Improving the Employee Portal at Mercedes-Benz USA
K.J. Smilde
University of Groningen, the Netherlands Faculty of Management and Organization May 2004
Author: Kris Smilde Student Number: 1064673
University: Rijksuniversiteit Groningen Faculty: Management and organization
Major: Industrial engineering and management science
1st Supervisor: Prof. Dr. G.B. Huitema 2nd Supervisor: Dr. A. Boonstra
Supervisors Mercedes: A. Branch S. Huang
Place: Montvale, New Jersey, USA
Groningen, The Netherlands
Date: 05-10-2004
Preface
For some people writing a thesis an easy job, but not for me. It took me a lot of time and energy to make what you see now. A preface is a good medium to thank all those people that helped me achieving this result. I am indebted to many, because it took them much time and energy as well.
I had the luck to come to Montvale and have an internship at Mercedes-Benz USA. It was a very good experience; I got a lot of responsibilities, drove a 1983 Ford Crown Victoria Ltd. and made lots of friends. It was great to work in the Web Operations Center with such great colleagues. In the first place I would like to thank Rudi Arts, for giving me the opportunity to come to MBUSA. Also my sincere thanks to Simon Huang and Alan Branch, my supervisors, they were a great help even after I had left the US. Luke and Brijesh were the best cubicle-mates imaginable and I really enjoyed working together with Vincent, Chris, Diane, Jim, Patrick, Naru, Kaushik, Lark, Bob, Olivia, Inigo, Lisa, Shelia, Somya, Liz, Kobe, Sven, Glenn and of course Louis.
A few months after my return to Groningen, I changed the research objective and almost my entire research. The process of writing a new version was hard; the next one after that was still another Mount Everest, like the successive versions. I am grateful to a number of people for their support during this time. First of all I thank Prof. Dr. G.B. Huitema for his supervision, time and patience. I am aware of the fact, that you have probably supervised easier students. Dr. A. Boonstra was a great assistance with his feedback.
My parents, Els, my sisters and my friends gave me a lot of energy and support when needed. The person I would like to thank most is Leonie, for her visiting me in the US, reading my thesis, giving me fruit baskets and all the other things you have done for me!
Kris Smilde
Groningen, May 2004
Abstract
At the end of 2002 an Employee Portal was introduced to Mercedes-Benz USA (MBUSA). The launch version of the Employee Portal is a big collection of links to parts of the Internet and the intranet placed on 5 Tabs. Because of the overwhelming amount of information employees have trouble to find what they need to find. There is also a lot of outdated content and a very bad search function. This research is a search for a new version of the Employee Portal in order to improve the information provisioning.
The objective of this research is to make a functional design of an Employee Portal in order to improve the efficiency of the information provisioning for the employees of Mercedes-Benz USA.
The objective is presented in the following main question:
What should be implemented into the Employee Portal to improve the efficiency of the information provisioning between employees at Mercedes-Benz USA?
To answer this question the current information provisioning situation is analyzed together with the employees’ experience.
Business System Planning shows a couple of things about the current information provisioning situation. First of all, using e-mail, paper and telephone are more expensive ways to provide information than using for instance the intranet and the Internet. Secondly, the document library, a Lotus Notes database is used together with PeopleSoft and the intranet for the same kind of information. Employees may have to search in two systems to find information. And thirdly, most departments are involved in projects using only the expensive information provisioning systems to keep the project information.
The users’ experience learn that it is hard to find the right information in a timely manner, because the Employee Portal is too overwhelming, the search function does not work and a lot of the content is outdated. Employees also spend a lot of time using alternative ways to find the information.
The best additions to the Employee Portal, in order to improve the efficiency of the information provisioning are:
♦ search function with taxonomy;
♦ personalization;
♦ content management tool with a publishing function;
♦ project management tool;
♦ document library integration.
There were many complaints about the search function; that is why a better search function has to be implemented. Taxonomies may also be of great value here, good taxonomies can produce more specific results per employee.
The problem of too much information can be solved by personalization. It will make the Employee Portal less overwhelming, less cluttered. It will be easier for the employees to locate the information required, because the Employee Portal will only show those Portlets they need to see. Taxonomies will help to achieve this personalization. Bad navigation was also one of the drawbacks of the launch version of the Employee Portal. Employees sent suggestions for a better navigation, because there was too much information on each Tab of the Employee Portal. The current navigation will work well if the information on each Tab becomes less. Personalization helps reducing the content on each Tab.
The content management system will be able to detect outdated content. The accuracy of the information will improve and error costs will reduce accordingly. The publishing function of the content management tool helps to get more consistency in the appearance of the intranet sites, thus improving the usability. The publishing function will also be time-saving for the intranet site developers.
Since a lot of employees are involved in projects, a project management tool can help to store the information from all projects in one place. Everybody involved will be able to find the most accurate information and deadlines, without having to search much through e-mails or other locations.
The document library stores information from the same information classes as the intranet. By integrating the document library with the Employee Portal, employees will only have to search one application for the information. The search functions can also be integrated.
