ERP IMPLEMENTATION CHALLENGES

Vendors’ Perspective

Jim Odhiambo Otieno

School of Business and computing Science, Middlesex University

Burroughs Roads, NW4 4BT, U.K.

Keywords: ERP, Challenges, Critical Success Factors, Implementation, Kenya.

Abstract: Enterprise Resource Planning (ERP) systems have transformed the way organizations go about the process

of providing information systems. They promise to provide an off-the-shelf solution to the information

needs of organizations. Despite that promise, implementation projects are plagued with much publicized

failures and abandoned projects. Efforts to make ERP systems successful in organizations are facing

challenges. The purpose of the study reported in this paper was to investigate the challenges faced by

organisations implementing ERP systems in Kenya based on consultant point of view. Based on the factors

identified from the interview, a survey was administered to ERP consultants from five Kenyan organisations

that were identified as having a key role in ERP systems implementation in their firms in order to assess the

criticality of the identified challenges. A factor analysis of these items identified six underlying dimensions.

The findings of this study should provide to management of firms implementing ERP systems a better

understanding of the likely challenges they may face and put in place appropriate measure to help in

mitigating the risk of implementation failures.

1 INTRODUCTION

An Enterprise Resource Planning (ERP) system

is an integrated set of programs that provides

support for core organizational activities such as

manufacturing and logistics, finance and accounting,

sales and marketing, and human resources. An ERP

system helps the different parts of an organization

share data and knowledge, reduce costs, and

improve management of business processes. In spite

of their benefits, many ERP systems fail (Stratman

and Roth, 1999). Implementing an ERP system is a

major undertaking. About 90 percent of ERP

implementations are late or over budget (Martin,

1998) and the success rate of ERP systems

implementation is only about 33% (Zhang et al,

2003, Arif et al., 2005).

Over the past few years limited research has been

conducted about ERP implementation issues: mainly

case studies in individual organizations have been

reported. A major problem with such ERP case

studies is that very few implementation failures

challenges resulting to these failures have been

recorded in the literature, and thus the reasons why

implementations fail are not known to practitioners

and researchers. That is a motivation toward

conducting empirical studies to explore challenges

that affect ERP systems implementation. In the

context of ERP project implementation, challenges

represent major pitfalls which if not addressed then a

project stands little chance of success. This study

examines the challenges faced by user and vendor

organizations implementing ERP systems in Kenya.

Consultants from five consulting organizations, who

were identified as having a key role in ERP systems

implementation, were interview in stage 1 of the

research in order to empirically identify challenges

faced by their client organizations during ERP

implementation. A survey was then conducted to

assess the criticality of the identified challenges

which were perceived by respondents as posing high

risk to successful ERP implementation.

This paper is organized in four sections. First

ERP-related literature is reviewed. The next section

introduces the research methodology, followed by

the presentation of the results. The paper ends with

the conclusions and implications for future research

and practice.

505

Odhiambo Otieno J. (2008).

ERP IMPLEMENTATION CHALLENGES - Vendors’ Perspective.

In Proceedings of the Tenth International Conference on Enterprise Information Systems - DISI, pages 505-510

DOI: 10.5220/0001714605050510

 SciTePress

2 RESEARCH METHOD

The choice of an appropriate research methodology

is critical in guiding researchers on how best to meet

research objectives. In this study, the purpose was to

discover the perceptions and experiences of

consultants involved in ERP implementation with

regard to the challenges faced by the client

organizations during implementation of ERP

systems. The research was conducted in two stages.

First staged involved an interview with consultants

drawn from different functional areas in five

consulting organizations who had been involved in

ERP implementation at some stage. The second

stage involved conducting a survey on the factors

identified in stage 1 to assess their critically. The

analysis has enabled the identification critical

challenges experienced by organizations

implementing ERP systems based on vendors’

perceptions and experiences.

