ENTERPRISE RESOURCE PLANNING SOLUTION SELECTION CRITERIA IN MEDIUM-SIZED SOUTH AFRICAN COMPANIES

The purpose of this study is to identify the factors that medium-sized South African companies consider important in the acquisition of an Enterprise Resource Planning (ERP) solution. Through an extensive literature review, fourteen ERP selection criteria were identified. Using these fourteen criteria, a questionnaire was developed to test and rate these criteria on a 10-point scale. Results from the survey show that manufacturing and non-manufacturing companies rate reliability, compatibility, service, and support as the most important criteria when selecting an ERP solution.


INTRODUCTION
In recent years, the Enterprise Resource Planning (ERP) market has undergone significant change with notable consolidation between role players.Three of the five leading mid-market ERP vendors in 2000 (JD Edwards, PeopleSoft, and Baan) no longer existed as independent companies by 2007.The ERP market also shows signs of maturation as many large companies have completed significant ERP implementations, resulting in sales slowing down in recent years.In response to this trend, leading ERP vendors have sought out new customers in the 'mid-market', with product and marketing strategies being redesigned to match the needs of these medium-sized companies.
Historically, ERP systems have been targeted at large organisations that are willing and able to spend tens or hundreds of millions of dollars on an integrated software system.Focused attention on the mid-market is a relatively new challenge -one that the software industry is still reorganising to meet.While ERP vendors face challenges in positioning, marketing, pricing, and supporting mid-market clients, clients themselves also face challenges.The ERP market has hundreds of products, services, and methodologies that may not all be applicable to the mid-tier customer base.Customers will have to be armed with the right tools to avoid serious errors and make purchases that will generate good returns.
The purpose of this paper is to identify the factors that medium-sized companies consider important in the acquisition of an Enterprise Resource Planning (ERP) solution.Insights developed from this research can be used to develop guidelines for the structured selection of ERP solutions in the medium-sized market.

RESEARCH PROPOSITION
Companies invariably turn to technology in their efforts to achieve or sustain competitive advantage.There is a constant need to streamline internal processes, cut costs, raise quality, and achieve tighter levels of integration.One way of realising these objectives is through the use of Enterprise Resource Planning solutions.The selection of an enabling technology platform becomes an important strategic issue because the realisation of these competitive targets largely depends on whether a company's systems can provide the required support.The process of selecting, purchasing, and implementing an enterprise solution is a major strategic decision that can either enhance or compromise the strategic position of a company.
Conversely, it must be remembered that ERP implementations can be very risky, and are often considered to be "a wrong purchase…[that] can adversely affect the organisation…even to the point of jeopardising the very existence of the organisation" [22].Leading companies such as Dell and Allied Waste have abandoned ERP projects after initially making large investments in the technology [3].While it is clear that the implementation of ERP systems is an important step towards remaining competitive on a global level, it is also clear that the actual selection of an ERP must be a carefully considered process.
In deciding which selection criteria are to be used for this important decision, the following propositions have been defined to aid the selection process: exchange, and storage standards.Siriginidi [18] highlights the importance of a technically upgradeable product that uses the latest technology.

c) Functionality
"Software features and their functionality are the most significant issue of all" [16].The functionality of the software encapsulates the key elements of value that are being procured.Sprott et al [19] point out that in some application areas, such as finance and accounting, there are globally accepted practices that help to promote a degree of standardisation, while in other application areas, practices may differ between countries.

d) Service and support
Because of the breadth and complexity of issues that must be addressed in the implementation of ERP, it is important that the service provided by the ERP vendor during and after implementation is of a high quality.In a study by Bernroider et al [4], 100% of small-to medium-sized companies rated "good support" as being either very important (42.1%) or important (57.9%).Other researchers also hold the view that service and support is an important ERP selection criterion [8,16,17].

e) Vision of the vendor
"It is desirable to purchase from suppliers who will maintain or improve their competitive position in terms of their products and services" [15].Their vision includes the initiatives and investments that a vendor plans to pursue in the future.These are important to a buyer because they involve key decisions about product evolution, key technical issues, future pricing, efforts to influence the general ERP sector, and the availability of complementary factors such as support, training, and add-on modules.The Gartner Group considers vision important enough to make it a dimension of its 'Magic Quadrant' -a tool that is used for rating product vendors [11].

f) System reliability
"Faults in the system can occur due to disconnected lines, system crashes, or an inappropriate, unreliable or wrong response.Such faults not only decrease productivity but also diminish the confidence in the system" [17].In some businesses, even a few minutes of downtime can cost millions of rands.With the increasing use of the Internet as an enabler of commerce, many online businesses cannot afford to lose system service or experience material degradation in service quality.

g) Compatibility with other systems
ERP solutions will often be integrated with other systems in order to satisfy the unique needs of an organisation.It is important that the selected ERP package can easily be integrated with existing and future software and hardware products [17,19].