Table of Content
PREFACE ...4
ABSTRACT...5
CHAPTER 1 INTRODUCTION...9
1.1 Mercedes-Benz USA ... 9
1.2 The end of the InfoBahn... 10
1.3 The launch of the Employee Portal ... 11
1.4 The Employee Portal during my Internship ... 13
1.5 Drawbacks of the launch version of the Employee Portal ... 15
CHAPTER 2 RESEARCH METHODOLOGY...17
2.1 Problem definition ... 17
2.2 Objective and main question ... 20
2.3 Information sources and measurements ... 23
2.4 Approach of this research ... 24
CHAPTER 3 INFORMATION PROVISIONING AT MBUSA...27
3.1 Processes Approach ... 28
3.2 Object-oriented Approach... 29
3.3 Business System Planning ... 31
3.4 Selection of suitable modeling method ... 32
3.5 Information provisioning analysis... 35
CHAPTER 4 THE LAUNCH VERSION OF THE EMPLOYEE PORTAL ...47
4.1 Employee Portal ... 47
4.2 The Employee Portal experiences ... 50
CHAPTER 5 EMPLOYEE PORTALS ...54
5.1 Introduction ... 54
5.2 Employee Portal building blocks... 56
5.3 Efficiency improvement... 60
CHAPTER 6 CONCLUSION ...65
ACRONYMS ...69
LITERATURE...71
APPENDICES ...76
Chapter 1 Introduction
In this information and communication revolution, knowledge becomes more and more important (Drucker, 1999). And because there is so much to know and information changes so fast, organizations are working more on concepts to find and to deliver information. Finding information is not always easy for employees in larger organizations. Research of Pitney Bowes (1998) has pointed out that 71%
of the workers say their main job is tracking down information, while 80% of the information that is being filed, is never used. Wurman (1990) even states:
“The amount of information available today is doubling every five years, and corporations are able to keep up with, and use, less than 7% of the data they produce.”
The new Information Revolution is about concepts, not data. Going forward, IT will focus less on the T (Technology) and more on the I (Information). It is a question of quality, not quantity (Drucker, 1999). There is clearly a need for companies to organize their information and reduce the time that employees need to find the information. When employees would spend less time on tracking down information they would save more time for other things. A better organization of the information might result in easier finding information. Mercedes-Benz USA (MBUSA) is aware of this fact and is currently changing the way the information provisioning is organized, in particular the intranet.
This research is carried out for Mercedes-Benz USA. In the next section a short introduction about MBUSA is given; after that the intranet situation before the change of process and the current situation are covered. The need of this research is discussed in the last two sections of this chapter.
1.1 Mercedes-Benz USA
This research is conducted for Mercedes-Benz USA. MBUSA has about 1200 employees in 128 departments, and is responsible for the sales, service and marketing of Mercedes-Benz products in the United States. MBUSA is not responsible for the production of cars; these are made in Germany.
"Knowledge is of two kinds. We know a subject ourselves, or we know where we can find information on it."
Samuel Johnson (1709 - 1784)
MBUSA is a wholly owned subsidiary of DaimlerChrysler and part of the business unit Mercedes Passengers Cars. The vehicles are actually sold through one of the 313 Mercedes-Benz dealers, which are not exclusively owned by MBUSA. These dealers do not have access to the intranet of MBUSA.
More information about the history and the organization of DaimlerChrysler and MBUSA can be found in appendix A.
1.2 The end of the InfoBahn
Up to the end of 2002 MBUSA used to have the InfoBahn as the intranet (Figure 1.1). This InfoBahn was available for every employee. Information provided by the InfoBahn comprised information about the company, projects, departments and other work related content. Each department had its own home page. Despite the huge amount of information, employees had trouble in finding the information they needed. This was a result of several problems.
Figure 1.1 The InfoBahn
• First of all, the difficulty in finding information was caused by the structure of the InfoBahn.
The department that created the content placed it on its own departmental site. That is why employees had to know which department created the content they were looking for. This was not always clear, since MBUSA has 128 departments and sub-departments with each a website.
• Secondly, there was the problem of the poor design of the websites. Per department there was always one content provider present, who maintained the departmental website. In most departments, the person responsible for the site only had one introductory course about designing web sites and using HTML and PageMaker. Another reason for the poor design was the lack of a design standard, available at MBUSA. Every content provider designed websites in another way. This also contributed to the fact that employees were facing difficulties in finding information they needed.
• Thirdly, employees were facing the problem that the content of the information was often outdated. Most sites were updated irregularly. When the content provider left the department, it was hard to find a person in the department to maintain the department site. The new content provider sometimes did not even update the site at all.
1.3 The launch of the Employee Portal
To deal with the previous problems in finding information, Mercedes-Benz USA decided to implement an Employee Portal. This is a new intranet technology, of which a launch version was implemented on December 10, 2002. MBUSA launched an Employee Portal, replacing the InfoBahn, as one of the changes in the information provisioning situation.
Employee Portal is the term widely used in business e.g. in companies like MBUSA, Plumtree, Forrester and Gartner. Sometimes the terms Corporate Portal, Enterprise Portal or Enterprise Information Portal (EIP) are used. In this research all of these terms have the same meaning.
An Employee Portal is a merger of software applications that combines strengths. An Employee Portal can manage the information, analyze and distribute it all in one. The distribution of the information occurs via the intranet and the information is shown via an Internet browser. An Employee Portal is much more dynamic than a flat HTML web page. An Employee Portal may appear different when viewed at various intervals or by other people at the same time.
