they felt satisfied with the support they have gotten
during the project.

35
5 Analysis & Discussion
The following section presents interpretations and discussions of the results of the case study
evaluation. The terms ‘cloud migration process’ and ‘organization’ are used to describe the
evaluated cloud migration process and organization in the case study.

5.1 Triangulation of data

The purpose of having multiple data collection methods was to achieve triangulation by
having different data pointing to the same results. Yin (2012) argues that triangulation of data
is a critical method to strengthen the validity of a case study. For this study, there were
multiple cases of triangulation where findings from the document review, archival data
review, or survey confirmed results from one of the other data collecting methods. The survey
contributed primarily with quantitative data from the perspective of end users, which made
findings from the document review and archival data review more plausible and objective.

The open-ended questions in the survey did not contribute with any new insights, although
they did provide some explanations to findings from other questions. For instance, several
respondents answered that they feel that member recruiting is time-consuming and
complicated, which provided a possible explanation to why the rating of handling of members
was lower than other functions in the Likert scale on functions.

5.2 R1 How did the cloud migration process compare to the process
described in established cloud migration frameworks?

The cloud migration process had many similarities to the process described in the frameworks
on cloud migrations, particularly regarding recommendations of factors to consider and
activities to perform. For instance, the cloud migration process contained the activities:
looking at the current state of business processes and feasibility, specifying requirements,
assessing a solution, migrating, and maintenance. These are activities described and
recommended in the frameworks by Khan & Al-Yasiri (2016), Moscoso-Zea et al. (2018), and
Zbořil & Svatá (2022), even though the order and execution varies between frameworks.

36
Differences between the cloud migration process and the process described in the frameworks
were primarily regarding the order of execution and the absence of a comparison of cloud
providers. Another difference was the focus on data security, governance, and risk
management, which was emphasized by the frameworks by Khan & Al-Yasiri and Zbořil &
Svatá (2022). For the case study organization, these aspects were mostly orchestrated by
constructing a SLA between themselves and the cloud provider (with the help of the
deployment partner), and implementing recommendations for IAM of the cloud application.
Reasons for this approach are most likely due to the fact that the cloud application was of the
service model SaaS, and that the organization trusted security recommendations from the
cloud provider and deployment partner. The deployment partner had previous experience with
deploying and maintaining the exact same application for commerce for other organizations,
which likely made the case study organization trust their guidance and recommendations.
Even so, the emphasis of security and risk management during the cloud migration process
differs from what is portrayed in the evaluated frameworks for cloud migrations.

All of the similarities and differences are discussed further below in relation to each step of
the cloud migration process: current situation - as is, providers analysis, requirements
specification, assessment of migration, migration phase, and maintenance.

Current situation - as is
The first step of the cloud migration process consisted of establishing the current situation and
specific conditions of the organization, which is similar to the first phase of all three evaluated
frameworks. The frameworks by Khan & Al-Yasiri (2016) and Moscoso-Zea et al. (2018)
both depict selecting a service model and deployment model according to the size and level of
the organization. For the case study organization, these factors were most likely affected by
the decision to implement an ERP system, which was already decided upon when determining
the need for a cloud application for commerce. The frameworks by Moscoso-Zea et al. (2018)
and Zbořil & Svatá (2022) also focus on establishing the customer role, value-offer and
business case at the first stage of the process, which is consistent with the cloud migration
process in this stage as well. For instance, the organization had expressed early on that a
business goal was to improve customer offers and relationships, by having a system that
would allow the store personnel to focus on customers when performing their working
activities.

37
Requirements specification
Specifying requirements is an activity included in the cloud migration process described in all
of the evaluated frameworks, although with differences in order and characteristics. The
framework by Khan & Al-Yasiri (2016) does not have a specific phase of establishing
requirements, however, it has a step called ‘identifying stakeholder needs’ in the ‘cloud
preparation stage’ where user needs are pinpointed. The framework by Moscoso-Zea et al.
(2018) places requirements specification as step 2 in the cloud migration process. For Zbořil
& Svatá (2022), their framework includes specifying requirements in the phase of vendor
negotiation, by focusing on requirements of the cloud provider.

