Customer-Driven Sales Processes
An exploration in the field
By
A.B. Craens Student nr. S1419439
craens@gmail.com
MSc Business Administration – Specialization Business & ICT University of Groningen, The Netherlands
Monday 30th of January 2006
Supervisors:
Prof.dr.ir. J.C. Wortmann
Faculty of Management & Organization University of Groningen
Dr. G. A. Welker
Faculty of Management & Organization University of Groningen
Ir. Paul R.P. Kimmel
General Manager
Sofon BV
3
Management Summary
Sofon makes software for customer-driven sales processes. These processes make customer-specific products. Although some organizations have standardized their products, they make all quotes during a sale. Quotes are made before the customer and the selling organization come to an agreement.
An exploratory research has been conducted in 25 organizations to give Sofon more insight into the perceived problems in new branches of organizations. The nature of these organizations is very diverse. Furthermore, one third of the interviewed persons did not experience any problems. However, the research has resulted in a rich collection of information and problems. The five most frequent problems are:
1. Missing management information about the sales process
2. Account managers do not see the benefit of software or software is not being used accurately.
3. Software has been experienced as not user-friendly.
4. Customer-driven sales processes have several sources of communication problems.
5. Managers like to improve the integration of software.
The above problems are related to each other. The origin of the shortage of management information goes back to issue two and five. In addition, some software has no functionality to generate overviews at all. However, a shortage of information is not restricted to management information. CRM
1information can also be an issue of information shortage. This example illustrates that the problem of information shortage is more comprehensive. Several factors of technology as well as factors relating to the account manager influence the degree to which software is used accurately. Simplicity and the added value of the software for the account managers seems to be the two most important variables to reach a solution.
Another finding of the research are the communication problems within customer- driven sales processes. These problems can have several origins and are in some situations related to the accuracy of software usage. In more automated organizations, this problem has shifted to input problems.
1 Customer Relationship Management
4
Preface
For me, the days of the Winter of 2005-2006 were short, but long at the same time. A final thesis is often tough, but it has its charms. In this case, the emphasis was on exploring a new area.
The preliminary results were very satisfying. This gave me the motivation to give myself for more than 100%. In addition, the several end-users of this report challenged me to create a valuable product. This is not to say that there were not a hundred other details which made this a pleasant project. An informative time that also gave me new insights in myself, and in the world outside the university.
Of course, Sofon is the main customer of the report. However, more parties may be interested. First, in my opinion, the field of customer-driven sales processes is a new area on scientific study. I hope the results of this research can be used for future research at Sofon and other topic related research.
Secondly, all co-operating organizations can benefit from this report. I hope this report will give inspiration and insight into the possibilities those organizations have for the future. I am very aware of the fact that this knowledge can only contribute to a better world if it can be applied in practice. Of course, I hope this will be possible with the support of Sofon.
I would like to thank Professor Hans Wortmann for creating this opportunity and for supervising this project. I also want to thank Gera Welker for her flexibility, extra effort and excellent feedback. Furthermore, I would like to thank Paul Kimmel as supervisor at Sofon and for the wonderful time I had in Son (Eindhoven, The Netherlands). The latter was not possible without the employees of Sofon. They make Sofon an inspiring environment and a great place to work.
Special thanks go to all cooperating organizations. First of all, I’m grateful for their warm welcome and thousands of times of wishing me good luck. Secondly, I would like to thank all for their time, effort, and for the interesting interviews (I had with them).
Last, but not least: Dutch public transport, which has transported me thousands of miles through the Netherlands. Although I would like to write another thousand words, it is now the time we all waited for: The next pages of this report!
André Craens
Groningen, the Netherlands, January 2006
5 In this field everyone is still learning
22 Miles and Huberman about ordering and explaining cross-case data [Miles, 1994].
6
Table of content
1. INTRODUCTION...8
1.1 A
SHORT INTRODUCTION INTO CUSTOMER-
DRIVEN SALES PROCESSES...8
1.2 S
OFON...9
1.2.1 A short introduction about Sofon...9
1.2.2 Sofon’s software-evolution from a functional viewpoint ...9
1.2.3 Current and future software modules ...10
1.3 C
ONTEXT...11
2. WHAT IS A CUSTOMER-DRIVEN SALES PROCESS?...12
3. AT WHICH CUSTOMERS IS SOFON AIMING?...14
4. STATEMENT OF THE PROBLEM ...15
4.1 R
ESEARCHO
BJECTIVE...15
4.2 P
ROBLEM STATEMENT...15
4.3 R
ESEARCHQ
UESTIONS...16
4.4 C
ONSTRAINTS...16
4.5 M
ETHODOLOGY...17
5. RESEARCH DESIGN ...18
5.1 C
OMMON DESIGN ISSUES...18
5.1.1 Purpose of the study...18
5.1.2 Interference ...18
5.1.3 Study setting ...18
5.1.4 Unit of analysis ...18
5.1.5 Time horizon ...19
5.2 S
UMMARY OF THEA
PPROACH...19
5.3 D
ETAILED DESIGN ISSUES...21
5.3.1 Selecting organizations...22
5.3.1.1 Selecting branches ...22
5.3.2 The right number of cases...22
5.3.3 Approaching organizations...23
5.3.3.1 Selecting and finding the right persons...23
5.3.4 Interview Preparation & Techniques...24
5.3.5 Interview structure and questions...25
6. CONCEPTUAL MODEL ...26
7. PROCESSING DATA ...29
7.1 Phase one ...29
7.2 Phase two ...30
8. CRITICAL NOTES CONCERNING THE RESEARCH DATA ...31
8.1 M
EASURABILITY...31
7
8.2 C
OMPLETENESS...31
8.3 O
THER ISSUES...32
