Maarten van den Muijsenberg ERP implementation at Heineken: improvements for information management?

(1)

ERP implementation at

Heineken:

improvements for information management?

Maarten van den Muijsenberg

Master Thesis

Technology Management

June 2009

University of Groningen

Faculty of Economics and Business

Project Company

(2)

ERP implementation at

Heineken:

improvements for information management?

Maarten van den Muijsenberg

Master Thesis

Technology Management

June 2009

Project Company

Heineken Netherlands BV

Supervisor: Mr. A.M. Werring

University of Groningen

Faculty of Economics and Business

(3)

Preface

This report describes the research I conducted at Heineken for finishing the master program of Technology Management at the university of Groningen. This research was part of an internship at Heineken, which started Jan 2009 and lasted for six months. The research was conducted in Zoeterwoude, where the Heineken headquarters of the Netherlands are situated. This research was conducted at the Horeca division of the Netherlands. The majority of the data used in this research was gathered there.

First of all I would like to thank my supervisor at Heineken Loek Werring for giving me the opportunity to conduct my research at the Horeca Division of Heineken. Also, special thanks to Rico van Burgh for providing me with this specific assignment. Furthermore, I would like to thank all the people within the HeiSale project team for helping me during the project and providing information when asked for. I would also like to thank all other employees at Heineken I visited for interviews or for discussing various end products.

Also, many thanks to Henk Sol, my supervisor at the University of Groningen who helped me develop ideas and structure my research.

(4)

Management summary

This report is the result of a research conducted at the Horeca division of Heineken Netherlands. The Horeca division is preparing for an ERP implementation at the end of 2009. Besides the ERP system, a BI solution and Every Angle are developed as well. The systems that are being developed are not fully adjusted to the information need of the Dutch Horeca organization. Because the systems that are being developed are also going to be implemented in other European countries. Also, Heineken does not know which manager needs what information.

The Horeca division wanted to know how the different tools could be used to satisfy their information need. Therefore, the following research question is answered. How

should the Horeca division of Heineken NL use the different BI tools to ensure that every manager is provided with appropriate information for the coming years?

During analysis of the current reporting situation, some problems concerning reporting came about. After analyzing the current situation, I identified if the new systems would solve any of the problems. The systems did not solve the problems identified in the current situation. There are four causes to the problems identified at Heineken:

- there is no control structure for reporting; - information need of managers is not clear; - many systems can be used to generate a report;

- employees do not always work according to work instructions.

(5)

1 Introduction Heineken HeiSale

In this chapter general information about Heineken and the Heisale project is presented.

1.1 Company information

Heineken was founded in 1864 when Gerard Adriaan Heineken acquired a small brewery in the heart of Amsterdam. Since then, four generations of the Heineken family have expanded the Heineken brand and the company throughout Europe and the rest of the world (Heineken Annual Report 2008).

Heineken has grown to a company that is present in many countries. Heineken’s most famous brand is Heineken, but the group brews and sells more than 170 different beers and ciders (Heineken.com).

In 2008, the average number of employees employed was 56,208 and the group beer volume was 125.8 million hectoliters. Heineken NV made a net profit of 209 million euro’s in 2008. This is less than the profit of 1119 million euro’s in 2007. This is due to the acquisition of Scottish and Newcastle (Heineken Annual Report 2008 and 2007).

Heineken is active in many countries and within countries Heineken has many business units. This research is conducted at the Horeca division of Heineken Netherlands.

1.1.1 Horeca division

(8)

1.1.2 IT Strategy

The IT strategy of Heineken is based on four interdependent fundamental pillars: foundation, information, innovation and organization. The HeiSale project is related to the foundation pillar. The foundation pillar stands for automating business processes in a standard way and for leveraging standard IT solutions. The goal is to manage common business processes using common systems and shared services (Heineken Intranet). The HeiSale system is such a common system for Heineken Europe.

1.2 Project information

Heineken’s Horeca organization uses many information systems to support their processes. This has resulted in a very complex system landscape (A description of the TO BE and AS IS system landscape can be found in Appendix 1). Heineken Europe decided to replace some of their legacy systems by a new SAP system; HeiSale. This SAP system consists of an ECC (ERP Central Component, ERP stands for Enterprise Resource Planning), CRM (Customer Relationship Management) and LEO (Logistics Execution Optimizer) component. Heineken Netherlands is the first in Europe to implement the new systems. After rollout in the Netherlands, the rest of Europe will implement the systems.

1.2.1 Scope HeiSale

Figure 1 describes the Heineken organization of the Netherlands in a little bit more detail. The Horeca organization is part of the commercial organization of the Netherlands. Only the Horeca organization of Heineken NL is going to use the new SAP systems. Retail, Marketing and all other entities are not involved in the project and are not going to use the new systems.

H e in e k e n N L H e in e k e n N L s u p p ly C o m m e rc ia l o rg a n iz a tio n V ru m o n a H e in e k e n N L b re w e ry s e rv ic e s H o re c a R e ta il M a rk e tin g

(9)

The Horeca organization of Heineken consists of various entities. Except for the HR (human resource) manager and the horeca secretary, every entity in figure 2 is going to use the new systems. A more detailed description of the Horeca organization is presented in Appendix 2. Horeca Manager Management Service Organization (TSO) National Sales Management National Account Management National Category Management National Trade Marketing Management Business Improvement Manager Horeca Secretary Controller HR Manager

Figure 2: Horeca Organization (Adapted from HNL organization v07.ppt)

National sales management is by far the largest department. HeiSale has its biggest effect there. Within the responsibility of national sales management are the different sales and logistical units in the Netherlands. The majority of the people work for those units and the units are involved in the majority of the processes. In figure 3 and 4 the responsibility areas of the different sales and logistical units are shown. The areas responsible for sales and for the logistical operations are somewhat different.