Although the placement of establishing requirements varies in the frameworks by Khan &
Al-Yasiri (2016), Moscoso-Zea et al. (2018), and Zbořil & Svatá (2022), it was performed
considerably late in the cloud migration process for the case study organization. This was also
recognized by the project management as one reason for delay of the migration phase. There
were a few specifications early in the cloud migration process, although the majority of
requirements were specified simultaneously as designing the solution in the assessment of
migration phase.

Providers analysis
Comparing and choosing a cloud provider is a stage of the cloud migration process that Zbořil
& Svatá (2022) describe as the second step of the cloud migration process. Khan & Al-Yasiri
(2016) and Moscoso-Zea et al. (2018) consider this step to belong later in the cloud migration
process, after establishing requirements and initial planning. Additionally, Moscoso-Zea et al.
(2018) divide the process of assessing and choosing a provider into three separate steps:
providers analysis, providers assessment, and decision.

For the case study organization, an analysis of providers was not performed in the cloud
migration process, because of the precedent decision to implement an IaaS by one of the
major cloud providers. Therefore, the analysis of providers consisted solely of establishing
that the cloud provider offered a cloud based solution for commerce. The decision to choose
the same cloud provider was most likely based on factors such as interoperability and
streamlining, since it would be an advantage for the organization to have fully integrated
systems. This approach is similar to that of the guidelines by cloud providers Microsoft and
Amazon that prompted Zbořil & Svatá (2022) to add ‘vendor negotiate’ to their framework.

38
However, this could reflect the reality of other organizations when determining what cloud
application to implement. If an organization uses other applications or infrastructure by one
cloud provider, it is not far-fetched to assume that they would be more inclined to choose the
available alternatives by the same cloud provider.

Assessment of migration
The next step of the cloud migration process was the assessment of migration, which could be
compared to step 4 and 5 (providers assessment and decision) in the framework by
Moscoso-Zea et al. (2018) and step 3 and 4 (plan and ready) in the framework by Zbořil &
Svatá (2022). In this phase, the case study organization evaluated and planned a feasible cloud
solution according to business objectives and available alternatives by the cloud provider,
which is similar to the recommendations by Moscoso-Zea et al. (2018). The organization also
established the roles and responsibilities of the project management and development team,
and organized support flow and IAM. These are also activities specified in the two phases
‘plan’ and ‘ready’ in the framework by Zbořil & Svatá (2022). Thus, for this phase of the
cloud migration, there were similar characteristics to the cloud migration processes described
in the frameworks by Moscoso-Zea et al. (2018) and Zbořil & Svatá (2022).

Migration phase
The migration phase of the cloud migration process shared many similarities to the
framework by Moscoso-Zea et al. (2018) in regards to the activities carried out within the
phase. The main reason for this is most likely that Moscoso-Zea et al. (2018) specified this
stage in a more detailed account than Khan & Al-Yasiri (2016) and Zbořil & Svatá (2022) in
their respective framework. The stage of migration consisted of building a support network,
training users, testing and developing with the UAT, migrating data, setting up options for
maintenance and deploying the application. These are all steps included in the framework by
Moscoso-Zea et al. (2018), although based on the guidelines from Amazon.

Maintenance
The frameworks by Khan & Al-Yasiri (2016), Moscoso-Zea et al. (2018), and Zbořil & Svatá
(2022) included monitoring and logging of performance as a final phase of the cloud
migration process. The evaluated cloud migration process showed a phase of maintenance by
monitoring and logging after the critical migration phase as well. There was an initial phase of
high attention where the cloud application for commerce was monitored closely by the project

39
management and development team that eventually shifted to low attention. All three
evaluated frameworks did not go into much detail regarding this phase, which also indicates
that the cloud migration process did not differ from the description provided by the
frameworks.