9. DISCUSSION RESEARCH DATA ...34
9.1 C
OMMON FINDINGS...34
9.1.1 Differences within the unit of analysis...34
9.1.2 Software usage ...35
9.1.2.1 CRM & ERP Systems...35
9.1.2.2 From external sales department to internal sales department ...35
9.1.2.3 How quotes are created ...36
9.1.2.4 From sales department to operations department ...36
9.2 T
HE COLLECTION OF PROBLEMS RELATED TO THE SOFTWARE...37
9.2.1 Perspective from the software...37
9.2.2 Overall perspective ...38
9.3 M
OST FREQUENT PROBLEMS IN DETAIL...39
9.3.1 No problem...39
9.3.2 Product specificness...39
9.3.3 View on user-friendliness...40
9.3.4 Communication ...41
9.3.5 Missing information ...42
9.3.6 Other issues...44
9.4 C
ONCEPTUAL FRAMEWORK BASED ON THE RESULTS...45
9.5 C
ONCEPTUAL FRAMEWORK FOR FURTHER RESEARCH...47
10. RECOMMENDATIONS...48
10.1 C
OMMON RECOMMENDATIONS...48
10.2 S
OFTWARE DEVELOPMENT AND FUNCTIONALITY...49
10.2.1 Simplicity...49
10.2.2 Insight ...49
10.2.3 Integrated & Flexibility ...50
11. CONCLUSION ...51
REFERENCES...52
APPENDICES ...54
8
1. Introduction
The context of this research will be given in this chapter by discussing the subject of customer-driven sales processes. The research was conducted on request of the company Sofon. Sofon will also be introduced in this chapter.
First, an overview of the report will be given. Customer-driven Sales Processes will be discussed in more detail in the next chapter. Chapter three sets out at which customers Sofon is aiming in general. The statement of the problem will be given in chapter four. After this, the research design will be presented in chapter five. Chapter six will introduce the conceptual model that is used during the research. The next two chapters discuss the case studies. In chapter seven the approach is explained. After this, the results are presented in chapter eight. Chapter nine will give several recommendations. The report ends with the main conclusions.
1.1 A short introduction into customer-driven sales processes
The customer-driven sales process can best be described as a process that is selling a customer-specific product or service. From this point on in the thesis, only the word/description ‘product’ will be used when describing one of these products of services.
A few samples of these products are given here. All these products have in common that they all have a unique customer-driven sales process.
• Industrially manufactured products
• Electricity for business or large organizations
• Insurance (for business)
• Office or object cleaning
• Telecom (business service providers and network architects)
3Furthermore, they have in common that they all create quotes during the sales process. The next chapter will discuss customer-driven sales processes in more detail.
3 One of the interviewed organizations (ORG2) defines itself as a Network Architect. Several names can be used. Nevertheless, these kinds of organizations design and build communication networks.
9
1.2 Sofon
This section will make the link between Sofon, customer-driven sales processes and this research.
1.2.1 A short introduction about Sofon
Sofon is selling software for customer-driven sales processes. At Sofon this is called
‘Guided Selling Software’. From a functional viewpoint, this software can be best placed between CRM and ERP systems. Creating quotes is an important feature of this software. The philosophy at Sofon is to make a quote motivated by customer requirements instead of product attributes.
The history and future of Sofon’s software will be discussed in the following sections.
1.2.2 Sofon’s software-evolution from a functional viewpoint
Let us start with the functionality of Sofon’s software. The software development at Sofon started with a product configurator. A valid configuration of a certain product can be made with this kind of software. More vendors of this software can be found in Appendix III.
Sofon’s software has shifted to more service-oriented functions after several extensions (see Figure 1). The current released version can make all kinds of documents, such as contracts, faxes, manuals and quotes.
Figure 1. Software functionality made by Sofon
Time
Price-calculation Document-generation Product-configuration
Contract-configuration
Products
Services
Contract change configuration
Quote-configuration Project-configuration
10 1.2.3 Current and future software modules
Sofon has developed three software modules. Starting with the oldest module these are:
• The sales configurator: Calculations, quotes, contracts, orders, visualizations, and all kinds of documents can be made with this software.
• The E-configurator: This software allows customers to make quotes online.
• The sales organizer: This module can be used to manage the sales funnel.
More information about the sales funnel can be found in Appendix IV.
The exact content of these modules is not important. The picture illustrates Sofon’s chancing view on manufacturing software.
Figure 2. Products made by Sofon
Currently Sofon is developing a document management system. Sofon can serve a wider group of organizations with the new software. However, the general manager at Sofon does not know what these new groups want. An important part of this missing knowledge is what problems these organizations experience. Sofon likes to create maximum benefit for their current and new customers. Therefore, this research was started with the broader purpose of creating insights into future customer segments of Sofon. Furthermore, the general manager is wondering if there is a market for this next generation software. However, the latter question is not the direct goal of this research. The following section will give a further explanation.
Time
Sofon Sales Configurator Quote-generation via MS Windows PC
Products
Services
The year 2000
Sofon E-Configurator Sofon via internet
Sofon Sales Organizer Sales process management
Sofon Dossier Manager Dossier management
11
1.3 Context
This section explains in more detail why the research has been performed. Customer- driven Sales Processes and Sofon will now be explained. These two are combined with the research in the figure below (Figure 3). This can be seen as a small kind of conceptual model. The aim of this picture is to give an overview of the research context and the role of the research project itself.