Sales responsibility areas wholesale: 1.1 national: National Acc Mgt (NAM) 2.9 regions: Noord Nederland

3. Noord Holland (Alkmaar) 4. Amsterdam 5. Zuid Holland 6. Zuidwest Nederland 7. Utrecht 8. Oost Nederland 9. Zuidoost Nederland 10. Limburg 11.3 islands: Terschelling 12. Vlieland 13. Ameland

(10)

Logistical resp. areas wholesale:

1.11 region DC Noord Nederland (Drachten) 2. Noord Holland (Alkmaar) 3. Amsterdam

4. Zuid Holland Rotterdam 5. Zuid Holland Wateringen 6. Zuidwest NL Etten-Leur 7. Zuidwest NL Hulst 8. Utrecht

9. Oost Nederland

10. Zuidoost Nederland (Oss) 11. Limburg

12.3 islands: Terschelling 13. Vlieland 14. Ameland

15. Centraal Magazijn (W’veen) 16.2 cellar beer Den Bosch

17. Wijlre

Figure 4: Logistical responsibility area's (Derived from: HNL sales & logistical organization 08 08 01.ppt)

1.2.2 Processes

Heineken Europe and ARIS consultants identified 6 major process groups for the Horeca division. The HeiSale system is based on these process groups. Heineken NL identified another process group (market to order) that was not stated in Europe. The process groups HeiSale is based on are described below.

Market to Order: to make sure customers by the products of Heineken for a price

that corresponds with profit margin goals of Heineken.

Order to Cash: for effectively and efficiently acquiring and storing goods and

services against good conditions.

Purchase to Pay: for taking customer orders and fulfilling these orders logistically,

financially, administratively as efficient and effective as possible.

Record to Report: process for recording and reporting financial facts for external and

internal reporting and analysis.

Master Data Management: for on time, complete and consistent registering of

master data.

Customer Technical Services: for technical support for customers. For installing

installations and performing after sales services.

(11)

1.2.3 Project planning

The project consists of two implementation waves. Not all departments are going to be supported by the new systems simultaneously.

Wave 1

The first wave starts in the last quarter of the year 2009. In wave one HeiSale is first implemented in the regional units and the islands, starting with Alkmaar. After roll out in the units, HeiSale is implemented for Cellarbeer and the Central Warehouse. All processes, except for customer technical services, are going to be supported at this time. Finally, in wave 1, HeiSale is implemented for the technical service organization. National account management is gradually adapted to HeiSale, because national account management customers are not situated in a specific region.

Wave 2

(12)

2 Problem statement

2.1 Problem description

Besides developing the HeiSale system for supporting their operational processes, Heineken is developing a business intelligence (BI) solution based on the HeiSale system. This business warehouse in combination with the transaction SAP system and Every Angle (a SAP add-on for reporting) should support the information need of the Horeca organization. It should provide employees with information that enables them to monitor firm performance from day to day.

SAP, BI, and Every Angle must provide the following type of reports:

- operational reports;

- reports about process control, two important measurements are:

o On Time In Full

o First Time Right;

- strategic/tactical reports and KPI (Key Performance Indicator) reports.

Because the systems are developed for Heineken Europe they are not fully adjusted to the information need of the Horeca organization of the Netherlands. The Horeca division does not know how the systems being developed can be used to satisfy the information need of the various managers in the organization. Also Heineken NL does not have a process to determine which manager needs what information. One of the symptoms is that definitions of KPI’s are changing every year.

2.1.1 Problem definition

The information need of managers at the Horeca division of Heineken NL is not clear and it is unclear how the systems being developed can and will satisfy that information need.

2.1.2 Research objective

(13)

2.1.3 Main research question

How should the Horeca division of Heineken NL use the different BI tools to ensure that every manager is provided with appropriate information for the coming years?

2.1.4 Sub-questions

The following sub questions are constructed for answering the main research question.

1. How are information requirements of managers determined and how is information provided to managers in the current situation and what are the consequences for the information delivered to the managers?

The first aim of this sub question is identifying how information requirements of managers in the Horeca organization are determined. The second part of this question is concerned with assessing how the Horeca division provides information to managers. The departments involved in reporting and the systems used are identified. Finally, the consequences for the information delivered to managers are described. This sub-question is dealt with in chapter four.

2. What changes in the Horeca organization do the new systems enforce and what are the implications for the way information is delivered to managers?

This questions aims to identify what changes take place in the Horeca division when the systems are implemented. Identified is if work tasks change, if the departments responsible for providing information stay the same and if possible changes resolve the problems identified in the current situation. This sub-question is treated in chapter five.

3. What are feasible directions for overcoming the problems related to reporting that remain after the new systems are implemented?

(14)

4. How should the departments responsible for reporting decide if information is

appropriate for a manager and make sure managers receive that information by using the new systems?

By now, problems concerning reporting and feasible directions for overcoming those problems are identified. When information is appropriate for a manager is also defined. Now, the Horeca division must make sure that managers only receive appropriate information. Answering this sub-question deals with that. Advice is given for achieving this and for overcoming the other problems identified in previous sub-questions. Chapter seven is concerned with answering this sub-question.

2.1.5 Scope

The scope of the research will be implementation wave 1 up till implementing HeiSale for Cellarbeer. This affects the processes that are in scope and the managerial entities that are in scope.

Organization

Horeca division of Heineken NL

Process groups - Market to Order - Record to Report - Order to Cash - Purchase to Pay Managerial entities

- National Sales Management - National Account Management - National Trade Marketing - National Category Management

Type of reports

(15)

2.1.6 Stakeholder analysis

Many groups are influenced by the introduction of HeiSale and many groups influence the successful introduction of HeiSale. In other words, various stakeholders can be identified. This corresponds with the definition of Freeman, who defines stakeholders as “any group or individual who can affect or is affected by the achievement of the organization’s objectives” (1984, p. 46). There are a lot of people involved in the introduction of HeiSale and there are a lot of people who are going to use the new systems. All of them are stakeholders, but I’m going to narrow them down for scoping reasons.

Following the sub-questions, two groups of stakeholders must be identified. Sincet here is no agreed method for identification of stakeholders (Heidrich et al., 2009), I will narrow them down based on whether they have information needs or are (or going to be) involved in reporting.