In this research the Employee Portal is defined as:
“An amalgamation of software applications that consolidate, manage, analyze and distribute information across the organization (Dias, 2001). An Employee Portal is a customized, personalized, ever changing mix of news, resources, applications and e-commerce options that becomes the desktop destination for everyone in the organization and the primary vehicle by which people do their work (MBUSA B2E strategy 2003).”
The launch version of the Employee Portal at MBUSA (Figure 1.2) is a direct translation from the old InfoBahn. The departmental structure of the organization of the content is replaced by five Tabs that contain a huge collection of links to all the content on the intranet. The problems of the InfoBahn are partly solved.
♦ The first problem, that employees had to know which department created the content in order to find the information, is solved by this new organization. With the Employee Portal, employees only need to know which Tab holds the content.
♦ The second problem with the InfoBahn was the poor design of the departmental sites. This problem is dealt with in two ways. First, a web style standard is set. New departmental sites are still separate sites, but all are made in the same template. The menu structure and the layout are the same for all sites. Secondly, a lot of content is directly linked to the Employee Portal. By skipping the departmental sites, the users only see the Employee Portal with its standard layout and not the poor designed departmental sites.
♦ Regarding the outdated content, the new structure of the Employee Portal has improved the quality of the content. Due to the new structure, information is much more visible and employees will detect outdated content much earlier, so that the content can be updated or removed earlier.
Figure 1.2 The Employee Portal
Despite the improvements made by replacing the InfoBahn with the Employee Portal, there appeared to be some drawbacks of the new situation. These will become clear in the next section, where my experiences at MBUSA are discussed.
1.4 The Employee Portal during my Internship
I did my internship at the Web Operations Center (WOC) at MBUSA. This is a sub-department of the Information Technology department, which is responsible for the Business to Employee (B2E) strategy. B2E represents the flow of information and transactions between a company, its employees, their families and outside organizations over electronic channels. (MBUSA B2E strategy 2003) The B2E strategy at Mercedes has the following mission:
"To make our employees more productive today and in the future through facilitating and improving online processes that empower them to build and maintain internal relationships, to grow their knowledge and skills.”
Facilitating and improving online processes for the employees is the key task of the WOC. WOC’s main responsibility is maintaining the intranet, which includes the implementation of the Employee Portal.
During my internship, I had several tasks regarding the Employee Portal. Among other things, I was responsible for: handling the feedback about the Employee Portal, helping content providers maintaining and redesigning their departmental web sites and giving training to the employees of the Customer Assistance Center. Through these tasks I received a lot of feedback and experiences of the employees with the Employee Portal.
I received the feedback from the users of the Employee Portal by e-mail and via the Employee Portal feedback form. This form pops up when users close the Employee Portal. The Customer Assistance Center (CAC) delivered the biggest amount of feedback and questions. Customers can call the CAC from their car any time when they have questions or emergencies. Because of CAC’s big amount of feedback and questions and their direct contact with the customers, the WOC decided to give training as how to use the Employee Portal and how to find information. I made a guide for these training sessions that can be found in appendix B. Beside these training sessions the WOC also took care of a
“lunch and learn”, a presentation during lunch, and a stand near the cafeteria, where people could ask questions (figure 1.3). During the lunch and learn and the trainings, I got a clear view of the experiences of the employees with the Employee Portal and I found some drawbacks of the Employee Portal, which will be discussed in the next section.
Figure 1.3 The Employee Portal stand near the cafeteria
1.5 Drawbacks of the launch version of the Employee Portal
Although the Employee Portal partly solves some of the problems of the old InfoBahn, the employees are still having a lot of troubles in finding the information they need. The following drawbacks were found:
• too much information;
• bad navigation;
• search function does not work;
• still a lot of outdated content;
• too difficult to build and maintain sites.
At first people complained about the overwhelming amount of information on the Portal. An employee, working at the Customer Assistance Center even stated that she was not using 80% of the Portal. The Employee Portal also contains more links than the InfoBahn. This is due to the fact that a lot of links point directly to the documents, skipping the departmental sites.
The appearance of the Employee Portal was extra overwhelming, because MBUSA choose to use one- level-navigation, all the links were divided over only five Tabs. I received a lot of suggestions for two- level-navigation; for instance: with an additional left side menu or sub-Tabs. Because there are so many links on each Tab, it has become harder for the employees to find fast what they need.
A search function was integrated to the Employee Portal, but does not quite work as it should. Because employees have a hard time finding information, they try to use the search function a lot. The search function does not work well for several reasons. For a start it only uses the metadata of the intranet documents and most documents do not have any metadata. The metadata of a document contains information about the document, but it does not influence what you see when you open the document.
Secondly, most intranet sites are built with different frames; every frame contains one document. As most index pages consist of different documents, it is hard to find full pages.
The last two problems, the outdated content and the problem of the content providers to maintain their site can be mentioned together. As the departmental sites are still existing and the way of posting content have not changed, the content providers still have a lot of troubles posting information and maintaining their site. That is why the problem of outdated content has not been solved so far.
This research will focus on how an Employee Portal can help resolving these problems. The objective and the main question, which will be dealt with in the next chapter, will cover these drawbacks.
Chapter 2 Research Methodology
This chapter contains the research methodology. This chapter begins with the problem definition.
Then the objective, main question and sub-questions will be discussed. In the third section the approach of this research will be dealt with. The last two sections of this chapter contain the information about the measurements of this research and the information sources.