5.3 R2 What challenges did the case study organization encounter during
the cloud migration process?

The results showed that challenges during the cloud migration process related to
characteristics of the technology, interoperability, project management, project complexity,
and support. There were three identified themes based on the results of all three data
collection methods: technical challenges, project management and project complexity, and
support and troubleshooting, as seen below. These are explained further below.

Technical challenges
The end users reported a few initial technical difficulties with the cloud application for
commerce during the cloud migration process, which is depicted as a barrier to cloud
acceptance by Amron et al. (2022) and Morgan & Conboy (2013). Amron et al. (2022) argue
that if a user struggles a lot with a cloud application it can potentially discourage usage and
cloud acceptance. The results of the survey showed that the three most common reasons for
contact were regarding the application for backend administration, the system freezing or
showing slow performance, and problems with campaigns/discounts. The Likert scale
regarding User Experience of different functions also showed a lower score for the item
‘handling of discounts/campaigns’, with a mean below 4 (see table 4). These results coincide
with the created tickets in the incident report system, where common descriptions were
assistance with the administration of workers (assistance with the application for backend
administration), incidents of the system freezing or having slow performance, and problems
with campaigns/discounts. Even so, it is hard to determine if these challenges were in fact
related to the implementation of cloud technology or not, since initial problems with an
application may occur with any new technology.

The incidents when the cloud application for POS temporarily went down, have a more
apparent connection to specific characteristics of cloud technology, such as availability,
network connection, multi-tenancy, and scalability. Gholami et al. (2016) argue that problems

40
with slow performance or system failure could be related to loss of network connection,
outages in cloud units, or multi-tenancy where quality of service is affected by other cloud
consumers' usage of resources. The cause for the majority of incidents were outages in cloud
service units, which affected availability since the cloud application was unusable during that
time. Scalability, in combination with multi-tenancy, was the cause of another incident where
the auto-scale logic resulted in performance issues when the usage of CPU drastically shifted
due to a high increase of sales on Black Friday. Interestingly enough, scalability has not been
depicted as a challenge for cloud migrations by a majority of research articles on barriers for
cloud migration. Gholami et al. (2016) describe how scalability can become a challenge when
deploying legacy applications to run on cloud, since it does not equate that the legacy
application is able to take advantage of the elasticity of the cloud. This was not the case for
the case study organization, since the application used was a SaaS application. Shuaib et al.
(2019) mention scalability as a potential barrier for cloud migration, yet do not specify
reasons for this other than pointing to the complexity for project management to navigate
through settings for scalability.

Another technical challenge during the cloud migration was regarding interoperability, due to
integration of existing hardware with the new hardware and new software, which at times
caused compatibility issues. Interoperability is pointed out as a challenge specific for cloud
migration projects by Tatić et al. (2020), and Shuaib et al. (2019). For the case study
organization, it was at times difficult for the support staff, support managers, and development
team to determine if occurring problems were related to the cloud application or not, which
complicated troubleshooting of incidents and problems.

Project management and project complexity

There were a few challenges during the cloud migration process that can be linked to project
management and project complexity, which are common barriers to cloud migration
according to Shuaib et al. (2019). Firstly, there were changes of staff members of the project
management during the initial phases of the cloud migration process, which may have
complicated project planning. The project was also structured according to agile
methodology, which was the first time this methodology was used by the organization, adding
to an already high level of project complexity (Tatić et al., 2020). Lastly, there were several
projects of dependence being executed at the same time that involved cooperation between

41
project management of various degrees. This led to challenges with the planning and
execution of the cloud migration process, and ultimately, a delay of the migration phase.

Support and troubleshooting

The results of the archival data review also showed an increase in the number of tickets
created in the incident report systems, which indicates that there was an increased need of
user support during the migration phase of the cloud migration process. This finding also
triangulates with the results of the survey, where the majority of respondents stated that they
contacted the service desk or super users 2-4 times per month. Furthermore, Khajeh-Hosseini
et al. (2010) claim that an increased need for user support or customer support is a common
challenge for cloud migrations.