(16)

3 Research design and methodology

This chapter presents the research approach chosen. Also the methods and sources used for answering the sub-questions are presented.

3.1 Design science

Hevner et al. (2004) combine behavioral science to explain and predict phenomena related to the business need and design science for building and evaluating artifacts to meet that need. This research aims at identifying how appropriate information can be provided to the managers of the Horeca division using the systems being developed (the business need). To prevent confusion, the information need of managers is not the business need. The business need is identifying how the systems being developed can be used to provide managers with appropriate information.

An artifact must be created to meet the business need. According to Hevner et al. (2004) an artifact can be a construct, model, method or instantiation. The artifact created in this research is a method for using the BI tools. This method helps identify if information is appropriate for a manager and how this information must be provided. Therefore, this research can be characterized as design science as formulated by Hevner et al. (2004); a method is developed to meet the business need of the Horeca division of Heineken.

(17)

Figure 5: Design-Science Research Guidelines (Derived from Hevne et al. (2004))

3.2 Systems design

According to De Leeuw (2002) a design should not only be a measure for solving the problem, but should provide steering measures in changing circumstances. This corresponds with the objective of this research: design the organization within the Horeca division of Heineken NL using the different BI tools to provide managers with appropriate information for the coming years. The coming years in the objective implies that circumstances may change. The method developed in chapter 7 must be applicable in changing circumstances. If this is true is also evaluated in chapter 8.

3.3 Development process

This research aims at providing the Horeca division with a design for using the different BI tools to ensure that every manager is provided with appropriate information for the coming years. It is sensible to use a development process when designing something. Although this research does not aim at developing a physical product, a design process still helps coordinating and planning the project (Ulrich & Eppinger, 2003).

(18)

Planning: during this phase the goal of the project and constraints are determined.

This is done in previous chapter.

Concept Development: during this phase the needs of the Horeca division are

identified. The specifications and features of the design are determined in this phase. Next two chapters deal with this part of the development process.

System Level Design: includes decomposition of the product in subcomponents and

components and a definition of the product architecture. Here it entails the design of the components of the proposed solution. This includes a description of the information that is appropriate for various managers, systems that can provide that information and who must provide the report containing that information.

Detail Design: for product design this phase includes specification of parts, materials

etc. In this paper it entails the detailed design of the general solution. However, the distinction between this and previous phase is not made in this paper.

Testing and Refinement: in this phase the solution is presented to the managers

responsible for using the BI tools. Refinements and adjustments to the initial design are made. The outcome of this phase is an accepted version of the design that is used after implementation of HeiSale.

Production Ramp Up: implementation of the solution in the Horeca organization of

Heineken NL. The roll out of the HeiSale systems starts in the last quarter of the year 2009. This phase is therefore out of scope. However, when the stakeholders accept the proposed solution, implementation is just using the solution.

3.4 Methods and sources

(19)

1. How are information requirements of managers determined and how is information provided to managers in the current situation and what are the consequences for the information delivered to the managers? (Chapter 4)

How information requirements are determined is answered by interviewing both the managers of the Horeca division and the managers responsible for delivering reports identified in the stakeholder analysis. How information is provided to the managers is also answered by interviewing managers of the Horeca division and the managers responsible for reporting.

The last part of this sub-question is answered by interviewing the managers of the Horeca division. These managers are asked which reports are delivered to them, what KPI’s are presented in these reports and which other information is presented in these reports. They are also asked what measures they can take to influence the content of these reports. The theory for effective control of De Leeuw (2002) is used to identify the problems with the information currently delivered to the managers.

The managers of the Horeca division and the managers responsible for reporting presented in figure 6 were interviewed for answering the different parts of the sub questions. The column ‘information requirements’ describes who was interviewed for determining how information requirements are gathered. The next to columns show who was interviewed for identifying how information is provided to managers and what information is delivered to managers in the Horeca division.

Part sub-question Information requirements How is information provided Information delivered

Reporting managers

Head of control X X

Logistics controller X

National accounts controller X

Business analysis managers X X

Information management managers X X

Financial services managers X

Horeca Managers

Regional sales managers X X

National logistics manager X X

National account managers X X

National category managers X X X

National trade marketing manager X X X

(20)

2. What changes in the Horeca organization do the new systems enforce and what are the implications for the way information is delivered to managers? (Chapter 5)

This sub question is answered by interviewing employees in the project team concerned with the implementation of the new systems and analyzing documents about the project. Also the Heineken Intranet is used to identify what processes currently exist in the Horeca division. The results of these interviews and the analysis of the project documents are compared with the current situation. Identified is if the current problems will be solved due to changes enforced by the systems.

3. What are feasible directions for overcoming the problems related to reporting that remain after the new systems are implemented? (Chapter 6)

Based on the outcome of previous sub-questions, directions for overcoming the remaining problems are given by exploring existing literature on the problems and when no literature exist, by using common sense. The directions identified where discussed with the managers of control, business analyses, information management and the HeiSale project team for feasibility.

4. How should the departments responsible for reporting decide if information is appropriate for a manager and make sure managers receive that information by using the new systems? (Chapter 7)

(21)

4 Current situation at the Horeca division

This chapter aims at answering the first sub-question “How are information

requirements of managers determined and how is information provided to managers in the current situation and what are the consequences for the information delivered to the managers?”. First, the way information requirements are determined is

described. Subsequently, the way information is provided to managers is identified. Next the information that managers receive is evaluated using the theory and models of De Leeuw (2002). Finally, the identified problems are related to the way information requirements are determined and the way information is provided to managers.

4.1 Determination of information requirements

The two types of reports managers of the Horeca division receive are reports about KPI’s and non-KPI reports (strategical/tactical reports).

4.1.1 KPI reports

KPI reports contain information on the KPI’s of managers of the Horeca division. The management team of the Horeca division and the Horeca controller determine the KPI’s each year. The management team of Horeca consists of the head of Horeca and the heads of the departments national sales management, national account management, national category management and national trade marketing management. They determine the KPI’s based on the KPI’s they get from the board of directors of Heineken NL. The board of directors of Heineken NL determines KPI’s based on the KPI’s they get from the board of directors of Heineken Europe.