Three parties are involved in this research:
1. Organizations with a Customer-driven Sales Process 2. Sofon, who develops sales supporting software (SSS) 3. The researcher
The aim of this research is to gain knowledge to innovate current and new software at Sofon. SSS and sales processes of organizations are investigated to achieve this goal.
Figure 3. Research context
World of Sofon
World of Researcher World of Organizations
Knowledge about problems Innovation in
software
Research
Legend:
Dependent variable Independent variable Intervening variable Moderating variable
Has influence on
Second importance Additional information (Sales) Supporting
Software Sales Process
Made by Sofon or another vendor
12
2. What is a customer-driven sales process?
This chapter will describe customer-driven sales processes. The differences with other sales processes and the role of quotes will be explained.
There are different types of sales processes. For example, a simple retail store has a marketing campaign. Customers react and take the initiative to go to the store and take some products to the cash register. In a more complex situation, the customer wants an article that is not available. The store orders the article and the customer buys the product from the store. In the same way, a customer can order a product with certain options. Every order can be customer-driven.
These kinds of sales processes are relatively simple. The expected outcome of the sale is clear from the beginning for both parties. Even if the customer can choose some options. Take for example customer-driven manufacturing. The outcome is a collection of previously known situations. The product, or collection of options, is constant and only the price is a negotiable variable during the sale. In other words, the choice of the customer is limited and determined by the supplier.
Within a customer-driven sale, the product is more unique. The product or service is often customer-specific, tailored to the customer’s need. Often, both parties do not know what the exact needs are. The costs and/or exact needed product are often unclear at the beginning. In addition, for the selling side it is not clear what the customer wants or needs (see Figure 4 below).
These kinds of sales processes are customer-driven. The customer requests a quote.
This can also be named a proposal. Only the word quote will be used in this report. In some way, the selling organization needs to know the customer situation and desires.
At this moment, an agreement on paper is needed. Often, a final agreement has been
made after several iterations of change in the quote. Afterwards, both parties agree on
paper and the deal is closed. Note that the selling organisation offers a product for a
specific price under certain conditions. The customer can determine all these
conditions. In all situations, the quote gives a formal shape to the commitments from
supplier to customer and vice versa.
13 It is important to realise that a quote has a very important position in a customer- driven sales process. Note that the quote is not a single goal, but the process to come to a quote is. Prospects
4can request quotes from several organisations. It is not unusual that the prospect decides to buy a product depending on the quote. As a result, the quote has three functions:
1. Make an attractive offer.
2. Get acquainted
5with each other and come to an agreement.
3. Formalize for both parties what they have committed themselves to.
Figure 4. The functions of a quote.
4 For more information about this term, consult Appendix III.
5 Origin Latin accognitare, from cognoscere ‘come to know’ (Source: Compact Oxford English Dictionary)
Quote
Attractive offer
Commitments Commitments
What does the customer want?
What can the seller
offer me?
14
3. At which customers is Sofon aiming?
In chapter one it is mentioned that Sofon wants to serve a broader market. Sofon is aiming at organizations that have a customer-driven sales process. The previous chapter has discussed this subject. This chapter will provide all details that are considered during the selection process of Sofon
6.
Let us start with two variables that are important within the research design.
1. The size of the organization: The revenue yield of Sofon’s software is linked to the size of an organization. Therefore, in general, Sofon has chosen to sell to organizations with more than one hundred employees.
2. The organizations have to be located in the Netherlands. The current customers are mainly located in the manufacturing industry. With the new generation software, Sofon aims at new branches of industry. For example, financial services and energy providers. Launching a new product in combination with entering a new market leads to undesired risks.
Besides the two variables above, there are more ways that Sofon uses to select their customers. To give some examples:
• As named at the beginning of this chapter, the sales process should be customer-driven. The customer should determine the final state of the sales process. In other words, the customer makes the product unique.
• Kind of product or service the particular organization is selling. The product should be customer-specific, but not too unique. Too unique products are not interesting to model.
• The sales process has a high information intensity.
• The sales process has manual parts. In general: Completely automated processes cannot gain improvements by using more information technology.
• The organization has a back office (BO) or department, which processes the information until a delivered product or service.
• Level of ICT maturity: Complete automated sales processes are difficult to improve by Sofon’s software. Therefore, there should be space for improvement. For example, none or outdated systems.
• The installed base: The currently used software.
• High total cost of ownership (TCO) of current software.
• Financial aspects as turnover and a financial healthy organization.
It may be clear that it is not realistic to check all of the above issues. However, investigating organizations that cannot or will not use Sofon’s software will not give any valuable information. How then, should these organizations be selected? This will be explained after the next chapter (section 5.3.1). First, the statement of the problem will be discussed.
6 Be aware that this research does not the aim to find new customers (directly) for Sofon. Participating organizations stay anonymous. Besides this, the research has more value for Sofon than some new prospects or customers.
15
4. Statement of the problem
De Leeuw states that a problem statement should answer the follow questions : What do you want to know, for whom, and why [Leeuw, 2001]? In chapter one “the for whom” and “why” are explained. This chapter will continue and discuss the “what to know” question.
4.1 Research Objective
Sofon BV provides Sales Supporting Software (SSS) for customer-driven Sales Processes. The general manager of Sofon is wondering
7which problems are perceived by managers of customer-driven sales processes. If these problems have been described, Sofon BV can develop new software that will meet more specific customer demands. In short, the research results will be used for the software innovations for the coming years
8. The research objective is:
Find out which problems are perceived by managers of customer-driven sales processes.