Other challenges during the cloud migration process can be associated with the shared
responsibility model of cloud computing. Incidents were reported to the support staff (of the
service desk) and thereafter escalated to the deployment partner, that in turn escalated to the
cloud provider. This support flow added extra steps to the process of troubleshooting, which
made support managers, support staff and deployment team more dependent on the service
and responsiveness of both the deployment partner and cloud provider. This resulted in the
support staff having less control over the issue, which Khajeh-Hosseini et al. (2010) describe
as a barrier to cloud migration. This issue can also correspond to the specific cloud service
model SaaS, which places more responsibility on the cloud provider, which consequently
results in less control for the cloud consumer.

5.4 R3 What success factors and learnings can be extracted from the
evaluated cloud migration process?

The cloud migration process contained several project management aspects that are
recommended for cloud migrations by Khan & Al-Yasiri, Moscoso-Zea et al. (2018), and
Zbořil & Svatá (2022). For instance, in-house expertise is promoted by Khajeh-Hosseini et al.
(2010), Tatić et al. (2020), and Shuaib et al. (2019) as an important success factor for cloud
migration projects. For the case study organization, the project management tried to
compensate for a lack of in-house cloud expertise by partnering up with a deployment partner
that had previous experience of the chosen cloud application. Gholami et al. (2016) also argue
that determining roles, responsibilities, and required expertise will help the project

42
management and development team during the cloud migration, which the case study
organization showed signs of by arranging a dedicated support group and business unit during
deployment. The project management also planned for an increase in user support by
extending working hours of the service desk, adding a separate communication line for
questions to ‘super users’ and having updated system manuals. This matches the
recommendations by Khajeh-Hosseini et al. (2010) and Amron et al. (2022) to plan for an
increase in user support during cloud migrations, to counteract a decrease in quality of service
and promote cloud acceptance. Furthermore, there are also indications that the support flow
and communication provided to the end users was overall successful, since the majority of
survey respondents stated that they were contacted by an installation technician prior to
installation, had received information prior to changes and did not feel that there had been
changes too often.

There are also indicators of project success of the evaluated cloud migration that are related to
cloud acceptance. There was a lot of effort put into organizing the migration project so that
the deployment of the cloud application should not affect customer interaction and working
activities of the store personnel. This was demonstrated by the planning and arrangement for
training, introduction, and design of the cloud application. The survey results showed that the
majority of respondents partook in the introduction and training that was arranged by the
project management. A majority of respondents also stated that they felt that they received
sufficient introduction/training and felt quite prepared or well prepared before the installation.
This points to adequate training and preparation during the cloud migration process, which is
an important factor for cloud acceptance and cloud migration project success (Morgan &
Conboy 2013; Khajeh-Hosseini et al., 2010). In fact, Morgan & Conboy (2013) state that the
challenges related to cloud migration projects are mainly user related, which was not the case
for this cloud migration project and therefore further supports the conclusion that the training
and introduction were success factors. The majority of the end users also rated the general
User Experience and User Experience of functions as overall positive, which points to cloud
acceptance (Amron et al. (2022) and that requirements and customizations for the retail
specific setting were successful.

43
6 Conclusion
This study explored a cloud migration project by evaluating the cloud migration process
stepwise from an organizational perspective. The study intended to answer the research
questions: ‘how did the cloud migration process compare to the process described in
established cloud migration frameworks?’, ‘what challenges did the case study organization
encounter during the cloud migration process’, and ‘what success factors and learnings can be
extracted from the evaluated cloud migration process?’.

The results showed that the evaluated cloud migration process included many activities
recommended by the established cloud migration frameworks, such as requirement
specification, establishing current situation, and performing maintenance. The steps included
in the migration phase of the cloud migration process were particularly similar to descriptions
by the frameworks. In contrast, the main differences between the cloud migration process and
the frameworks were that the case study organization had not performed a comparison
between cloud providers, and that there was less emphasis on data security and risk
management.