4.1.2 Non-KPI reports

(22)

4.2 The way information is provided

In figure 7, the reporting structure within the Horeca division is described. The units responsible for reporting are described, the reports delivered are described and the units receiving reports are described. This picture is a simplification of the situation at the Horeca division. In reality more departments, systems, reports and relationships exist. The most important departments and systems are mentioned.

Horeca Manager

National Sales Manager

Regional Sales Managers

Regional Logistics Managers National Logistic Manager

National Category Management

National Account Management

National Trade Marketing IM

Horeca Organization

Detail reports for Horeca organization, reports depend on

demand in organization

Reports on promotions, reports from StarBusiness (Promotool,

StarBusiness)

Business Analyse

Reports on volumes,cashflows, profit and losses (general ledger) Proost, WMS, Dolfijn, EDI,

Marketmaker, LotusNotes, MXP, Fieldvision, Enorm, Powerplay, SAP BW, Excel

Reports on suppliers (HBDW) Financial Services (accounts

recievable)

Business Control Proost, Rems, Floris,

Consist, Onquard

Overdue Payment Reports

Reports on suppliers (SAP, Proost, MarketMaker, Onquard)

Proost, SAP TM1, Heyperion and most other systems Financial Services (accounts

payable)

Reports on matching, workload, not processed payments BasWare, SAP, Proost

Adjusted reports, reports depending on

need management

(23)

The light blue boxes are departments that produce reports, the yellow boxes contain the systems used for reporting and the green box is the Horeca organization that receives reports. Business control is also made green, because it both receives and provides reports. The organizational units identified in the picture are described below.

4.2.1 Horeca organization

The Horeca organization of Heineken consists of the departments national account management, national sales management, trade marketing management, category management. Appendix 4 presents the processes that can be identified in the Horeca organization. Various departments deliver reports; the Horeca organization itself can even be a reporting source. This is because every department has its own controller. Controllers produce reports for the departments they work in and adjust reports made by other departments.

4.2.2 Financial services accounts receivable

They spread information through the organization about overdue payments and the amount corresponding with those overdue payments. These reports are provided monthly and contain information per region. This information goes to the legal department, unit controllers, regional sales managers, business control, etc.

4.2.3 Financial services accounts payable

They spread reports about matching, workload and not processed invoices to logistical departments. They upload their reports on the intranet.

4.2.4 Information management (IM)

Information management provides reports about volumes, Kikkers, logistics etc. These reports contain detailed figures. Managers within the Horeca division can ask for a certain report and IM decides if they will produce that report. If so, they make the report and spread it through the organization periodically.

4.2.5 Business analyses (BA)

(24)

Horeca Organization. Managers in the Horeca organization can ask business analyses to produce reports.

4.2.6 Business control

Defines KPI’s together with the management team and controls the Horeca organization based on values of those KPI’s. They adjust reports delivered by other departments and make own reports.

4.2.7 Systems

As can be seen, many systems are in use and there are even more systems that are not shown in figure 7. A description of these systems is outside the scope of this research. Important to remember is that different systems can be used to produce reports. The data in the various systems is not consistent. The systems are often not related to each other and data entered in one system is not automatically processed in other systems.

4.2.8 Comments on reporting structure

As explained at the beginning of this paragraph, the reporting structure presented above is not complete. All relevant organizational units are identified, but not all reports and relationships. For a general idea about the reporting structure and problems associated with the current structure, figure 7 is sufficient.

What can be seen is that reports can come from many sources; various organizational units are responsible for building and distributing reports. Many departments were mentioned that deliver reports to the managers of the Horeca division during interviews with those managers.

(25)

What also can be seen from the figure above is the large number of systems that are used for reporting. Reports on KPI’s and other information can be generated by multiple systems and made by multiple departments.

4.3 Information delivered to managers

This section first describes on which two aspects the information delivered to managers is assessed. The aspects are identified by using the theory on effective control of De Leeuw (2002). Subsequently, the information delivered to managers is assessed on those aspects. Finally, the KPI’s used at the Horeca division are analyzed in more depth.

4.3.1 Aspects the information delivered to managers is assessed on

The managers within the Horeca division control certain areas of the Horeca division. They each control parts of the Horeca organization. The manager and the area he controls can be represented by the BO BS model of De Leeuw (2002) presented in figure 8. Where BO is the control system (the manager in this particular case) and BS the system he controls (a part of the Horeca organization). The model represents the information going to the control systems (indicated by the left arrow), the measures from the control system to the controlled systems (indicated by the right arrow) and the control (BO) and controlled system (BS) themselves.

BO

BS

Figure 8: BO BS model

(26)

1 Goal

A system should have a goal and a mechanism to measure attainment of that goal.

2 Model of controlled system

For control of the controlled system, you should know the effect of measures, be aware of causality.

3 Information about environment and state of the system

For a measure to be effective one should know the state of the system and the environment.

4 Sufficient steering measures

The variety of steering measures should be at least as big as the variety of disruptions of the system.

5 Capacity of information processing

Information needs to be processed, so the control system needs to have sufficient capacity to do that.

The information managers receive is assessed on two aspects based on the five conditions above. The first aspect is if the information managers currently receive is sufficient to know the state of the system they control (condition 3 of De Leeuw (2002)). The second aspect is if the KPI’s and information managers receive is relevant to the controlled area of a manager (combining condition 2, 3 and 4 of De Leeuw (2002)). In other words, if the information or KPI presented in reports can be influenced by steering measures a BO (manager) can take. Because when managers cannot influence the values of KPI’s and information in non-KPI reports they should not get those KPI’s or reports. Analysis on the aspects is done by analyzing the BO BS configurations using the attributes below. Descriptions of the attributes are obtained by interviewing managers of the Horeca organization.