4.2 Problem statement
The research objective can be translated into the following problem statement:
What
9problems do current and potential customers of Sofon BV perceive in their sales process?
The focus is on the effectiveness of software in sales processes. Can software be seen as a problem? In other words; does the current software add value or is it a barrier in the sales process. However, problems in sales processes are not always related to the software. Various other variables have direct and indirect influence. Furthermore, the term organizations will be used in this report. This term denotes Sofon’s customers and potential customers.
7 This is a time-related problem (and thus not a structural problem). The conclusion given by this research project will be used for the next two to three years of software development.
8 Sofon BV does not need a competitor or market analysis at this moment.
9 What-questions are exploratory [Yin and Campbell,1994].
16
4.3 Research Questions
From the problem statement and research objective above the following five sub- questions can be derived:
1. What is a (customer-driven) sales process?
2. Which problems, related to information processing and management, are identified by these organizations?
3. How can these problems be identified?
4. What should be investigated to identify these problems?
5. Which software features and recommendations can be advised for Sofon’s software?
Question one has already been discussed in chapter two. Chapter nine will present the results for question two. Question three will be discussed in chapter five and question four will be discussed in chapter six. The answer to question five will be presented in chapter ten.
4.4 Constraints
The research project will have the following constraints:
• The research project has a time span of 16 weeks. It started on Monday 24 October 2005 and will finish at the latest on Friday February 10
th. In December 2005 this was changed to Friday January 27
th.
• Intellectual Property (IP) of Sofon BV (especially new products) has to be
kept secret.
17
4.5 Methodology
The research has an exploratory nature. The main objective of the research is to create new insights. Furthermore, no hypotheses were tested or comprehensive variables were examined. This research can be used to develop theories, which can be tested in future research projects.
The methodology is based on Sekaran’s Research Methods For Business [Sekaran, 2003], an overview of which is included (Appendix VI). The new research process scheme is presented below (Figure 5).
Figure 5. Research methodology
This methodology has been worked out into a more detailed approach. Section 5.2.
summarizes this approach.
Zoom out PRELIMINARY
DATA GATHERING Interviewing Literature survey
PROBLEM DEFINITOIN Research problem
delineated
CONCEPTUAL MODEL Research areas
defined
RESEARCH DESIGN
DATA COLLECTION, ANALYSIS, AND INTERPRETATION
GENERATE THEORIES Hypothesis and
theoretical framework Report Writing OBSERVATION
Broad area of research identified
Zoom in
18
5. Research Design
The central question in this chapter is not what are these problems, but how to find these problems? The common design-issues will be discussed in section one. After this, the approach for the whole research will be summarized in section two. Section three presents some more detailed design issues.
5.1 Common design issues
The following sections will discuss the considered issues of the research design.
5.1.1 Purpose of the study
The goal of the research is exploration. It is unknown what kind of problems there are and where they occur. Besides this, the study creates insight into (relatively unknown) Customer Driven Sales Processes. Furthermore, it useful to know if problems are structural or incidental.
5.1.2 Interference
The extent of researcher interference has been minimalized by using open and non- suggesting questions during interviews. Closed questions were only used to confirm answers given.
5.1.3 Study setting
The setting was natural, to allow for an exploratory (case) study (non-contrived, thus not a lab experiment). One or some sales managers, sales director, commercial directors, or other people who have a clear overview of an organization’s sales process have been interviewed. The data was collected through semi-structured interviews. The interviews are approximately one hour (or less more). This is a relatively short time span per interview. Therefore, this research cannot be labelled as a case study.
5.1.4 Unit of analysis
The analysed units are organizations with a Customer-driven Sales Process. One or
two employees who have insight in this sales process will be interviewed. This
collection of organizations should consist of existing and potential customers of
Sofon. The group of organizations will be specified in chapters six and seven.
19 Furthermore, the focus is on the activities (and software) between quote request and order processing. All activities following the order exchange between the sales department and operations exceed the focus of this study.
5.1.5 Time horizon
The research provides a snapshot of the field; it is a cross-sectional study. This means that several units were investigated once. The organization visits had to be finished before the end of 2005. This leaves seven weeks for approaching (calling) and visiting the relevant organizations. As organizations are often changing continuously, it is possible that the relevance of information gathered has a short life span.
5.2 Summary of the Approach
Yin and Campbell state that a research design is a logical plan for getting from here to there [Yin, 1994]. A short summation of the steps to get there is as follows:
1. Introduction to Sofon and the research problem
2. Investigation of a broad area of literature and other sources (for example competitors of Sofon) relevant to the problem.
3. Problem definition 4. Conceptual model 5. Research design
a. Definition of the customers that Sofon is aiming at b. Selection of organizations and interviewees
6. Contacting of organizations willing to cooperate in the project 7. Visitations and interviews (i.e. data collection)
8. Case reports of visits
9. Contacting of interviewees for additional questions 10. Analysis of reports
11. Induction, generalization and summary of findings 12. The final write up of the report.
The first step, step one, is described in chapter one. Step four is discussed in chapter
four. The next chapter will discuss the fifth step. Issue 6a has already been discussed
in chapter three. Section 5.3.1 discusses step six a. The sections after 5.3.1 discuss
steps seven to ten. The chapters eight and nine discuss the step eleven. Step twelve is
discussed in chapter nine.
20 In a picture the research area, in combination with the time variable, looks like a funnel (see Figure 6). The top represents the beginning of the research and the bottom represents the end of the research. The space in pyramid reflects the investigated area.