2.1 Problem definition
As discussed in the previous chapter, the drawbacks of the Employee Portal are the main reason for this research. MBUSA wants to improve the Employee Portal, this is rather vague. The business-to- employee (B2E) mission of the Web Operations Center, the experiences of the employees with the Employee Portal, and a literature review are taken into account in specifying the objective.
The B2E mission and the experiences of the users
The search for the objective of this research starts with the B2E mission of the WOC and the experiences of the employees with the Employee Portal. This research has been initiated because of the experiences of the employees and it has to align with the B2E mission of the WOC. The B2E mission is:
"To make our employees more productive today and in the future through facilitating and improving online processes that empower them to build and maintain internal relationships, to grow their knowledge and skills."
As stated in the B2E mission statement, the WOC has to improve the productivity of the employees through facilitating and improving online processes. Productivity is defined as the rate at which goods or services are produced, especially output per unit of labor (Webster’s dictionary). The Employee Portal takes care of the process of information provisioning for employees. The Employee Portal has to improve the productivity according to this mission statement.
The experiences of the users show that there are opportunities to improve the way in which information is provided. Using the Employee Portal, employees have trouble in finding the information they need swiftly. This is time consuming, because of the overwhelming amount of
information, the bad navigation, the inadequate search function and the big amount of outdated content.
Using productivity as the measurement of success of the Employee Portal logical for this research. If the employees spend less time on their search for information, they will have more time left to produce products and services, with the same amount of labor units. The productivity will increase. However the difficulty with this objective is, that the productivity of the employees is influenced by many factors besides the Employee Portal. In fact, it has never been shown that investments in IT have improved the productivity, as will become clear in the next section.
Productivity: not a good measurement of success
Productivity is defined as “the rate at which goods or services are produced, especially output per unit of labor” The influence of Information Technology on the productivity of employees is not unequivocal. While some studies have found positive association between IT spending and employee productivity, or firm performance, most find weak or even negative associations (Barua, 1995; Hitt, 1996; Strassman 1990). This is called the information technology productivity paradox. The introduction of an Employee Portal does not necessarily influence the productivity of employees as is assumed by the WOC.
There are several interpretations about how this paradox can be explained. One explanation is that managers have a hard time bringing the benefits to the bottom line, especially if investment in IT is not accompanied by changes in business processes (Dehning, 2003). The result could be for MBUSA that investments in IT lead to increased organizational slack rather than productivity and improved financial performance. The process of information provisioning does not only determine the productivity at MBUSA. Other business processes also influence the productivity of the Employees.
Another explanation is that managing IT projects is inherently difficult, and a high percentage of IT projects encounter significant problems such as cost overruns, coordination problems, or failure during implementation (Shao, 2002). Sometimes poor usability is blamed for the failure of the system (Mulder, 1992). In short, IT projects are often overestimated in performance and underestimated in costs.
A third explanation is that competitors make the same IT investments. They will not experience a significant performance advantage relative to their competitors (Shao, 2002; Dehning, 2003). This explanation implies, that not making the IT investments will lead to a harmful effect on the
competitive position. Often productivity is seen as a function of (technical) efficiency and technical change. One of these theories is the Malmquist productivity index (Malmquist, 1953).
In order to make the productivity better by improving the Employee Portal is not a fair objective for this research, because there is no proven correlation between investing in an information system and an upgrading of the productivity. Therefore other measurements for the success of the Employee Portal were taken into account. These will be dealt with in the next section.
Alternative measurements of success
Because productivity cannot be used in the objective of this research, alternative measurements have to be found. Alternative measurements have to align with the B2E mission and with the feedback from the employees. There are many theories about what will influence the productivity of employees.
Efficiency (Taylor, 1911; Paul, 1996; Lin, 2000; Grosskopf, 1993), effectiveness (Sloan, 1963; Paul, 1996), adaptability (Harrington, 1991) and satisfaction (van der Zwaan, 1999; Haselhoff, 1987) are often regarded as factors influencing the productivity. All four will be discussed below.
Adaptability is the flexibility of the process to handle future changing customer expectations and today’s individual, special customers` requests (Harrington, 1991). Adaptability and flexibility have a lot to do with the structure of the organization and business and production processes (Jagers, 1999) If the Employee Portal succeeds in making the organization, or processes more flexible, the information gathering will be just as hard. This might improve the productivity but it does not deal with the drawbacks from the Employee Portal, as given by the employees.
The Employee Portal will probably not effect the employee satisfaction much. The organization of labor does have a big influence on the employee satisfaction (van der Zwaan, 1999). Paul (1996) points things out like, working conditions, labor relations and conditions of employment. Employees might find it pleasant when they can find information faster, but there is no study proving that might influence the satisfaction much.
Efficiency and effectiveness are often used together with productivity. All three concepts look at the input and at the output of a process, or an organization. Efficiency and effectiveness both align with the B2E mission statement of the WOC. To be able to determine what to use in this research, efficiency, effectiveness or both, it is necessary to explain the difference between efficiency and effectiveness.
The definitions of efficiency and effectiveness are (Paul, 1996):
Effectiveness is the extent to which the output of the process meets the goals.
Efficiency is the extent to which the output of the process meets the goals by the use of as little resources as possible.