Description BO Description of control system

Objective Objective of control system

(27)

Information from BS Information received by BO (manager)

KPI KPI’s used by BO

Steering measures Measures the BO (manager) can take.

Figure 9 graphically presents the BO BS configurations that can be identified in the Horeca division of Heineken. In Appendix 5, the majority of the BO BS configurations are described using the attributes above. The reason for not describing all entities and relationships is because it was not possible to interview every manager of the Horeca division.

(28)

4.3.2 Scores on the two aspects

Information sufficient for knowing state of the system?

During interviews with managers from the BO’s a number of problems related to information delivered to them came about. One should keep in mind that not all managers experienced the problems described below. Still it is important to mention them, because these problems indicate that the information presented to managers of the Horeca division is not always sufficient to know the state of the system they control.

Some managers mentioned getting too little information; regional sales management does not get reports on costs, commercial costs (which they can influence directly) and profit. Category management does not get information about suppliers for example. They need this information to choose suppliers that score good on OTIF (on time in full deliverances) and charge the lowest prices. Also logistics does not get information on OTIF of their warehouses. National account management does not get information about profitability per customer.

Also, reports do not always come on a regular basis; some months, managers get a report, the other month they do not (regional sales management). Some managers get multiple reports on the same KPI with different values on that KPI. For example, a volume sold report made by business analyses showed a 10 % difference on volumes sold compared with the same report made by a controller. Business analyses used SAP and the controller used the Proost system for producing the report. The SAP report considered volumes leaving the factory as volumes sold, while Proost considered volumes delivered to the customer as volumes sold.

Another problem mentioned was that information and values on KPI’s are sometimes experienced as unreliable (regional sales manager, trade marketing, category management).

Is information presented to managers relevant to their controlled area?

(29)

they can take action, because the information is not relevant for their department. For example, category management gets reports and KPI’s on volumes per channel, which is impossible to influence by them. Trade marketing managers have KPI’s on Kikker activities (activities for acquiring sales points for new concepts or products). But for getting the targets corresponding with that KPI, he needs to be able to control salespersons, which he can’t. KPI’s and information managers get are not always based on measures that manager can take. The KPI’s used at the Horeca division are analyzed in more depth in subsequent section.

4.3.3 KPI’s used at the Horeca division

The KPI’s used at regional sales units of the Horeca division can be found below. To show the changes in KPI’s each year, the KPI’s for a regional sales unit in 2008 and 2009 are given. Still there are more KPI’s. Logistics has its own KPI’s, even as trade marketing and category management. Logistics has KPI’s on costs per coli, trade marketing on the success of Kikker activities (measured by sales points), and category management on volumes, margins and sales. The KPI’s for the sales units are best documented and used for explanation purposes.

KPI’s 2008 KPI’s 2009

Volume tap beer concern ATL / BTL (commercial costs) Revenue incl. discounts Revenue

Volume non-alcoholics Volume non-alcoholic Fit 2 Fight: (reducing costs) Discounts on tap beer EBIT regional office Discounts on non-beer DSO (days sales outstanding) DSO (days sales outstanding) Overdue payments Overdue payments

Kikker: sales activities Visits to clients

Amstel voetbal: sales activity Capex Loans (niet EHF) KIT: sales activity Capex tap installations

Jillz: sales activity Fixed Expense (incl. energy)

(30)

Comments on KPI’s used at the Horeca division

Parmenter (2007) defined seven key characteristics of KPI’s. One of them is that a KPI should be a non-financial measure. A KPI is a result indicator when you put a dollar sign on it. The majority of the KPI’s of 2008 and 2009 are result indicators. They are costs (bonuses, investments, expenses), income or income related KPI’s (DSO, overdue payment). Only ‘visits to clients’ is no result indicator. Also, all KPI’s are past indicators, measuring results and events in the past. They do not predict (except for visits to clients maybe) performance in the future.

Furthermore, regional sales managers cannot influence all of their KPI’s. When volumes or revenues decline, there is no measure a manager can take to overcome declining volumes and revenue. Volume and revenue highly depend on external factors like the economic situation (credit crunch) and rules and regulations (smoking ban). Paying more visits to customers does not increase volumes. The customers of the Horeca division depend on the willingness of people to spend their money in their pubs and restaurants. A regional sales manager is unable to influence that. Another example of a bad KPI is overdue payments. Regional sales units do not handle payments; this is done by financial services accounts receivable.

Managers should have KPI’s and get information that is relevant to their controlled area. They should be able to influence values of KPI’s and information in reports by measures they can take within their controlled area. Then, they can act upon the information they receive. At this moment, they cannot influence all KPI’s and information presented to them in reports.

4.4 Causes to the problems

The interviews with managers of the Horeca division revealed many problems related to reporting and information. This section relates the problems to the way information requirements are determined and the way information is delivered to the managers.

4.4.1 Problems related to information requirement gathering

(31)

information is appropriate for a manager. The departments responsible for reporting are not aware which information is relevant for various managers. The KPI’s used in the Horeca division are not always based on measures managers can take, but follow from KPI’s stated by higher management.

4.4.2 Problems related to the way information is provided

The cause of multiple reports with conflicting values on KPI’s being delivered and reports not being delivered on a regular basis is because there is no control structure for reporting. The various departments are not aware if reports made by them are also made by other departments and how often reports should be provided. Also, multiple systems can be used to make the same report which can lead to conflicting values in different reports.

4.4.3 Cause of unreliable information

Unreliable information in reports cannot be related to the way reports are delivered or information requirements gathered. This is due to another reason. This is because employees do not always work according to work instructions and enter data the wrong way. This is explained by the following two examples.

When visiting a customer for Kikker activities (activities for acquiring sales points for new concepts or products), a sales person must process the result of the visit in MarketMaker (CRM system) and Proost (Administration System). The result can be: new sales point, no new sales point or no decision made yet. This result must be processed directly after a visit. For a number of reasons this does not happen.

(32)

The second example is about sales points won and lost; salespersons immediately register won sales points, but are not that fast with registering lost sales points. This distorts data as well.