Figure 6. Time and research area visualized
Observe broad area (seek literature)
Define customers
Do interviews (the case reports)
Analysis
Conclusions
Time Area of research
21
5.3 Detailed design issues
This section answers the question of how the research data has been collected. First, several data collection methods are considered, following Sekaran who discusses several data sources and data collection methods [Sekaran, 2003]. Below are two hierarchical lists with the arguments for including or excluding sources and collection methods.
• Data sources
• Primary sources of data
o Individuals, respondents Æ chosen, see below.
o Panels: have not been chosen because the research does not focus on the situation of one organization. The research does not look for one unanimous answer. The different situations in practice form the units of analysis.
o Focus groups: See panels.
o Unobtrusive measures: Not feasible.
• Secondary sources: Although these sources would have saved some time, they cannot give any answer to the problem statement.
• Data collection methods
• Interviews
o (Semi-) structured Æ Chosen, see below.
o Unstructured: Not chosen because the aim of the interview was clear. Besides this, the interview results should be comparable.
o Face to face Æ chosen, see below.
o Telephone interviews: Not chosen because the duration of the interviews was at least one hour. In addition, no visual aids could have been used.
o Computer-assisted interviews: Not feasible.
• Questionnaires
o Personally administered questionnaires: This is not an easy way to get an insight into one’s process structure. Beside this, follow up questions would not (easily) be possible.
o Email questionnaires: See above
From the following main requirements, a choice was made:
• A longer interaction is needed to get a clear understanding
• The effort should be minimal for the interviewee
Besides above possibilities, indept case studies or indept interviews were a choice.
This was chosen because it would request extra effort from the cooperating persons. A
semi-structured interview at the location of the concerning organization did meet
these needs. In addition, indept interviews allowed for visual aids.
22 Another reason for choosing interviews is that the gathered information is more reliable. With questionnaires, people are less accurate in their answerswith giving an answer. Furthermore, it is easier to find the real problems because it is easier to ask specific follow-up questions.
5.3.1 Selecting organizations
Finding existing customers of Sofon was not that difficult. These organizations can be found in a list, including the contact persons. Most of the existing customers are industrial manufacturing organizations. As mentioned in the first chapter, Sofon wants to serve a broader group of organizations. Thus, the first question at this stage was which branches would be potential customers for Sofon’s software?
5.3.1.1 Selecting branches
To answer this question first a list was studied with all Dutch organizations grouped in branches. This list is an extension of the International Standard Industrial Classification (ISIC) and is approximately 120 pages long. The extension
10is called
‘Businessview Bedrijfsindeling 1995’, in short DVBI’95
11. Therefore, the six layers above the branches were studied first. These layers have a hierarchy, which divides the branches in sections and (sub) groups.
From here, interesting groups of branches were selected. Chapter two and three contain the basic requirements for selection. In addition, the general manager and Director Business Development of Sofon examined the first results.
Groups were selected on the basis of amount of employees. To reduce the total amount, the minimum amount of employees was set at 500. Although in chapter three states that more than 100 suffices, the amount of organizations with more than 500 employees is better workable. Besides this, it gives a larger margin to prevent an incorrect part of the selection. This query resulted in 410 organizations (see Appendix IV for the exact queries). The Dutch chamber of commerce
12was considered but not selected. Unfortunately, their site could not process large queries.
5.3.2 The right number of cases
Although there were some duplicates among the 410 selected organizations, this number was too large. Some binary variables could be examined. However, these kinds of variables did not elicit the response covering the research objective. Malhotra indicates that quantitative research has the sample number of non-representative cases [Malhotra, 1999]. This means no claims can be made about the whole population.
10 The ISIC has been extended with the European Union, and the list of this study was extended with the Dutch Chamber of Commerce and the extension of this list is the DVBI’95.
11 Note that 95 is the year of version 1.0. For this research version 7.0, released in November 2004, has been used.
12 http://www.kvk.nl
23 Hence, the results are not necessarily true for all Customer-driven Sales Processes.
Miles and Huberman (page 29-30) state that 15 cases is enough for a quantitative multiple case study. Nevertheless, there was a risk that persons would cancel their appointments. Therefore, this study aimed at having appointments with more than 15 organizations, to be sure of a useful number of case reports.
5.3.3 Approaching organizations
With a list of possible organizations, a few steps were needed before an interview appointment was made. These steps will be described in the following section. Then, the interview and processing data will be discussed.
5.3.3.1 Selecting and finding the right persons
For every listed organization, the follow steps were taken before making an interview appointment.
• Check if the organization has a web-site. Important questions could often be answered from the website. Which products are most customer-driven and which department should be contacted.
• Find a telephone number on the site or in a business directory.
• Find the best person for the interview. It was assumed that the sales director or commercial director would be the right person. Several times this person was spoken to and he or she redirected the request to another person of the sales department. In other cases, it was not clear to the person on the other side of the line who the right contact person would be. In all cases, the criterion was that the person should have a good insight into the process of the creating of quotes and the accepting of orders.
Approximately 90 organizations were called. At the end of this stage, 27 appointments were made. It has to be taken into account that this number includes two existing customers of Sofon, who were contacted directly without going through the above selection procedure.
The 27 organizations can be divided into the following branches.
IND Industrial manufacturing: 3 111 * AC Advertisement companies (publishers): 3 111 ENE Energy selling companies: 5 11111 **
TEL Suppliers of fixed and mobile networks: 3 111
INS Insurance 5 11111
CLE Cleaning 4 1111 ***
SoH Services (or hybrids) 4 1111
* (one is medical equipment)
** (later two appointments were cancelled)
*** (one is industrial cleaning)
24 Approaching the organizations to ask for cooperation was carefully monitored to include at least three or four organizations of the same branch.