The main challenges for the cloud migration process were related to interoperability and
project complexity, by having multiple system integrations and simultaneous projects of
dependence. Other challenges were related to support management, since there was an
increased need for user support after deployment, and limitations in troubleshooting due to
having a third-party service provider. The success factors of the cloud migration process were
primarily related to project management and cloud acceptance, since the cloud migration
process showed signs of accurate planning for support flow and deployment, proper training
and introduction, and sufficient user support.

The evaluated cloud migration process shared many aspects and qualities with the cloud
migration frameworks that were used for comparison in this study. Thus, a contribution of this
study is the indication that these frameworks are in fact representative of cloud migration
processes, and could thereby be used as guidelines for organizations looking to migrate to
cloud. Lastly, an additional contribution of this study is the identification of challenges and
success factors encountered during the cloud migration, which are aspects that could be
important for organizations to consider when preparing for cloud migration.

44
6.1 Further studies
This case study evaluation explored a cloud migration process where no framework for cloud
migration was used to support or facilitate the process. For future studies, it would therefore
be interesting to explore outcomes of cloud migration projects where the migration process is
performed according to established frameworks. This could further determine the usefulness
of such frameworks, as well as provide means for review and refinement of them.

Another suggestion for future studies is to explore the concept of post-project evaluation of a
cloud migration process. During the case study evaluation, the cloud migration process was
most likely still at the maintenance phase, for which the inclusion of a post-project evaluation
may have added an extra dimension to the phase. Moscoso-Zea et al. (2018) name the last
phase of the cloud migration process ‘follow-up’, yet specify that it is the follow-up of
performance by monitoring and logging. It might, however, be beneficial for an organization
to do a follow-up of the entire cloud migration project to learn from project outcomes and
evaluate if business goals have been met. This could be of value for the organization’s project
management, stakeholders and end users. Future studies could therefore investigate the
impact and usefulness of a post-migration evaluation and if it should be included as a step of
the cloud migration process.

45
7 Ethical reflection
It is important to note that the author of this thesis has worked as a consultant for the case
study organization. Although the author did not receive any financial support for performing
this study, it is still relevant to acknowledge the potential risk of personal bias. An important
part of mitigating personal bias was to include a survey as part of the data collection, to gather
the perspective of end users who experienced the project first-hand. Lastly, as mentioned in
the method section, triangulation of data was an important factor to provide transparency into
the study, and thereby mitigate potential bias.

46
References
Amazon Web Services. (January 2023). Six advantages of cloud computing. AWS.
https://docs.aws.amazon.com/whitepapers/latest/aws-overview/six-advantages-of-cloud-comp
uting.html

Amazon Web Services. (February 2023). What is cloud computing?. AWS.

https://docs.aws.amazon.com/whitepapers/latest/aws-overview/what-is-cloud-computing.html

Amazon Web Services (February 2023). AWS Application Auto Scaling. AWS.
https://aws.amazon.com/autoscaling/

Amron, M. T., Ibrahim, R., & Bakar, N. A. A. (2022). The Impact of Technology Readiness
on User Acceptance of Government Cloud Computing Services. 2022 International Congress
on Human-Computer Interaction, Optimization and Robotic Applications (HORA),
Human-Computer Interaction, Optimization and Robotic Applications (HORA), 2022
International Congress On, 1–7. https://doi.org/10.1109/HORA55278.2022.9799952

Bouali, H., & Kolinsky, R. (2023). Source evaluation: Components and impacts. Thinking
Skills and Creativity, 47. https://doi.org/10.1016/j.tsc.2023.101250

Braun, V. & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research
in Psychology, 3, 77-101.