Effectiveness has to do with the outcome of the process, while efficiency mainly focuses on the process and the input. Effectiveness is doing the right things, whereas efficiency is doing the things right.
Efficiency: the measurement of success
The information provisioning is effective when the employees get the information they need.
According to the feedback and the experiences of the employees, that is not the problem. The employees can find what they need, but sometimes it is much time consuming. Efficiency is the extent to which the output of the process meets the goals by the use of as little resources as possible.
According to the employees, the resource time is used too much in the process of information provisioning. This might be reduced by a redesign of the Employee Portal.
(Technical) efficiency is often used as a factor of the productivity of a firm together with technical change (Grosskopf, 1993; Färe, 1994; Lin, 2000). One of these theories is the Malmquist productivity index (Malmquist, 1953). Because technology will change, firms have to become more efficient to avoid a downfall in productivity according to this theory. Research shows that the efficiency often increases, whereas the productivity might not increase (Grosskopf, 1993; Färe, 1994; Lin, 2000; Shoa, 2002).
Becchetti (2003) shows that IT investments, especially in telecommunications (such as an intranet), improve the efficiency of a firm. The Employee Portal has not improved the efficiency much yet.
Therefore efficiency of the information provisioning will be the measurement to examine in this research.
2.2 Objective and main question
In this research the Employee Portal will be redesigned in such a way that the efficiency of the information provisioning will increase. According to Employee Portal suppliers (Plumtree, IBM, Brio,
Viador, Intraspect), there are many things that can be implemented separately. In this research these additions to the Employee Portal will be called Employee Portal building blocks. This research will look for the building blocks that improve the efficiency of the information provisioning.
Objective:
The objective of this research is to make a functional design of an Employee Portal in order to improve the efficiency of the information provisioning for the employees of Mercedes-Benz USA.
Main question:
What should be implemented into the Employee Portal to improve the efficiency of the information provisioning between Employees at Mercedes-Benz USA?
Answering sub-questions will help to answer the main question. The following sub-questions are chosen for this research:
1. What is the current information provisioning situation at MBUSA?
2. What features does the launch version of the Employee Portal have?
3. What are the possible building blocks of an Employee Portal?
4. Which parts of an Employee Portal help to improve the efficiency of the information provisioning at MBUSA?
To determine how an Employee Portal can make the information provisioning more efficient, it will be necessary to find out how the information provision has been arranged in the current situation. It is also necessary to discover; what is the information needed to be provided to the employees, which systems do so, and how. This needs to be examined because the next version of the Employee Portal will be implemented in the current information provisioning situation. New Employee Portal building blocks may influence the information provisioning situation and make the information provisioning more efficient.
The launch version of the Employee Portal needs to be looked at more closely to be able to determine a desired design. What functions have already been implemented and where the problem areas are.
The feedback from the employees will be very useful here. This part helps to discover how the launch version of the Employee Portal can be made more efficient.
Of course, it is also necessary to know what an Employee Portal may consist of; which functionalities it may meet. Research in literature and information from Employee Portal suppliers is needed to answer this sub-question. The most important Employee Portal building blocks require to be examined.
The final step is to find out what part of an Employee Portal will help to improve the efficiency of the information provisioning at MBUSA. This will be assessed with the help of the current information provisioning situation and the feedback from the employees. The building blocks will be selected on how they improve the efficiency of the current information provisioning situation and how they improve the efficiency of the Employee Portal.
The situation of this research can also be shown as a conceptual model (Figure 2.1). This model is a global view of this research and will serve as a guide in this research.
As shown in figure 2.1 the employees of MBUSA obtain their information by means of the information provisioning systems. The information provisioning systems can be the Employee Portal, E-mail but also the telephone, paper and conferences. The information provisioning systems and the information available at MBUSA determine the information provisioning situation at MBUSA. The employees provide feedback about the Employee Portal. This feedback will help, together with the information provisioning situation, to select those Employee Portal building blocks, which improve the efficiency of the information provisioning. How this can be done is explained in the next section.
Employees of MBUSA Information
provisioning systems
Information Information provisioning situation at MBUSA
Feedback about the Employee Portal
Employee Portal building block selection
Efficiency of the information provisioning
Figure 2.1 The conceptual model of this research - improving the efficiency of the information provisioning
+
2.3 Information sources and measurements
To be able to select Employee Portal building blocks, the efficiency of the information provisioning has to be measured per building block. Maximizing the efficiency of the information provisioning equals minimizing the amount of resources, needed for the employees to find information. Time is the most important resource, as can be concluded from the experiences and the feedback from the employees. Money is also an important resource. New projects at Mercedes-Benz USA are only implemented when the savings or the profits of the project exceed the investments. At MBUSA time is money. That means that an Employee Portal building block can be implemented when the estimated time saved has a bigger value than the money invested.
Some researchers use a very hard definition of efficiency (Shoa, 2002; Grosskopf, 1993). To emphasize that they use a hard definition of efficiency, the term technical efficiency is often used often. Technical efficiency either focuses on the output side or the input side of a production process.
An indicator of technical efficiency can thus be actual output versus expected output (given some input amounts) or resources actually consumed versus resources expected to be consumed (for producing a certain level of output) (Shoa, 2002). At MBUSA it is hard to determine what the actual input and output will be in the new situation. The actual improvement of efficiency can only be measured when the new version of the Employee Portal is implemented. Then gaining in time can be measured as well as the actual costs.