5.3.4 Interview Preparation & Techniques
After the phase of making appointments the organization visits started. This involved the semi-structured interview, which will be discussed in this section.
A standard checklist with several themes and issues was made for the interview (see Appendix VIII). Also an A3 poster was used as a visual aid. This poster contained possible departments within and related to the sales process (see Appendix IX). The activities and communications have been recorded in this paper. One major advantage of the poster was the space for notes and the visualization during the interview. As an extra preparation, before each interview the organization’s website was revisited. This often saved time during the interview itself.
An indication of what was feasible within approximately one hour emerged during the first interview (see Appendix ORG1). After this interview the content for a one-hour interview was determined. This first interview can be seen as a pilot interview.
A few small improvements were made after several interviews. For example, many organizations had several customer segments and products. Often the sales process was different for these segments or products. Therefore, an initial question like:
Can you give a short introduction about your organization?
Was later been changed into:
Can you say something about the product you are selling? Or, are their differently organized sales processes within the sales department?
Such focusing questions saved a lot of time. During the interview, mostly open (unbiased) questions were used. After a response to an open question, some closed questions could be used to ensure that the answer:
• Was interpreted correctly
• Was complete
• Did not consume too much time in relation to the complete interview The following questioning techniques were used [Sekaran, 2003]:
• Funneling
• Unbiased questions
• Clarifying questions
• Taking notes
25 5.3.5 Interview structure and questions
Every interview started with an introduction. After this, the focus was set, drawing on the first interviews. Organizations often sold more than one product and/or to different segments of customers. Therefore it was possible that the sales process structure was divided into several specialized processes for one product (or customer segment). To prevent an unnecessary complex interview a suitable focus was set at the beginning. Often this reason was explained first to prevent a very long introduction about the product.
After this the poster was used with the first question:
What happens if a prospect requests a quote?
After this had been discussed the second question followed:
What happens if the prospect says yes to the quote?
Often this resulted in one flow of documents, other communication (i.e. face to face, e-mail), activities, departments and persons. If this was complete and clear, some numbers and time indications were asked. Problem areas and bottlenecks were discussed. Furthermore, time allowing, some optional questions were on the checklist.
These questions gave insight into how the sales trajectory was organized.
After the structure of the process a software inventory was made, the advantages and disadvantages of the software were discussed. Subsequently, the following question was asked:
Could you tell me about any advantages and disadvantages of these software systems (or packages)?
This question gave insight into which software tools were used within the sales process. Besides this, the opinion of the interviewee about this software instantly became clear. The use of MS Office suite was often omitted because most interviewees perceived this software as necessary or useful. Most interviews were conducted at the end of the year 2005. From these results two additional questions were formulated.
1. What would you like to improve within the sales process?
2. Which information are you missing within the sales process?
The sequence of these questions is no coincidence. The broad question is asked before the narrower question. These questions gave a good insight into the problems in general and the software problems specifically.
The four organizations visited in January 2006 were used as a control group.
26
6. Conceptual Model
Miles and Huberman claim that qualitative reports often end with a conceptual framework, instead of beginning with one [Miles, 1994]. Although this may be true, it is useful to have a clear view of the investigated area. Sekaran describes a theoretical framework as the conceptual foundation of the research [Sekaran, 2003]. For an explorative research as this, it is unknown what is important. Therefore, a viewpoint on the problem statement and research is needed. This viewpoint will be made in this chapter.
Customer-driven sales processes are the investigated area. More than one unit is analyzed, thus these contexts can be compared to each other (see Figure 7). This is done with processing times, numbers of quotes and orders. The research goal is uncovering unknown problems. The question of how to find these problems has been discussed in the previous chapter. The question of what these problems are is the focus of Figure 7.
As the sources of the problems are unknown, some sub areas could be investigated.
As stated in chapter four, the problems should be related to software and information processing. Hence, these areas are the expected areas of the problems, without naming concrete problems. The following areas are chosen:
• Sales process structure (see also Appendix IX).
• What does the workflow of the sales process look like?
• What does the information flow look like?
• What are bottlenecks or problem areas?
• What would one change in the existing situation?
• ICT: What are the advantages and disadvantages of currently used software?
• Information: Is there any information missing?
Of course more areas could be selected. This could prevent overlooking other unknown areas within and outside the context (see Figure 7, variable X and Y).
However, many research areas have disadvantages. The three most important disadvantages are:
• Time. Time is limited and should be used wisely.
• Overviews. A too wide research area will result in too much data.
• Less added value. What will be the added value for a software developing company?
As stated in the last issue, a software manufacturing company is more interested in software related problems than in, for example, psychological factors. As mentioned in chapter two, the focus is on software.
Besides the areas mentioned above, the problems can directly be investigated: the
problem area can be questioned directly. However, customer-driven sales processes
27 may function very good. This information can also be a source of learning more about the problem situations.
Figure 7. Conceptual model
Context ORG A Context ORG B
Context ORG N
Problems within a customer driven
sales process
Time and amounts
Existing Sales Process Structure
Desired Situation
Missing Information ICT
(Software)
Legend:
Unknown area Investigated area
Is connected with
Second iteration Variable Y
Variable Y
28 By studying the context of the problem, the findings will have a foundation. The context is the relation to the reality. Without contextual embedding, these findings (about problems) would be loose facts, which could give the risk of comparing apples to pears. Moreover, research data would be superficial without a context. Often the research area is investigated in a preliminary investigation. This provides enough theoretical framework to set a narrow focus. However, defining a narrow research area is not the aim of this study, as mentioned in section 4.2. Setting the focus too narrow would result in overlooking sources of problems (ingredients). On the other hand, a very broad focus would consume too much time and generate too much data.