4.5 Problem summary

For clarification, the problems and their underlying causes identified in this chapter are clustered below.

Problem 1

Managers mentioned that they get too much information. Other managers mentioned getting too little information. For some decisions managers have to take, there is no information available at all (profitability per customer). Another problem is that the information in the reports is not always information about the control area of managers. Managers cannot always influence KPI’s and the information presented to them in reports.

Cause of problem 1

There is no process for determining which information is appropriate for managers. KPI’s and information managers receive are not always based on measures a manager can take.

Problem 2

Also, not all reports are provided on a regular basis, some reports are not provided every month. Other problems occur because the information they get is conflicting; different reports on the same KPI’s can show different results.

Causes of problem 2

(33)

Problem 3

Management cannot always make decisions based on the reports they get, because information in the reports is unreliable.

Cause of problem 3

(34)

5 To Be situation

This chapter aims at answering the sub-question “What changes in the Horeca

organization do the new systems enforce and what are the implications for the way information is delivered to managers?”. First it is identified if work tasks change due

to the new systems. Next, it is identified if the existing departments in the Horeca division and the departments responsible for providing information change. Finally, it is determined if those changes solve any of the problems identified in previous chapter.

5.1 Work processes

The work processes presented in figure 10 are the processes on which the SAP systems being developed are based. These processes are going to be implemented at the Horeca division together with the implementation of the SAP systems in 2010.

(35)

ARIS consultants defined the work processes for the To Be situation. ARIS consultants have defined common processes for Europe based on best practices within Europe. Below the new work processes are compared with the processes that can be identified in the current situation. The processes in the current situation are described in appendix 4. This comparison is only made for the 4 process groups that are in scope; market to order, purchase to pay, order to cash and record to report.

5.1.1 Market to order

The processes in market to order highly correspond with the commercial work processes in the current situation (appendix 4). The processes are generally the same, but category management processes are added as well. This is a different process group in the current situation.

Departments involved in the process group

- National category management - National trade marketing

- National sales / National account management

5.1.2 Purchase to pay

The processes within purchase to pay consist of part of the logistical work processes and part of the financial work processes (accounts payable) in the current situation. The difference with the current situation is that processes for formulating contracts with supplier are added. The assortment definition processes are also added.

- National category management - Logistics

- Financial services accounts payable

5.1.3 Order to cash

(36)

- Telesales - Logistics

- Financial services accounts receivable

5.1.4 Record to report

The processes within record to report are not described in the current situation. It are processes for external and internal financial reporting. They are described in section 4.2. The roles do not change in the To Be situation.

- Business Control - Business Analyse

- Shared Financial Services - Information management

5.1.5 Result of changed processes

The process groups of the current situation do not longer exist. Category management work processes, financial work processes, logistical work processes, telesales work processes and commercial work processes are replaced by the process groups market to order, purchase to pay, order to cash and record to report. The general work processes do not change, but the allocation of the processes to process groups is somewhat different.

Although the general processes stay the same, work instructions can differ due to the ARIS descriptions. Work descriptions will be different due to using the new systems. The Heisale project team is going to formulate the new work descriptions. Those detailed work descriptions are not ready at the moment. Work descriptions in the To Be situation will change, although the general work processes are still the same.

5.2 Departments

(37)

that the reporting structure will not differ from the current situation. The only thing that needs to be determined, according to the Horeca division, is which reports are needed and how the new systems can deliver those reports. No serious effort has been made though. The To Be reports are identified by taking the reports used in the current situation as input for the tools and systems that are being developed. Only some minor adjustments have been made.

5.3 Do the changes solve the current problems?

For the new situation, new work tasks are defined based on the new systems. Work descriptions will be formulated that differ from the way work is performed in the current situation. In previous chapter was identified that employees at Heineken do not always work according to work descriptions. They do not enter data as they are supposed to do. When employees at Heineken do not work according to the work descriptions in the new situation, the data stored in the systems stays unreliable.

It would be better for the Horeca division to work with systems based on how work is performed in the current situation. When it is impossible to capture the content of work in standard work processes, you have to account for that. Steering measures will not have any effect, or a different effect as intended, when they are based on processes that are not being followed. You lose steering possibilities when using a system based on non-existing work processes. Then you are in exactly the same situation as before implementation of HeiSale. Measurements on KPI’s and other attributes will not contain reliable information. The effect of employees deviating from the work processes on which the SAP systems are based on values of KPI’s and other measurements needs to be taken into consideration. This problem has become more relevant in the new situation, because the work descriptions will change.

(38)

systems available for reporting, at least HeiSale BI, HeiSale and Every Angle. And there are also the other problems left that are not going to be solved by the HeiSale systems. Summarizing:

- employees are not likely to work according to work instructions;

- the information need of different managers is still not known and KPI’s are still not based on what can be controlled by a manager;

- there are still various departments responsible for reporting without a control structure for overcoming problems associated with it;

- different views on the data are still possible, this means values on KPI’s can be different depending on the view or system that is chosen.

(39)

6 Overcoming the problems

This chapter aims at answering the sub-question “What are feasible directions for

overcoming the problems related to reporting that remain after the new systems are implemented?”. The inputs for this chapter are the problems identified by answering

the first two sub-questions. In this chapter, directions for overcoming the four causes to the problems at the Horeca division are presented. Based on the situation at the Horeca division it is determined which problems are worth solving and which solutions are feasible to implement. Based on that outcome a solution for overcoming the problems is proposed in the chapter 7.

6.1 Problems that must be solved

HeiSale will not solve the problems identified in the current situation. When HeiSale is implemented, consistency in data will be better because of fewer systems. However, various systems (HeiSale BI, HeiSale or Every Angle) and different views on data can still be used when making reports. Furthermore, using fewer systems does not make sure reports are consistent, delivered in a timely manner and contain appropriate information. Also, reports still can come from various departments. A clear vision on reporting is still needed. For solving the problems a solution must be developed that makes sure that:

- employees work according to work instructions: to make sure data in the systems is correct and reliable (paragraph 6.2);

- managers only get the information on which they can take action, in other words, information they can influence (paragraph 6.3);

- there is a reporting structure with clear responsibilities and tasks: to make sure reports come from one source; no multiple values on KPI’s or other attributes are presented to managers (paragraph 6.4);

- a clear view on the data is chosen and communicated throughout the organization: to makes sure one system is used for generating certain reports (paragraph 6.5).