The target of the study was the consulting

organizations that have been involved in ERP

implementation in Kenya. The key informant

method was used for collecting information in a

social setting by surveying (or interviewing) a

selected number of participants. Five consulting

organisations participated in the study. We contacted

the ERP project managers of each company in

charge of ERP implementation. Questionnaires were

sent to the ERP project managers of each firm, who

forwarded the questionnaires to the project team

members in charge of individual processes. Eight

questionnaires were distributed and a total of six

(75%) were returned. Of the companies that

responded half are wholly locally owned while the

other half are fully foreign owned with both

categories having a workforce 20 to 40 staff. The

vendors implemented a cross section of ERP

solutions ranging from SAP, Oracle financials to

others like Sage line 500. Of the vendors that

responded, 38% implement Oracle Financials,

followed closely at SAP (25%), BAAN, Navision

and Ebizframe all equally had 13%.

The questionnaire consisted of two main parts:

the company background and statements which

expressed the challenges identified in stage 1 one of

the study. The first part was designed to determine

characteristics such as size of the company,

ownership, location of company etc. The second part

consisted of nineteen statements about the success

factors of ERP systems implementation, derived

from the literature review. The language used in the

survey was both English and Kiswahili (national

language in Kenya). Translation was rather easy

because Kenyans used original English terms for

many technical and management concepts and

especially for information systems and computing

concepts.

The factors identified at the interview stage were

included in the questionnaire and respondents were

asked to state the extent they disagreed or agreed

with the statements in a likert scale of 1-strongly

disagree, 2- Disagree 3- Indifferent, 4-Agree and 5-

Strongly agree. This method was employed on the

grounds that it avoids the problems of having to

consider ten challenges simultaneously in order to

rank them. The data collected was then analyzed by

using SPSS. Based on the responses, descriptive

statistics, factor analysis (FA) and reliability tests

were carried out to identify the critical challenges

faced by user and vendor organizations during ERP

implementation in Kenya and data validity

respectively.

3 RESULTS

Based on the responses from the vendors and

literature review, we identified 10 factors that could

be considered as major challenges faced by

organisations implementing ERP in Kenya. Table 1

show the factors in order of ranking.

Table 1: Ranking of challenges.

Challenges Mean

High cost involved 4.99

Lack of IT skills among users 4.94

Poor ICT infrastructure 4.83

Data conversion problems 4.81

Inadequate time allocated for training 4.71

Integration problems 4.68

Government ICT policies 4.67

High staff turnover 4.61

Scarcity of funds 4.56

Inadequate preparation by employees 4.50

The individual mean value of the Likert rating

scale is the popular usage indicator for measuring an

item's importance, without regard to the other items:

so the higher the value the more important the

factor. All items are rated above the 3.0 scale (mid-

point).

ICEIS 2008 - International Conference on Enterprise Information Systems

506

Table 2.

Challenges Mean F1 F2 F3

High cost involved 4.99 .792

Lack of IT skills among users 4.94 .682

Poor ICT infrastructure 4.83 .689

Data conversion problems 4.81 .815

Inadequate time allocated for training 4.71 .792

Integration problems 4.68 .610

Government ICT policy 4.67 .556

High staff turnover 4.61 .651

Scarcity of funds 4.56 .699

Inadequate preparation by employees 4.50 .815

Eingenvalue

2.58 2.21 2.07

Percentage of variance

18.5 15.54 10.78

Cumulative percentage of variance

37.96 22.7 40.16

Cronbach alpha coefficient

0.83 0.76 0.68

3.1 Factor Analysis

In an attempt to reduce the number of item

(challenges), and to understand their underlying

structure, a factor analysis (FA) was performed. FA

is a data reduction technique that uses correlations

between data variables. The underlying assumption

of FA is that a number of factors exist to explain the

correlations or inter-relationships among observed

variables (Chatfield and Collins, 1992). For the

present study, FA was performed on all nineteen

variables using principal components extraction

(Tabachnick and Fidell, 1989). The goal of this

method is to extract maximum variance from the

data set within each factor. It is basically used to

reduce a large number of variables down to a smaller

number of components. The measure of sampling

adequacy for the twenty items was 0.87 indicating

that the items were suitable for factoring (Kaiser,

1974).