To conclude this chapter: The research area can be divided into two areas. The first is
the context, which is the customer-driven sales process. The second are the possible
problem areas. Together they are most meaning, and can lead to solving the problem
statement of this study.
29
7. Processing data
The generated data was processed in two phases. The first phase was after every interview and the second phase was after all interviews. This chapter will discuss both phases.
7.1 Phase one
During all interviews several pages of notes were made. In addition, an empty poster of the sales process was used for note taking (see Appendix IX), as well as the interview checklist could be used to make notes. The interviewee provided extra information in several cases. Altogether, there were four different sources of notes and information. This information has been digitalized as fast as possible. This was the first activity that involved analysing the results. A standard template was used for the text. This had the following layout:
• Short information
• Branch
• Type of person
• Related interviews
• Date of the interview
• General introduction about the organization and their product
• Acknowledgements: Often these are the constraints of the content.
• Quantitative information
• General description of the software
• General description of the sales process
• Problems
• Missing information
• Desired situation
• Other issues
• Remaining questions
The list can be summarized as containing an introduction, context and problems.
Sometimes some optional extra sections were used. Some examples are:
• Flexibility
• Knowledge (spots)
• Documents
The posters and interview notes were translated into Actor Activity Diagrams (AAD).
These diagrams give a clear picture of the performed activities and involved systems,
departments or functions.
30 Above reports have all been approved by the interviewed persons. Before this, a call had been made about the two additional questions (see section 5.3.5). Often this was been combined with questions concerning the content.
7.2 Phase two
Meta-matrices have been made to create more insight into the collected data. One is for software and one for problems in general (See Appendixes XI & XII). In addition, all interviews have been labeled. The chosen labels and the results can be found in Appendix XIII. All three matrices, together with the interview reports (the contexts), constitute the foundation for chapter nine.
Miles and Huberman state that human affairs are not monolithic [Miles, 1994]. In
contrast, they are complex networks of conditions and effects. The key problem is to
formulate a general theory based on several of these unique networks. Chapter nine
performs this step.
31
8. Critical notes concerning the research data
This chapter will give a critical view on the quality of the research data.
8.1 Measurability
People are subjective in that they can exaggerate their experiences with software usage. Therefore small issues can weigh more then a structural problems. However, this does not disturb the research data because the perceived situation also provides information. A hypothesis can be that this is extremely low for all sales processes (or all business processes). Alternatively, outdated software is less satisfying, and thus a part of the interviewed organizations has outdated software. It is not possible to make these kinds of claims from this research.
The quantitative part of the research does not give an accurate image of the performance. The conversion rate from quotes to contracts has a relation with the qualifying phases of the sales funnel (see Appendix III). An example would be the conversion rate from quotes to contracts (or orders). This is not a good performance indicator. Although it can give an indication, the total revenue gives a better indication of the performance of a sales process. The reason for this is related to the qualifying stage
13during the sales process. It may be possible for organizations with a high conversion rate to make quotes for less valuable customers. In short, a few high value contracts are better than many less valuable customers. Therefore, the percentage of quotes resulting in a contract is not accurate.
8.2 Completeness
People do not notice all problems at the time of the interview. A perfect example is the question about which information is missing. During the interviews, no participants indicated that missing (or delayed) information is a problem within the sales process. However, most answers to this question indicated that this was a problem within the sales process. This issue can also relate with the measurability, because people do not experience everything as a problem.
In general, it is realistic to assume that an interviewee can have forgotten some software, activities or other relevant issues. Therefore, the interview reports could be incomplete. However, this research did not have the aim to collect all kinds of possible problems. The focus was on the experienced problems.
In addition, the time limitation of the interview itself could have been a factor.
Although there was enough time for most subjects, some parts could have been discussed in more detail. For example, the exchange of information between the sales
13 During this stage the customers shows interest. On the other side, the company decides or the customer is profitable or not.
32 department and operations. However, to find the real problems it should be better to contact the operations department directly. The problems perceived by operations do not have to match the problems of the sales department.
Often only one person has been interviewed. Although the research was aimed at managers, many organizations have several managers within their sales process. This inconsistency can affect the completeness of the research data.
8.3 Other issues
For several organizations their sales process was in a transition phase. Others were planning or implementing new software. Most organizations change continuously. As mentioned in the research design, this research provides a snapshot of the field
14rather than that it presents several measurements over a time interval.
Miles & Huberman mention three sources of analytic bias:
1. The holistic fallacy: see more patterns than there really are,
2. Elite bias: to predominantly have data from articulate, well-informed informants at the cost of data from less articulate informants.
3. Going native: losing one’s own perspective and ability to place cases in the same group. Co-opted into the perception and explanation of local informants.
These three sources have been considered during the data analysis. Furthermore, Miles and Huberman give thirteen tactics for testing or confirming findings [Miles, 1994]. These will be discussed below.
1. Checking for representativeness a. Positive, amount of units
b. Negative, amount of time (A hour or less more) and amount of interviewees within a case. There is also a high degree of product uniqueness.
c. Although all interviewees had a good insight into the sales process, not all interviewed persons were managers (see Appendix XI).
i. Fourteen managers have been interviewed.
ii. Four managers of supporting departments have been interviewed.
iii. Five internal sales employees have been interviewed.
iv. Two account managers have been interviewed.