Brobeck, W.M., Givins, J.G., & Meads, P.F. (1976). Point-of-sale system and apparatus
(Patent US 3946220). https://patents.google.com/patent/US3946220

Cisco. (April, 2023). What is Identity Access Management (IAM)?

https://www.cisco.com/c/en/us/products/security/identity-services-engine/what-is-identity-acc
ess-management.html
Cloud Security Alliance (CSA). (April, 2023). SLAS.
https://ea.cloudsecurityalliance.org/display.php?id=data_info4003&_ga=2.218520923.331475
919.1681822635-588150760.1681822635

Cloud Security Alliance (CSA). (2022). SaaS Governance Best Practices for Cloud
Customers. Retrieved from:
https://cloudsecurityalliance.org/artifacts/saas-governance-best-practices-for-cloud-customers/

Fincham J.E. (2008). Response rates and responsiveness for surveys, standards, and the
Journal. American Journal of Pharmaceutical Education, 72(2).
https://doi.org/10.5688/aj720243

Fisher, I., & Ziviani, J. (2004). Explanatory case studies: Implications and applications for
clinical research. Australian Occupational Therapy Journal, 51(4), 185-191–191.
https://doi-org.proxy.lib.ltu.se/10.1111/j.1440-1630.2004.00446.x

Gholami, M. F., Daneshgar, F., Low, G., & Beydoun, G. (2016). Cloud migration process—A
survey, evaluation framework, and open challenges. Journal of Systems & Software, 120,
31–69. https://doi.org/10.1016/j.jss.2016.06.068

Hassan, J., Shehzad, D., Habib, U., Aftab, M.U., Ahmad, M., Kuleev, R., & Mazzara, M.
(2022). The Rise of Cloud Computing: Data Protection, Privacy, and Open Research
Challenges-A Systematic Literature Review (SLR). Comput Intell Neurosci.
https://doi.org/10.1155/2022/8303504

International Organization for Standardization. (2017). Information technology — Cloud

computing — Interoperability and portability (ISO/IEC Standard No. 19941). Retrieved from:
https://www.iso.org/standard/66639.html

Jacobsen, D. I. (2017). Hur genomför man undersökningar? Lund: Studentlitteratur.

Joshi, A., Kale, S., Chandel, S., & Pal, D.K. (2015). Likert Scale: Explored and Explained.
British Journal of Applied Science & Technology, 7(4), 396–403.
https://doi.org/10.9734/BJAST/2015/14975
Kankia, Y. B., Che, X., Ji, W., & Ip, B. (2022). Comprehensive Technology Readiness
Adoption Model: A Combined Approach to Assessing Cloud Technology in Higher
Education in Nigeria. 2022 22nd International Conference on Advances in ICT for Emerging
Regions (ICTer), Advances in ICT for Emerging Regions (ICTer), 110–117.
https://doi-org.proxy.lib.ltu.se/10.1109/ICTer58063.2022.10024101

Khan, N., & Al-Yasiri, A. (2016). Framework for cloud computing adoption: A road map for
Smes to cloud migration. International Journal on Cloud Computing: Services and
Architecture (IJCCSA), 5(6). https://doi.org/10.5121/ijccsa.2015.5601

Khajeh-Hosseini, A., Greenwood, D., & Sommerville, I. (2010). Cloud Migration: A Case
Study of Migrating an Enterprise IT System to IaaS. 2010 IEEE 3rd International Conference
on Cloud Computing, Cloud Computing (CLOUD), 2010 IEEE 3rd International Conference
On, 450–457. https://doi-org.proxy.lib.ltu.se/10.1109/CLOUD.2010.37

Leavy, P. (2014). The Oxford handbook of qualitative research [Web resource]. New York:
Oxford University Press.

Microsoft Dynamics 365. (May, 2023). What is CRM?

https://dynamics.microsoft.com/en-us/crm/what-is-crm/

Microsoft Dynamics 365. (May, 2023). What is ERP?

https://dynamics.microsoft.com/en-us/erp/define-erp/

Microsoft Learn. (February 2023). Describe Infrastructure as a Service.

https://learn.microsoft.com/en-us/training/modules/describe-cloud-service-types/2-describe-in
frastructure-service

Microsoft Learn. (January 2023). Microsoft Cloud Adoption Framework for Azure: Strategy.
https://learn.microsoft.com/en-us/azure/cloud-adoption-framework/strategy/
Microsoft Learn. (February 2023). Monitor performance for scalability and reliability.
https://learn.microsoft.com/en-us/azure/architecture/framework/scalability/monitor-scalability
-reliability