In this research a soft definition of efficiency will be used. By looking at the information provisioning at MBUSA, the feedback from the employees and the information from the Employee Portal suppliers, it will be estimated accordingly whether the building blocks increase or decrease the efficiency of the information provisioning. The conclusions will be drawn using quantitative, as well as, qualitative information.
In order to obtain all the information needed for this research the following methods are used:
• scientific literature and information from Employee Portal suppliers;
• information on MBUSA’s intranet;
• interviews with employees of MBUSA;
• feedback send by the users of MBUSA’s Employee Portal.
First of all scientific literature is used to find information about efficiency, information provisioning and Employee Portals. Websites from Employee Portal suppliers contributed to get more knowledge
about the Employee Portal building blocks. A lot of information about the Employee Portal building blocks was given at the ‘No Empty Portals’ workshop from Plumtree, one of the biggest Employee Portal suppliers. The intranet of Mercedes-Benz USA also provides much information. Especially information about the organization, the processes and the strategy is found here. The intranet really helps to understand the organization and its processes. By interviewing employees of MBUSA in a structural way, through face-to-face interviews and a questionnaire, information was obtained about their feelings toward the Employee Portal. Much feedback was received in unstructured interviews. On the Employee Portal is a “send feedback” function available. A lot of Employees use that function to answer questions and provide feedback. This feedback is put in a database and analyzed.
2.4 Approach of this research
This research is done in different parts, as mentioned below:
• research introduction;
• the information provisioning at MBUSA;
• the launch version of the Employee Portal of Mercedes-Benz USA;
• the Employee Portal building block description and selection;
• the conclusions.
The research introduction part covers chapter 1 and 2. Chapter 1 is the problem exploration and chapter 2 deals with the research methodology. The chapter about the information provisioning at MBUSA is chapter 3. The first sub-question is answered here. The current information provisioning situation is described and analyzed. The launch version of the Employee Portal at MBUSA is dealt with in chapter 4. The functionalities of the launch version of the Employee Portal are described and the feedback from the employees is explained in further detail. This part deals with sub-question 2.
Chapter 5 contains descriptions of Employee Portal building blocks and a selection of those building blocks. Literature and information from Employee Portal suppliers will be used to find the Employee Portal building blocks. The determination whether the Employee Portal building blocks will improve the efficiency of the information provisioning is done with the help of the previous two chapters. The conclusion will be drawn in chapter 6. In this chapter the answer to the main question is given. This approach can be represented by figure 2.2.
Research introduction
Chapter 1 and 2
Information provisioning at MBUSA
Chapter 3
Employee Portal building block description and selection
Chapter 5
Launch version of the Employee Portal
Chapter 4
Conclusion
Chapter 6 Figure 2.2 The approach for this research
Chapter 3 Information provisioning at MBUSA
In this chapter the information provisioning at MBUSA in the current situation will be described and analyzed. The first sub-question is dealt with.
1. What is the current information provisioning situation at MBUSA?
In answering this sub-question, a description of the current information provisioning situation is needed. It is necessary to answer the following questions:
• What kind of information is provided to the employees?
• Which information provisioning systems are used to provide the information?
• How are the several information provisioning systems used to provide the information?
It is necessary to find out what kind of information the employees need, because the next version of the Employee Portal may help providing that information. Whether the Employee Portal has to supply this information depends on it has been provided up to now and which information provisioning systems were used?
There are several ways to approach and analyze the current situation. Due to the fact that MBUSA is such a big company, it will be useful to first model the current information provisioning situation, before analyzing it. In the first three sections of this chapter, three of these methods will be discussed.
These are:
• the processes approach;
• the object-oriented approach;
• Business System Planning.
Section 3.4 contains a discussion about the best suitable method. In section 3.5 the best suitable method will be used and will help answering the first sub-question. This section will show the conclusions that can be drawn regarding the efficiency of the information provisioning in the current information provisioning situation.
3.1 Processes Approach
A lot of modeling methods show the information provisioning in processes. A process is drawn modeling information flows, actors, activities and decision points. An example of such a method is the Eriksson Penker Business Extension (EPBE) developed by Eriksson and Penker (2000). The process is presented as a big arrow with one or more triggers, input, a goal and an output. In the arrow the process is shown with sub processes, decision points and information and resource flows. Figure 3.1 shows an example layout.
This method clearly shows which information is needed by whom and trough which system. It also makes a clear delineation between relevant and irrelevant information, resources and actions.
Figure 3.1 Standard EPBE layout
The event, or trigger for the process of redesigning a departmental website, could be an e-mail from the content provider. The goal of the process is to make a better design of the website. The output is the redesigned website. The process of redesigning a website can be described in detail. Not only the information and resources required can be mentioned, but also the decision points, which can be represented by a rhomb. Example decision points are:
- Does the content have to change?
- Does the design follow the standard?
- Is the site user-friendly?
Another way to model a process is by using a sequence diagram (de Boer, 2001). A sequence diagram shows a process on time lines. On top of a sequence diagram the objects that play a part in the process will be found. The objects can be persons, information systems or other companies. Under each object
there is vertically a thick line representing time. There are horizontally thin arrows between these thick lines. They represent the messages send by one object to another one at a certain moment in time.