h) Ease of customisation (flexibility)
Verville et al [23] observed reluctance on the part of clients to change to fit the system.It was the view of an IT director in one case that "…it is a lot harder to change a process than to change software, and users…want …to change the software" and not their process [23].Bernroider et al [4] made similar observations, finding that 94.7% of small-to medium-sized companies rated adaptability and flexibility of software as either important (26.3%) or very important (68.4%).Other researchers have also highlighted customisation as a key factor in ERP procurement [2,16,17].

i) Market position of the vendor
"Selecting suppliers who will continue to be able to meet the firm's needs will minimize the costs of changing suppliers" [15].In a study by Bernroider et al [4], 55.6% of small-to medium-sized enterprises rated the market position of vendors as being an important factor in ERP vendor selection.ERP implementations often take a long time to complete, and involve large investments of money and effort.The purchase of an ERP system is a long-term commitment, which means that the "ERP provider must be a long-term partner" [2].This implies that the long-term viability of the vendor is an important selection criterion.The market position of the vendor is also an important factor because the 'road map' or vision of the vendor is a consideration in the purchase decision.Buyers must consider the capability of the vendor to continue delivering value-adding features over the long term.If the vendor does not hold a strong market position, their vision may be compromised by a number of factors such as takeovers, etc.

j) Compatibility with organisation structure (organisational fit)
ERP vendors provide mechanisms for customisation.These customisation features can be employed to make the ERP fit the organisation.ERP customisation has been observed to be a major contributor to ERP implementation costs.It is desirable to select the package that will most closely match the organisation's structure and processes.

k) Specialised industry knowledge
The software vendor must have knowledge of the industry in which the software is to be deployed [2,16,18].This enhances the likelihood that ERP implementation will succeed, since industry-specific issues and problems can be accommodated within the solution -or, if not, the company can be made aware of the consequences of its inability to fit the solution.

l) Vendor references
"Because ERP requires a substantial amount of capital investment, the feasibility study involves a greater degree of effort than the typical capital investment analysis" [3].External parties can assist in providing objective information about the performance of vendor products and services [2].Objective, valid data about vendor products and services can be difficult to find [8].Given this fact, it is important that buyers do not rely exclusively on the information supplied by the ERP vendor.Some companies rely on consulting firms to assist in product selection.Piturro [14] warns that consultants are not always fully independent -a fact that can cloud their ability to provide objective input.

m) Fit with parent/allied organisation systems (corporate compliance)
Global enterprises and large groups of companies often have minimal technology standards and guidelines for technology acquisition.The benefits of pursuing corporate standardisation include improved negotiating power with key vendors, a greater ability to redeploy resources without retraining, and ease of collaboration across functions and divisions.Bernroider et al [4] found that 68.4% of small-to medium-sized companies did not consider guidelines from a controlling company to be important in the ERP purchase decision.This was in contrast to responses from large companies in the same study, in which 42% of large companies considered guidelines from a controlling company to be important (18.9%) to very important (24.2%).

n) Implementation time
The cost of ERP implementation includes the indirect cost that companies incur by allocating key staff to the ERP initiative.Resources are often taken away from normal business and assigned to the ERP project on a full time basis.This can result in significant interruptions to normal business operations.In a study conducted by Bernroider et al [4], 95% of small to medium-sized companies considered implementation time to be a key factor in ERP selection.Extended implementations also run the risk of losing key project personnel.Mustacello et al [12] reported instances of ERP implementations that had failed owing to the resignations of key project team members and leadership.

DATA ANALYSIS
Research data was collected by means of a questionnaire, focusing on collecting the appropriate attributes for ERP selection, with a number of filters applied to group respondents.The major measurement scales for the attributes were itemised Likert rating scales.Although these are technically considered to be ordinal in nature, they were assumed to be interval for the majority of analyses [10].Owing to the fact that the data were not normally distributed (confirmed via a Kolmogorv-Smirnov Goodness-of-Fit Test), these scales were treated as ordinal and thus tested using nonparametric analysis.This type of testing was more appropriate (and accurate) than parametric testing for this type of data.The Cronbach's Alpha Test was used to determine reliability of constructs.Moreover, the level of importance for each statement was obtained by multiplying the number of responses on each Likert value for each issue.Thereafter, the means and standard deviations were calculated.Based on the mean value, the most important reason regarding the use of derivatives was determined.In order to rank the nonparametric variables, a Kendall's W Test was used as a measure of the agreement of the rankings of variables across cases [20].Manufacturing companies rated some assessment areas differently from non-manufacturing companies, and in the criteria of organisational fit, reliability, compatibility, functionality, infrastructure, flexibility, service and support, and the market position of vendor.