Microsoft Learn. (April, 2023). Noisy Neighbor antipattern.

https://learn.microsoft.com/en-us/azure/architecture/antipatterns/noisy-neighbor/noisy-neighb
or

Microsoft Learn. (April, 2023). What are Azure regions and Availability zones?
https://learn.microsoft.com/en-us/azure/reliability/availability-zones-overview

Microsoft Support. (April, 2023). All about processors (CPUs).

https://support.microsoft.com/en-us/windows/all-about-processors-cpus-06dc72ec-3de2-4eb8-
8cc2-7e5f2417b90b

Morgan, L., & Conboy, K. (2013). Key Factors Impacting Cloud Computing Adoption.
Computer, 46(10), 97–99. https://doi-org.proxy.lib.ltu.se/10.1109/MC.2013.362

Moscoso-Zea, O., Saa, P., Paredes-Gualtor, J., & Sandoval, F. (2018). Moving the IT
Infrastructure to the Cloud. Enfoque UTE, 9(1), pp. 79 - 89.
https://doi.org/10.29019/enfoqueute.v9n1.219

Paredes-Gualtor, J., Moscoso-Zea, O., Saa, P., Sandoval, F., & Rodas, P. (2017). Unified
Cloud Computing Adoption Framework. 2017 International Conference on Information
Systems and Computer Science (INCISCOS), Information Systems and Computer Science
(INCISCOS), 247–252. https://doi-org.proxy.lib.ltu.se/10.1109/INCISCOS.2017.58

Rad, B.B., Diaby, T., & Rana, M.E. (2017). Cloud computing adoption: a short review of
issues and challenges. In: Proceedings of the 2017 International Conference on E-commerce,
E-Business and E-Government, 51–55.

Sage Publications. (2007). The SAGE encyclopedia of social science research methods.
Thousand Oaks, Calif.: Sage.
Shuaib, M., Alam, S., Siddiqui, S. T., & Samad, A. (2019). Why Adopting Cloud Is Still a
Challenge?—A Review on Issues and Challenges for Cloud Migration in Organizations (Vol.
904). Springer Verlag. https://doi-org.proxy.lib.ltu.se/10.1007/978-981-13-5934-7_35

Sullivan, G.M., & Artino, A.R. Jr. (2013). Analyzing and interpreting data from likert-type
scales. J Grad Med Educ, 5(4), 541-2. https://doi.org/10.4300/JGME-5-4-18

Tardiff, A. B. (2022). Have a CCOW: A CRAAP Alternative for the Internet Age. Journal of
Information Literacy, 16(1), 119–130.

Tatić, K., Džafić, Z. Haračić, M., & Haračić, M. (2020). The Analysis of the Key Drivers and
Barriers of Cloud Migration in Companies in Bosnia and Herzegovina. Economic Review,
XVIII(1), 21–36.

Wolf, C., Joye, D., Smith, T.W., & Fu, Y. (2016). The SAGE handbook of survey methodology.
London: SAGE.

Yin, R.K. (2012). Applications of case study research. (3. ed.) Thousand Oaks, Calif.: SAGE.

Zbořil, M., & Svatá, V. (2022). Cloud Adoption Framework. Procedia Computer Science,
207, 483–493. https://doi-org.proxy.lib.ltu.se/10.1016/j.procs.2022.09.103
Appendix A - Survey questions

1. Vad är din arbetsroll?

⚪ Store manager
⚪ Super user (Välj detta alternativ även om du är både store manager och super
user)