Because the information provisioning systems can be taken as objects it gives a good view of all the information used by the Employees and where they get it from. An example is shown below (figure 3.2). Shown is the process of planning a one-on-one meeting. Lotus Notes is the program with the calendar of both persons. Not only the actors and the actions taken in this process are clearly shown, the exact order is also very clear.
Figure 3.2 A sequence diagram for planning a meeting
3.2 Object-oriented Approach
Object-oriented modeling is another popular method of modeling. The meaning of the object-oriented approach, during the analyze- and design-phase of the project, is to draw a picture of the reality as truthful as possible (Boullart, 1991). This picture will be drawn with objects, attributes, classes and relations.
An object is an independent system component with certain characteristics. These characteristics are called attributes. One person at MBUSA would be one object with, among others, the attributes name, job title and age. Objects with the same value for at least one of their attributes are considered a class.
Person 1 Person 2 Lotus Notes
Checks calendar for possible dates and times
Shows possible dates and times Gives possible dates and times
Checks calendar for possible dates and times
Shows possible dates and times Accepts one date and time
Puts meeting in calendar Puts meeting in calendar
The class “MBUSA Employees” contains all the people (objects), whose employer (attribute) is MBUSA (value of the attribute). An object can be part of a class, but this class can also be part of a bigger class. The class “MBUSA employees” is part of the class “working people”. In most object- oriented analysis the classes are described and not the individual objects (Florijn et al., 1996). For MBUSA the employees as well as the information and the information provisioning systems can be put into classes. This is a great advantage to get a better overview. Processes are not modeled;
processes can be objects in this method or can even be left out.
To make an object-oriented model it is not only necessary to know what objects and classes there are, but you also need to know what the relations are between the objects and classes. Employees create and use information and they use information provisioning systems. There are a lot object-oriented modeling languages. Object Modeling Technique (OMT) of Rumbaugh, Object Oriented Analysis and Object Oriented Design (OOA and OOD) of Coad and Yourdon and Object Oriented System Analysis (OOSA) of Shlaer and Mellor are examples (van den Goor et al. 1993). The relations are shown in most object-oriented modeling languages through lines or arrows between the objects. On the lines you can read what the relationship is between the objects. Often is shown if the relation is one to one, one to many or many to many. A small example is shown in figure 3.3. It is written in Unified Modeling Language (UML) (Rumbaugh, 1999). The Arrows above show that one class is a subclass of the other class. Man is a sub class of Person and Car is a sub class of Vehicle.
Figure 3.3 An object oriented example (Diskin, 2003)
Making classes from the objects reduces the complexity of the model dramatically, but the relations might still make it very complex. Especially, since most classes have several relations with other classes.
3.3 Business System Planning
IBM developed Business System Planning (BSP) in the 1960’s explicitly for establishing the relationship among large system development projects (Zachman, 1982). BSP is an analysis of the organization wide information requirements by looking at the entire organization in terms of organizational units, functions, processes, and data elements. It helps to identify the key entities (objects) and attributes in the organization’s data (Laudon, 1998). BSP is a very structural approach to deal within a very complex information and communication situation. It links data with departments, processes and with (information) systems and everything is well documented. The information, the employees and the processes are put into classes and in a matrix to show whether there is a relationship between these classes or not. The type of relationship can be given by using different signs in the matrix. The model that shows the connection between these classes is called the information cross (Cuppen, 2000). This information cross has four quadrants as shown in figure 3.4 (QI, QII, QII and QIV).
Quadrant I
The first quadrant shows what departments are involved in what processes. Per department the involvement is defined as major, some or none. This quadrant helps to get insight into the
Figure 3.4 BSP Information Cross
Processes
Systems Departments
Information classes Q IV
Q II Q III
Q I
S4 S3 S2 S1 D1 D2 D3 D4
P1 P2 P3 P4
I1 I2 I3 I4
organizational complexity and the importance of the processes. When the amount of departments involved in a process increases, the complexity of the information provisioning within the process will increase as well.
Quadrant II
In the second quadrant it will become clear which department generates which information. These will be defined as the creators/administrators. Administrating data means updating, deleting or modifying the information. The parts of the organization that use the information are also pointed out.
Quadrant III
This quadrant shows how the several information classes can be found. There are not only electronic information systems in this list of system, but things as telephone and conferences are also shown as systems that help to provide information. Several systems can be used to provide one data class.
Quadrant IV
The fourth quadrant shows which system supports which process. According to the BSP philosophy there should be all one-on-one relations between the processes and the systems to ensure that the processes are as efficient as possible. When people use more systems in one process, they will need to switch between systems, which might become time consuming.
3.4 Selection of suitable modeling method
All of these three methods can help answering the first sub-question. All three are able to show what the current information provisioning situation looks like. To decide which method is the best to use, several things will be considered. Each method will be examined on what information they provide about the information provisioning situation. The methods will also be examined on usability. In this context usability does not imply the ease to build the model, but the ease to use it, when analyzing the information provisioning situation. First will be dealt with the information provided by the three methods. After that, the three methods will be judged on usability.
Information provided by the methods
As stated in the beginning of this chapter, the following items need to be examined of the current information provisioning situation:
• Which information is provided to the employees?
• Which information provisioning systems are used to provide the information?
• How are the several information provisioning systems used to provide the information?
The information provided to the employees is clearly shown in the process oriented methods.