A three-stage factor analysis was conducted with

an orthogonal (varimax) rotation to obtain a stable

factor structure (Rai et al., 1996), resulting in easily

interpretable factors. Under this three-round factor

analysis, items were omitted according to the

following two criteria: (1) no loading greater than

0.45, or (2) loading greater than 0.45 on two or more

factors (Kim and Mueller, 1978).

Table 2 shows the results of this analysis. A first

factor analysis was conducted and produced five

factors. According to the two criteria, one items was

dropped. A second factor analysis on the remaining

9 items resulted in three factors and the dropping of

one item. Finally, a three-factor structure was

derived which kept a total of 8 items after three

iterations. Factor 1 accounted for 37.96% of the total

observed variation; factor 2 and 3 explained 22.7%

and 16.4% of the total variation respectively. The

three-factor solution (cumulative) explained 77% of

the total observed variation. The minimum

eigenvalue from a varimax rotation for which a

factor was to be retained was set at 1.0 in order to

satisfy the minimum eigenvalue criterion (Nunnally,

1978).

Factor 1

, named “High cost further escalated by

extensive customization'', comprises four items

relating to ERP system cost: High system cost, data

conversion too involving, scarcity of funds, and

integration problems. All consultant interviewed

alluded to the fact that cost is a major concern for

organizations implementing ERP systems in Kenya.

The cost of ERPs implementation averages between

Kshs. 100 to 500 million (USD 1.4 – USD 7

million). However, the cost of ERP implementation

varies from one ERP to another. Baan and SAP, two

of the ERP systems implemented by case

organisations, charge software license fee charged

based on: 1) the number users, 2) the type of users,

3) the number of master records in the database.

Users are categorised based on what they can do or

not do in the system.

Cost continues to be a concern for many

organisations. In the case of SAP, their costing

model seemed not favour Africa. For example, the

cost charged per day per consultant is USD 1,200

plus and an additional hotel and overnight stay

allowance of USD 400 charged per night. This is

ERP IMPLEMENTATION CHALLENGES - Vendors’ Perspective

507

three times above the amount charged in Asian

continent like in India that is USD 600 per day.

Implementation costs were found to be, on average

25% percent over budget. Organisations under-

estimated support costs for the year following initial

implementation by an average of 20%. IS maturity

had a major influence on support cost for the year

following initial implementation. Organisations with

low IS maturity experienced an increase in support

cost whereas IS mature organisations experienced a

decrease in costs. This disparity can be traced, in

part, to the widely varying maturity of pre-ERP

environments. Lack of regional standardization and

low budget for IT within Kenya and other African

nations makes it difficult for ERP companies to find

markets with enough potential to justify investing in

costly customisations of the products.

Consultants also noted that benefits of an ERP

application are limited unless it is seamlessly

integrated with other information systems.

Organisations face many challenges relating to ERP

integration – (1) the challenges of integrating

various functional ERP modules, (2) the challenge

of integration with other e-business software

applications, (3) the challenge of integration with

legacy systems. Integration further escalate the cost

of implementing ERP systems. Organisations noted

that legacy systems have accumulated vast amount

of data vital to the survival and operations.

Integration of ERP systems with legacy systems is

more complex than the integration of ERP modules

and Integration of e-business applications. It

routinely requires the installation of third-party

interface software for communication between ERP

software systems and legacy systems. Second

generation ERP systems use relational database

management system (RDBMS) to store enterprise

data. Data conversion from legacy systems to

RDBMS is often a time-consuming and tedious

process.

Integration of the business processes also faced

additional challenges related to new rules built into

ERP software being incompatible with the

established ways of thinking and the norms of

behaviour embedded in the existing work routines.

This is consistent with the idea of ‘best practice’

being situated. Assistant Purchasing in one of the

organisations while explaining that ERP could not

accommodate their work practice said that:

Given our unique requirement, Ebizframe

could not meet our need. The system could

not accept advance payments. It required us

to raise DN [Delivery Note] first then raise

sales invoice and they receive payment

against the invoice. We don’t work that

way.