(40)

6.2 Employees working according to work instructions

6.2.1 ERP adoption

Employees working at the Horeca division must accept the new work processes and the impact on their daily work enforced by the new systems. Overcoming resistance to change is necessary for a successful implementation of an ERP system. Resistance to change can be due to perceived job risk and resistance due to habits (Sheth, 1981). At Heineken the resistance is primarily due to habits: ‘we have always done it this way’. There is still no good way for overcoming that resistance. ERP literature is still evolving and has not build a theoretical base to overcome resistant to change (Aladwani, 2001).

Aladwani (2001) combines marketing and ERP implementation strategies for successful ERP implementation. The ERP field can benefit from the marketing perspective, because of similarities between the two perspectives: overcoming consumer (user) resistance to new products (ERP systems). Since the implementation at Heineken is an ERP implementation and the reporting tools extract data from the ERP system, his findings can be useful for Heineken. When the ERP system is not accepted, the reporting tools lose all value.

(41)

Figure 11: A model of successful ERP adoption (derived from Aladwani, (2001))

The project team has begun communicating the benefits of the ERP system to the organization. Not to all levels yet, because the project is still in progress, but everybody will be informed during this year. Communication of the general operations has not begun, because there is no system to show to people yet. Minimizing adoption costs has not been considered. Individuals and groups are involved in the introduction: key users, business process owners, managers and other employees.

(42)

Problem areas are:

- minimizing adoption costs of the project;

- securing opinion leaders support, no real opinion leaders have been identified; - a date for introduction is set, no matter what happens;

- top management commitment and support is easily lost when the first results are negative.

The focus will not be on the problem areas above. The project is on its way and the project team is aware of above issues. Priorities are not with those issues and the project team feels it has done everything in its power to make sure employees are going to adopt the new systems and the new work instructions associated with the systems.

6.2.2 Dirty data

When the above issues are addressed, there is still another issue to take into consideration. Although the ERP system might have been accepted and the implementation might have been successful, wrong usage of the system might lead to dirty data. Dirty data are inaccuracies or inconsistencies within a collection of data. Dirty data within a data source makes it difficult or unwise to use the data for analysis (Vosburg & Kumar, 2001).

When not working according to the work descriptions forced by the ERP system – HeiSale BI and Every Angle data is loaded from the ERP system- data gets dirty and becomes unreliable. Reports are of no value when using that data. Incomplete or incorrectly entered data is dirty data. This problem already exists in the current situation; data is not always entered correctly in the systems. There is also the possibility at Heineken that dirty data is due to unwillingness to work using the new work descriptions. Either way, dealing with dirty data must be addressed. Otherwise the reports generated by the systems are of no value.

(43)

- Migrate and clean dirty data from the legacy systems.

- Work performed by external consultants must be understood, for example structures and data definitions. It must be documented for future use.

- Ensure that every stakeholder understands responsibilities with respect to maintaining data integrity. Develop shared understanding and importance of data. Overcome comments like “we have always done it this way”. It takes time for users to comprehend and use integrated data; care should be taken to ensure that. - Test for all possible user groups in all possible situations, also test entering bad

data before implementing ERP. Implement the system in phases.

- Maintaining data integrity is an ongoing process and does not stop with implementation. On-going training, before and after implementation of the ERP system is needed. Experiences must be exchanged between users from the same work environment.

- After implementation: data audits should include daily integrity checks and regular audits performed by user groups. Tools for auditing are needed. Problems uncovered within audits must be shared (Kikker example). Employees must see their role in context of the business.

The first guideline has been taken into consideration. Data is cleaned or enriched when certain data fields within the various systems do not correspond. However, there is no detailed plan for migration of data, due to uncertainties about which systems disappear or stay and the exact specifications of the SAP system.

The second point is the responsibility of the international business process owner (IBPO). He knows which requirements every country has. The data definitions and rules are documented in detail. When a consultant wants to deviate from the standards, he has to discuss that with the IBPO. Those changes are also documented. IBPO’s check if consultants are building the systems according to the standards.

(44)

maintenance is not made available easily. On higher management levels, the importance and impact of data integrity is not fully acknowledged everywhere.

On lower levels, understanding of the importance of data is even worse. The benefits of disciplined working are hard to identify for them. Most of the time, benefits on individual levels are not even there, benefits only occur on higher levels. A driver does not care if the customer data is correct; he still knows where to go. Business process owners (BPO) working in the project are responsible for stressing the importance of data to individual users, but no plans for achieving this have been made though. No serious attempt or plan has been made for informing and convincing the end users of data importance.

The fourth guideline is addressed, although the testing period is too short to cover all testing. Also, although the importance of testing bad data is acknowledged, it is not sure there is sufficient time for ‘negative testing’. Implementation is done in phases, first Alkmaar, and then the other parts of the Netherlands.

The fifth guideline is only partly addressed. There is a master data manager (Gert van Zanten) responsible for data integrity and the BPO’s are responsible that employees use the systems appropriately. Also, employee training has been planned for. The duration of the training depends on the time and budget that is still available. Planned training time is 4 to 6 weeks. Also, the HeiSale team wants to test the employees using exams, if not passing exams has consequences is not clear yet. How this training needs to be done and how experiences between users can be shared is not specified.

The master data manager is responsible for continuous data audits after implementation. But how to share results of those audits and how users can see their role in the context of the business is not known.

6.2.3 What needs to be done?

- Stressing the importance of data integrity to all levels of the Horeca division (from management to users).

(45)

- Planning for after implementation training, evaluation of system use, data auditing and communicating results throughout the organization.