Information becomes clear as a trigger of a process, as something needed to accomplish the process, or as the output. In the object-oriented methods information can be modeled with all the attributes needed. It can show in what process the information is needed, but also where the information came from. Lots of other attributes can be added as well. Like the object-oriented approach, BSP puts the information in classes. It shows who has created and who has used the information. Putting the information in classes also implies a loss of detail, which might be a disadvantage.
Which information provisioning systems are being used is not always clear in the process oriented methods. It shows which information is needed from outside the process but it does not always show which system provides that information. It is easy to show which systems are used in the object oriented methods. The required attributes of the information provisioning systems can also be chosen.
BSP shows the same information provisioning systems, but it may show less attributes.
In the process oriented methods it is not always clear which methods are used to provide the information. For this reason it is sometimes hard to state how the several systems are used together to provide the information needed. The relations in the object oriented methods have become very important in order to show how the information provisioning systems are being used to provide the information. The relations are presented as arrows in the model. BSP shows which information is provided by which information provisioning system and it shows the information provisioning systems used in the specific processes. It is easier to understand how the information is provided by the information provisioning systems when their relations within the processes are shown.
When observing the kind of information the three methods provide for this research, the object- oriented methods are the best, closely followed by the BSP. The process oriented methods do not always show how information is provided. The modeled information provisioning situation might omit important information.
Usability of the models
The process oriented methods can provide a detailed model of the information provisioning situation with the information, the decision points and others. This is a very objective method, because most processes are put down in writing by the employees in procedures. To have a good overall view of the information provisioning at MBUSA, as a company, it is required to know most of the important processes within MBUSA and to model all the big and important processes. Since MBUSA is a big company with 128 departments and 1500 employees, it is hard to make a selection in the processes and even harder to get a good overview when all the processes are modeled.
The object-oriented approach solves the problem of big amount of processes. Processes can even be left out of the model. Another important advantage is the use of classes. Several objects can be put together into classes to get a better overview of the situation. Employees can be divided in sub- departmental classes and sub-departmental classes can be put into departmental classes. The model or models can become much smaller and much more comprehensible. This is nevertheless very subjective; the researcher will define which information is categorized as a class. By showing the important relations between the classes, the information provisioning at MBUSA can be modeled very well. In these modeling techniques the relations are shown with arrows. Because there are still a lot of classes in this situation, the information classes, as well as the information provisioning system classes, there are many arrows in this model. The model may cause confusion due to the big amount of classes and arrows. It may even become difficult to gain a clear overview of the situation.
Business System Planning is similar to an object-oriented approach. Objects can be put into classes and the relations are explained, not the processes. The disadvantage of subjectivity also marks this method. In the standard information cross four different kinds of object classes are used, the information classes, the employee classes, the information provisioning system classes and the process classes. But instead of working with arrows, BSP works with matrixes. By using matrixes the model will become more rigid, but also much better to be surveyed.
Conclusion
The values for the provided information and the usability for the three methods can be drawn in the following table (Figure 3.5). This table ranks the three modeling techniques on the provided information and the usability.
Information Provided
What information Which information provisioning systems
How is the
information provided
Usability
Process-oriented 1 3 3 3
Object-oriented 2 1 1 2
BSP 2 1 2 1
Figure 3.5 Evaluation the three modeling techniques
Figure 3.5 shows that the object-oriented approach may provide the biggest amount of information.
This is because of the attributes and the relations that can be shown. Business System Planning is the best method to use in this research, it provides all the information needed and it is much better usability. BSP provides a better overview than the object oriented methods and for this research, the information provided about the information provisioning situation is much better than the process oriented methods. Business System Planning will be used in the next section to analyze the current information provisioning situation at MBUSA.
3.5 Information provisioning analysis
Business System Planning is used to model and analyze the current information provisioning situation.
Matrixes show the relations between processes, departments, information classes and information provisioning systems at MBUSA. These matrixes are made in consultation with Simon Huang. The four quadrants of the information cross will be dealt with separately.
Q1 Processes-Departments
MBUSA has 128 departments and sub departments. To keep everything well organized, most of these sub-departments will be divided into classes. MBUSA will be split into 16 organization parts. The sub-departments are sub-classes of the main department classes. The processes are mainly derived from the responsibility of the departments. This is also done at a high aggregation level to improve the usability of the matrix. The matrix is shown in figure 3.5.
A ‘M’ is placed in those squares where the department has major involvement in a process. When there is some involvement of a department with a process a ‘S’ is placed. When a department is not, or not much involved in a process the square has been left empty.
A lot of the departments manage projects. Most of the IT department occupations are projects, but departments as Marketing and Engineering Services also work a lot on projects instead of doing routine work. It is very hard to define which processes are included in those projects in general, that is why a project is a process in itself in this scheme.
Looking at figure 3.5, it shows that many departments are involved in a lot of processes. Sometimes, the departments are working together and sometimes they work parallel to each other. With most IT projects only two departments are involved; an IT department and the department where the project runs; for instance with a redesign of a web site. The IT department has some involvement in all of the processes; that is because all the processes use at least one information system to complete it. Some processes have major involvement through the entire organization, like the Human Resources processes. There are employees (Human Resources) in all the departments.
A lot of department’s customers are other departments. The Quality Assurance Department conducts audits at other departments to improve their processes. Corporate Communications provides information from the executives or other departments, to the rest of the company.