Factor 2

, named “Lack of IT skills by users and

high staff turnover” comprises four items relating to

lack of capacity to cope with ERP on the part of all

organizational members at all levels due to the

inadequate time for training and high staff turnover:

lack of IT skills among users, inadequate training

time, high staff turnover, and inadequate preparation

by employees.

An all common complaint was the frequency

with which the case study organisations lose key

personnel experienced with ERP or supporting

technologies. Frequently reported problems were:

(1) losing key IT specialists and user representatives

working on the project while the project was going

on, often despite handsome retention bonuses, (2)

losing experienced people after the project was

complete. Many IT specialists thrive on project work

and view assignment as a ‘competence centre’ and

springboard to lucrative opportunities.

One of the major challenges facing ERP systems

implementation in Kenya is the non-existence of

well-qualified employees in implementing

organisations to manage the implementation process

of the system. In one of the case organisations, the

ERP project was supervised by the financial

controller (the ERP Project Manager) and the heads

of administrations (the key users). None of them had

any knowledge about the computer or the ERP

software. The organisations use India and South

Africa as resource base for implementation. The

MIS General Manager said:

All of the company’s leaders were not

qualified to use the computer…They only

trained on the beginnings of Windows and

DOS…Training was internal in finance

department for one week. One week was

not enough. It was just background

information.

Consultant noted that most of the employees in

the implementing organisations are unprepared for

the changes resulting from ERP implementation.

Implementing an ERP will bring in changes to the

way people work within the organization, processes

will change and there may be job cuts and

rationalization of responsibilities within

departments. All this will definitely evoke resistance

from the employees and this has to be managed

effectively before, during and after the

implementation of the ERP package. Consultants

noted that employees are often inadequately

prepared for the major undertaking of ERP

ICEIS 2008 - International Conference on Enterprise Information Systems

508

implementation. Most respondents noted that they

were neither aware of the reasons for the

implementation nor the business benefits that can be

expected by implementing a new ERP package. ERP

projects are therefore perceived as cost cutting

measure by organisations which lead to mass staff

layout. Consultants noted that there was no strong

change management team in place to oversee to

approve, implement and track the changes in the

organization, which includes the impact and detailed

structure (i.e. documentation) associated with the

life cycle of the ERP project.

Factor 3

, named “Poor ICT infrastructure and

ICT policies” comprises two items relating to

government policy on ICT and the state of ICT

infrastructure in the country: poor ICT infrastrucre

and government policy on ICT.

Consultants noted that poor telecommunications,

internet and intranet, and mobile coverage affect

ERP adoption and use in Kenya. Organisations in

Kenya face obstacles related to Kenya’s poor

telecommunication infrastructure in getting the ERP

systems up and running. Organisations operating

nationally, regionally, and internationally must incur

additional costs in providing communication

backbone. Consultant noted that most organisation

do not trust the reliability of local

telecommunications and thus are forced to install

alternative telecommunication means, the most

common being Very Small Aperture Networks

(VSAT). This is expensive by Kenyan standards, but

this is the only way to go about it. However, the

drawback with the VSAT system as mentioned by

the IT manager at a consultant is that “it would

support only batch data transfer (via FTP), which

caused significant delays…up to one-and-a-half

seconds per keystroke…that is not acceptable for

online transactions”.

Even with current networks communication is

quite slow. Two main aspects of ERP system by

infrastructure are the system availability and speed

as commented by a consultant at

PricewaterhouseCoopers “The system cannot run

full month without breaking” and when the system is

available it is likely to be slow.

A consultant noted that a company was forced to

completely abandon implementing web-portal in

their ERP due to lack of internet connectivity in

Muhoroni (town in remote location in Western

Kenya). The cost of setting up a private using VSAT

was exorbitant for the company to afford. Poor

internet connectivity is also hindrance to complete

value chain management and customer relationship

management.