2. Hur länge har du arbetat i organisationen?

⚪ Mindre än 1 år
⚪ 1 - 2 år
⚪ 3 - 4 år
⚪ 5 - 6 år
⚪ Mer än 6 år
3. Hur länge har du använt den nya kassan?
⚪ 3 - 6 månader
⚪ Över 6 månader
4. Vilken information tog du del av under introduktionen av nya
kassan? Flera svarsalternativ kan anges.
⬜ Inspelade demonstrationsfilmer
⬜ Testfall för träning i kassans testmiljö
⬜ Genomgång från produktansvarig eller närmsta chef
⬜ Kassamanual från intranätet
⬜ Annat
5. Upplever du att du fick tillräckligt med material och introduktion till
nya kassan?
⚪ Ja, det var tillräckligt
 ⚪ Nej, jag hade velat få mer
⚪ Varken eller
6. Hur förberedd kände du dig inför installationen av nya kassan?
⚪ Mycket förberedd
⚪ Ganska förberedd
⚪ Ganska oförberedd
⚪ Mycket oförberedd
⚪ Vet inte
7. Blev er butik kontaktad av installationstekniker inför installationen
av nya kassan?
⚪ Ja
⚪ Nej, inte alls
⚪ Nej, vi fick ta kontakt med Service Desk Retail för att därefter bli uppringda
⚪ Vet inte
8. Hur ofta kontaktade du Service Desk Retail/Super Users gällande den
nya kassan under de tre första månaderna? Välj det svarsalternativ
som stämmer bäst in.
⚪ 4 gånger/vecka eller mer
⚪ 2-3 gånger/vecka
⚪ 2-4 gånger/månad
⚪ 1 gång/månad eller mer sällan
⚪ Aldrig -> Gå vidare till fråga 10
9. Vad var de vanligaste orsakerna till kontakt? Max tre svarsalternativ
kan anges.
⬜ Problem med kringutrustning (skanner, kvittoskrivare)
 ⬜ Handhavande eller frågor om funktion
⬜ Problem med betalning (Klarna, Swish, kortbetalning)
⬜ Problem med rabatter/kampanjer
⬜ Kassan har hängt sig/långsam kassa
⬜ Backend administration (exempelvis lägga till personal)
⬜ Problem med lokala presentkort
⬜ Annat
10. Vad är din uppfattning om nya kassan gällande följande?

Mycket Ganska Varken eller Ganska bra Mycket bra

dåligt dåligt
Enkelhet och

användarvänlighet ⚪ ⚪ ⚪ ⚪ ⚪
Layout och design ⚪ ⚪ ⚪ ⚪ ⚪
på meny/knappar

Lätthet att lära sig ⚪ ⚪ ⚪ ⚪ ⚪

Pålitlighet ⚪ ⚪ ⚪ ⚪ ⚪
och stabilitet

Användbara ⚪ ⚪ ⚪ ⚪ ⚪
funktioner
 11. Vad är din upplevelse av nya kassan gällande följande funktioner?

Mycket Ganska Varken eller Ganska bra Mycket bra

dåligt dåligt
Öppning/stäng

ning av skift ⚪ ⚪ ⚪ ⚪ ⚪
Genomföra ⚪ ⚪ ⚪ ⚪ ⚪
köp

Hantering av ⚪ ⚪ ⚪ ⚪ ⚪
medlem (skapa, lägga till etc)

Hantering av ⚪ ⚪ ⚪ ⚪ ⚪
rabatter/kampanjer

Hantering av ⚪ ⚪ ⚪ ⚪ ⚪
returer

Söka kvitto ⚪ ⚪ ⚪ ⚪ ⚪
 12. Nedan följer några påstående som relaterar till när det har skett
uppdateringar eller förändringar i nya kassan. Markera det svar som
är närmast din upplevelse.

Instämmer Instämmer Vet inte Instämmer Instämmer

inte alls inte helt delvis i hög grad

Det har skett ⚪ ⚪ ⚪ ⚪ ⚪

förändringar
alltför ofta

Jag har blivit ⚪ ⚪ ⚪ ⚪ ⚪

informerad innan
förändringar har skett

Jag hade velat ⚪ ⚪ ⚪ ⚪ ⚪

få information om
förändringar tidigare

13. Finns det någon funktion du tycker saknas i nya kassan?

14. Har du någon ytterligare feedback du vill ge oss inför framtida

projekt?