Just as with ERP acceptance issues, the focus is not on the issues above. The project is on its way and plans have been made for addressing above issues. Not all issues have been addressed, but the project team is aware of them and has the intention to address them. Therefore, the final solution will be focused on the other three problem areas. A solution developed for those problems has to be used when the organization is using the systems already.

6.3 Appropriate information and KPI’s

The analyses in chapter 4 revealed that information delivered to managers and KPI’s used at the Horeca division are not always based on measures a manager can take. An alternative for the KPI’s used at the Horeca division is described below.

6.3.1 Alternative for KPI’s used at the Horeca division

The difficulty with stating good KPI’s for the Horeca division of Heineken is the nature of the business. It is not just selling beer to customers, but helping clients sell beer, beer related products and non-alcoholic drinks to their customers. The business has grown to providing clients with a “total horeca concept”. This consists of providing high quality, a diversified product assortment, loans and helping clients with the interior design of their bars, pubs, hotels and restaurants. Furthermore, the Horeca division helps clients with promoting and organizing events, promotion outside the bar (neon lights) and provides new (product) concepts like Extra Cold and GoegGetaptBier. The products must also be delivered in a timely manner, installations must be cleaned and the relationship with the client must be a good one. Also the image of Heineken plays a big role in wetter a client chooses to buy Heineken or another beer. The KPI’s used at Heineken do not reflect these activities.

(46)

Thus, a key performance indicator could be a combination of

• Total number of clients

• Won clients

• Lost clients

Since not all clients buy the same volumes and provide the same income for Heineken, you could weigh them according to volumes or income.

For example:

Volume < 50 HL a year: factor 1 50 HL < Volume < 250 HL factor 2

Volume > 250 HL factor 3

A survey could reveal the reason why customers choose Heineken as their provider/supplier. Controllers should be aware of these factors and intervene when the number of clients KPI goes down. Also lost clients can be interviewed on why they left Heineken. This indicates what goes wrong when customers leave or do not choose Heineken as their supplier.

The number of clients KPI should not stand on its own, profitability and revenue (per customers, unit and Horeca as a whole) should be taken into consideration as well. Supplying beer for free will result in lots of won customers, but it is easy to see that this is not the way to do business.

6.3.3 Feasible alternative?

(47)

The alternative described above is not feasible, the Horeca division wants to use the KPI’s of 2009. Another complication is that KPI’s change very often and the Horeca division must use those KPI’s. Because the solution proposed in subsequent chapter is not going to be used before 2010, KPI’s have changed by then. They just keep changing. The solution proposed in subsequent chapter has to be a generic solution. It must be determined who delivers information on which KPI. Since the KPI’s change each year, the solution should be flexible enough for changing KPI’s. So it should be determined what the types of information are within KPI’s based on the KPI’s of 2008 and 2009 and who in the Horeca organization can influence the value of these information types.

So far, only KPI’s are discussed in this paragraph, but managers also get other information than information about KPI’s. Not all reports delivered to managers are KPI reports; reports can also contain non-KPI information. This information should also be based on what a manager can influence. The solution proposed in chapter 7 takes KPI reports and non-KPI reports into account. Both have to contain information a manager can influence. Only when a manager can take measures to influence the values in those reports, he should get the report. How to make sure only those reports are delivered to managers is described in chapter 7.

6.4 Reporting structure

The reporting structure in the current situation is one of the causes to the identified problems. Three alternative reporting structures are described below. These structures can help overcome the problems related to the current reporting structure. After describing these alternative reporting structures, the structure is chosen that is feasible for implementation at Heineken.

6.4.1 Structure 1

(48)

involved in training users and identifying how reports have to be made. When using only a few KPI’s that stay the same over time, you also make sure everybody is talking about the same data. Individual managers and controllers are free in using the systems and analyzing whatever they want as long as they have good results on the KPI’s. This also corresponds with the work culture at Heineken. People have freedom in achieving their goals, without having to obey lots of criteria and KPI’s. This structure requires that departments disappear or departments losing a lot of their responsibilities.

6.4.2 Structure 2

Reporting responsibility is with one central department. This department should be responsible for reporting on KPI’s and delivering non-KPI reports. This department should deliver reports on a structural basis, on time and to the appropriate stakeholders containing the right amount of information. Reports come from one department, which resolves the problems associated with multiple reporting departments. Just one value on KPI’s and other attributes is presented to the managers. This structure also requires that departments disappear or lose responsibilities.

Controllers and other users

Controllers of the various departments and other users within the Horeca division are the ones using the analytical capabilities of the systems. They are the ones that have to be instructed on how to analyze the data presented to them in reports. Users have the freedom of playing with data and using the analytical capabilities of the systems in use, but must do this according to a general acknowledged composition of KPI’s and other attributes. They can make reports composed in other ways, but they know that information does not correspond with the way information and KPI’s are composed in reports made by the central reporting department. This makes sure everyone is looking at the same data.

6.4.3 Structure 3

Maarten van den Muijsenberg ERP implementation at Heineken: improvements for information management?

ERP implementation at

Heineken:

improvements for information management?

Maarten van den Muijsenberg

Master Thesis

Technology Management

June 2009

University of Groningen

Project Company

ERP implementation at

Heineken:

improvements for information management?

Maarten van den Muijsenberg

Master Thesis

Technology Management

June 2009

Project Company

University of Groningen

Preface

Management summary

Table of contents

1

Introduction Heineken HeiSale

1.1 Company information

1.2 Project information

2

Problem statement

2.1 Problem description

3

Research design and methodology

3.1 Design science

3.2 Systems design

3.3 Development process

3.4 Methods and sources

4

Current situation at the Horeca division

4.1

Determination of information requirements

4.2 The way information is provided

4.3

Information delivered to managers

4.4 Causes to the problems

4.5 Problem summary

5

To Be situation

5.1 Work processes

5.2 Departments

5.3 Do the changes solve the current problems?

6

Overcoming the problems

6.1 Problems that must be solved

6.2 Employees working according to work instructions

6.3 Appropriate information and KPI’s

6.4 Reporting structure