The poor ICT infrastructure in Kenya according

to consultants is because the Kenyan government

exerts excessive control over the national

communications operator. The government often

operates national communications directly through

Telkom and Coomunications Commission of Kenya

(CCK), largely because the private sector is often

incapable operating such an infrastructure. There is

hope that things that ICT infrastructure will improve

once Telkom is privitised early next year, 2009.

Although there has been a move towards

deregulation and privatization of Telkom, the

process is still slow and had often been opposed by

government telecommunication ministries national

security forces.

In combination with inefficient state-run

telecommunications, another major impediment to

ICTs in Kenya come from the onerous tariff

structure. It works out that the average

telecommunications revenue per subscriber line in

Kenya is twice as much in Europe, and four to six

times as much as in USA.

Cronbach alpha coefficients were calculated to

test the reliability of these challenges, as shown in

the last row of Table 2. The reliability of

coefficients obtained ranges from 0.67 (factor 3) to

0.83 (factor 1). Srinivasan (1985) proposed that a

coefficient of 0.7 or higher is acceptable, while a

coefficient of 0.5 or higher is considered sufficient

when dealing with exploratory research combined

with unvalidated data. Thus, the reliability

coefficients in this study are deemed acceptable. The

strength of factor analysis is that it provides a basis

for data reduction. Therefore rather than looking at

all ten items, just three factors can be examined.

That simplifies the rankings and clarifies the most

important items. Rather than focusing on individual

items, practitioners and researchers can focus on the

broad set of items represented by the essential

factors.

4 CONCLUSIONS

All factors – ten – were .rated as critical (rating > 4).

The findings of this study supports Huang and

Palvia’s (2001) view that ERP projects faces

additional challenges in developing countries related

to economic, cultural and basic infrastructure issues.

In fact, ERP implementations in Kenyan

organizations face some unique challenges some of

which are summarised below:

ERP IMPLEMENTATION CHALLENGES - Vendors’ Perspective

509

• Costs of the system are high for many

companies thus putting off many potential

buyers of the systems.

• Inadequacy in resources e.g. funds thus

delaying the implementation process.

• Poor IT skills among users of the system in

the companies.

• Poor change management among

implementing companies with users not

ready for change and minimal management

support in the systems implementation

process.

• Poor ICT infrastructure especially lack of

adequate network coverage nationally

• Inadequate training time allocated to ERPs

implementation. Companies always in a

rush to see the project completed without

due respect to work that has to be done.

• Data migration from existing systems very

involving, some times incompatible with

the new systems

The conflict, caused by the implementation of

ERP systems in developing countries, between the

globalization and localization of management work

practices is worth of further investigations.

However, this conflict is not a unique phenomenon

to developing countries.

REFERENCES

Arif, M., Kulonda, D., Jones, J., and Proctor, M. (2005),

“Enterprise information systems: technology first or

process first?” Business Process Management Journal,

11 (1), pp. 5-21.

Alshawi, S., Themistocleous, M., and Almadani, R.

(2004), “Integrating diverse ERP systems: a case

study” The Journal of Enterprise Information

Management, 17(6), pp. 454-462.

Al-Mudimigh, A., Zairi, M. and Al-Mashari, M. (2001),

“ERP software implementation: an integrative

framework” European Journal of Information Systems,

(10), 216-226.

Bingi, P., Sharma, M. K., and Godla, J. (1999), “Critical

issues affecting an ERP implementation, Information

Systems Management, 16(2), pp. 7–14.

Chatfield, C. and Collins, A.J. (1992) Introduction to

Multi-Variate Analysis, Chapman&Hall, London.

Davenport, T. H. (1998), “Putting the Enterprise into the

Enterprise System”, Harvard Business Review, 76(4,

July-August), 121-131.

Esteves, J., and Pastor, J. (2001), “Analysis of critical

success factors relevance along SAP implementation

phases” in Proceedings of the Seventh Americas

Conference on Information Systems, Boston, 1019–

1025.

Falkowski, G., Pedigo, P., Smith, B., and & Swanson, D.

(1998), “A recipe for ERP success”, Beyond

Computing, September, pp. 44–45.

ICEIS 2008 - International Conference on Enterprise Information Systems

510