RESEARCH FINDINGS
The research findings confirmed that the original fourteen criteria were an adequate and important means of selecting an ERP system, supporting proposition 1. Reliability is rated highest by medium-sized companies.This could be an indicator of the strategic nature of ERP purchases: it is critical for the software to deliver the value that it promises, and reliability is a key factor in this equation.Unreliable systems not only "…decrease productivity, but also diminish … confidence in the system" [17].With reliability being the leading ERP selection criterion, it makes logical sense that service and support was also rated high.This could also indicate that there is a strong correlation between the two selection criteria.Service and support is rated second out of a total of fourteen criteria.This is consistent with the findings and views of other researchers such as Bernroider et al [4], Sahay et al [16], Sarkis et al [17], and Hecht [8].
It is evident from the results that medium-size businesses rated service and support, reliability, and compatibility highest when selecting an ERP solution.This partially supports proposition 2. It is interesting to note that price (affordability) received one of the lowest mean ratings (7.99), appearing thirteenth out of fourteen ERP selection criteria.One reason for this could be the fact that other criteria closely related to price were rated high.Reliability was a key driver of cost of ownership; this criterion received the highest rating overall, and could be instrumental in leading customers to ERP solutions that have lower ownership costs.Compatibility was a driver of acquisition cost.High levels of compatibility with other software systems enable customers to engineer acquisition costs by selecting ERP modules from a range of different vendors.Compatibility also enables customers to select a mix of technologies that are compatible with internal skill sets.Another factor that might explain the reason for the low rating of affordability could be that, having observed the high rate of failed ERP projects, customers focus more on successful ERP adoption and less on inexpensive ERP adoption.It would be interesting to investigate whether customers are willing to pay a higher acquisition price if they perceive a greater chance of successful ERP adoption.
Further application of the Mann-Whitney U Test revealed that there were significant differences between the retail and non-retail sub-segments within the nonmanufacturing segment, thus supporting proposition 3.For manufacturing companies the top half of the ERP selection criteria list was dominated by criteria that provide guidance on the likelihood of successful implementation.These were factors that focused largely on the current state of the solution, such as how reliable it was, the quality of support available, and the extent to which it could be customised.Out of the top seven ERP selection criteria, only one (vision of the vendor) was focused on the future state of the ERP solution.Non-manufacturing firms appeared to place a stronger emphasis on the future (rather than current) state of the ERP offering.Three out of the top seven ERP selection criteria were focused on matters that relate to future plans for the ERP solution, the extent to which the vendor understands customer's industry, and the extent to which they were able to prove this.The fact that both manufacturing and non-manufacturing firms applied the same selection criteria for ERP selection suggests that they have a common scope of concerns.The fact that the ERP selection criteria were rated differently by manufacturing and non-manufacturing firms suggests that the two market segments emphasise different concerns when selecting ERP solutions.Manufacturing firms appeared to be predominantly concerned with ensuring successful ERP adoption, while non-manufacturing firms appeared to give equal weighting to ensuring a successful ERP adoption and enhancing that with a steady flow of value-adding features in the future.One reason for this difference between manufacturing and non-manufacturing firms could be the fact the ERP systems were initially developed for manufacturing firms, and were thus richer in manufacturing features than nonmanufacturing features.This state of affairs could also suggest that medium-sized non-manufacturing companies that have recently adopted ERP were early adopters, while their manufacturing peers were part of the later majority of all manufacturing companies.If this were true, this could also mean that the recent growth in the middle market was actually the intersection of two distinct markets that were evolving in different ways -that is, the manufacturing market captures smaller companies as it matures, while the non-manufacturing market captures larger companies as it matures.The illusion of a single expanding middle market could simply be the effect of both markets passing through a phase where they were serving similar sized companies.
Given the low number of respondents who proposed new selection criteria, it was concluded that respondents largely felt that the current set of selection criteria would continue to remain relevant in the foreseeable future.This may also suggest that respondents were largely ignorant or unconcerned about developments that may impact future selection criteria.This could be further explained by the fact that 70% of respondents had already implemented their ERP.

CONCLUSION
In 2006 The Economist reported that SAP wants to expand its customer base from 35,000 to 100,000 customers by "moving downstream".The same article referred to the 19.5 billion USD that Oracle has spent in pursuit of the same objective.ERP vendors have traditionally served a relatively small number of clients who are willing to pay a higher price for highly specialised software products and services.As ERP vendors expand into the middle market, there will be a greater number of customers who are willing and able to pay lower prices.ERP vendors clearly hope to sustain growth by building volume, but this will also mean that the cost of delivery will have to be reduced significantly.
Manufacturing and non-manufacturing firms emphasise different priorities when applying the ERP selection criteria.Although all fourteen ERP selection criteria were important to medium-sized businesses in South Africa, manufacturing and nonmanufacturing firms agreed exactly on the first three and differed in their rating of eight of the fourteen ERP selection criteria.It was clear that manufacturing and non-manufacturing companies emphasised different themes in the selection criteria.Non-manufacturing firms appeared to place a greater emphasis on the realisation of potential future gains from ERP investment.For these companies, the vision of the vendor, industry knowledge, and vendor references were the next three important selection criteria after reliability, support, and compatibility.On the other hand, manufacturing firms appeared to emphasise the criteria that determined how ERP can be made to work in their specific situations.Flexibility, infrastructure considerations, and organisational fit were the next three important selection criteria after reliability, support, and compatibility.Non-manufacturing firms appeared to a hold the view that ERP still had a lot to offer them in the form of new features and capabilities, while manufacturing firms appeared to be more focused on accessing the gains that have been realised by their larger counterparts.