2. Researcher effects
a. Of researcher on the case: minimal (most open questions or funnel technique).
b. Of case on the researcher: misleading information may be given.
However, a high enough number of informants was included.
Negative: only one method of data collection was used.
c. There can be an observer’s bias. Incorrect interpretations could be made. For example, instead of observing symptoms, conclusions may
14 One shot means a single snapshot, instead of several measurements with a certain interval.
33 be recorded. On the other hand, questions could be misinterpreted by interviewees. Two measurements are taken for all interviews:
1. Record the interview as close as possible (on the sentence and word level) in an interview report.
2. Have the interview report reviewed by the interviewee.
In case of doubt, interviewees were contacted by e-mail or phone.
3. Triangulating
a. One method, one data source and one data type is used by one researcher. For the quality of the data it would have been better, if there had been multiple sources.
4. Weighing the evidence
a. Two questions were collected later in most organizations + b. First-hand data +
c. The interview contains statements – d. Collected in a natural setting +
e. Respondents were alone with the researcher + 5. Checking the meaning of outliners.
a. When looking at the problems, products or quantitative data, no real outliners can be found.
b. Atypical cases, atypical settings, unique treatments or unusual events.
6. Finding extreme cases can lead to surprising answers.
a. When looking only at numbers, three extreme cases can be given:
i. ORG5 is an extreme case: almost no quotes are made.
ii. ORG6 is an extreme case: almost 50.000 quotes
iii. ORG24 is an extreme case: around 100 customers (this is low) 7. Follow up surprises.
a. A surprise is that SAP has disadvantages (not user-friendly), but the other ERP systems do not have disadvantages at all.
b. Almost no software, except MS Office, is used at ORG19.
8. Looking for negative evidence
a. ORG14 and ORG16 are almost identical, even though their problems do not match.
9. Making If-Then Tests
a. These tests are not explicitly carried out.
10. Ruling out spurious
15relations
a. No third or intervening variable is found for the conceptual models in the next chapter.
11. Replicating findings
a. Recommendation for further research.
12. Considering alternative explanations
a. Recommendation for further research.
13. Receiving feedback from informants
a. The interviewees reviewed the interview reports.
b. The interviewees did not comment on the final results.
15 Spurious originates from the Latin word spurius, which means false (Source: Compact Oxford English Dictionary).
34
9. Discussion research data
This chapter discusses the main findings of the interviews. As mentioned in chapter four, the focus is on software. First, some common findings will be mentioned. After this, some software and related problems will be discussed.
9.1 Common findings
What can be said after interviewing people/managers from 25 organizations? Of course, all organizations have a customer-driven sales process. And the interviews gave insights into that process. Two subjects will be discussed in this section. First, the differences within the unit of analysis will be discussed and second, the way software is used within the investigated customer-driven sales processes.
9.1.1 Differences within the unit of analysis
The research group of organizations is not a homogeneous group. First, the tender process and a common sales lead are different. Tenders are characterized by a formal way of communication and long running times (months). While common sales leads have less structured communication and often have short running times (hours or days). Furthermore, most sales processes include several iterations before a final quote is reached. In theory, for the tenders this is more intensive than for the standardized products of the insurance companies, but in some practical cases there may be no differences (for example ORG8).
Common sales leads Tenders
Both
Figure 8. Pie-chart with different types of customer-driven sales processes.
35 Second, some products are less customer-specific than others. Examples are:
• ORG6, which generally standardizes approximately 80% of their customers, spending more time on the last 20% of the customers.
• Both ORG10 and advertising companies. A complete marketing campaign is certainly customer-specific, while a certain (advertisement) space in a newspaper or magazine is not.
• Energy providers make most standardized quotes. However, their most important customers get a customer specific product.
Although this research does not focus on product uniqueness, it is an important variable to take into account, considering customer-driven sales processes.
Furthermore, this variable can differ within an organization. Even one of the cleaning companies
16had several standardized packages.
A third distinction can be made in the importance of the calculation and the importance of text. Although nothing can be said about the complexity, in some organizations this part costs relatively more effort than others. Some organizations have a calculation department or a specialized employee. Notice that insurance companies do not use term contracts. Instead, they use the word policy.
9.1.2 Software usage
This section will discuss how software is used through four themes. Before the next section introduces the problems, this section provides an insight into how the group of investigated customer-driven sales processes are organized and computerized.
9.1.2.1 CRM & ERP Systems
The offer of software vendors is very fragmented. Leading are the ERP systems of the vendor SAP (5 times). Two CRM vendors share the second place, both having three organizations. These are Siebel and Super Office.
A CRM system is used 13 times and an ERP system is used 16 times. Three of the 25 organizations did not have CRM and ERP systems and in eight organizations a CRM as well as a ERP system was implemented. From these numbers it can be generalized that CRM and ERP systems are often used in organizations with customer-driven sales processes. It should be taken into account that all interviewed organizations are relatively large (regarding the number of employees).
9.1.2.2 From external sales department to internal sales department
In ten organizations, one person or department (see Appendix XII) handles quote- and order-requests. The remaining 15 organizations have divided this flow in at least an
16 Namely: ORG8.
36 external sales department and an internal sales department. In the external sales department account managers serve existing and new customers. At some point, the internal sales department handles activities, for example, after the quote request, or after the order request. In these cases, employees of the internal sales department carry out administrative tasks. To perform these tasks information is needed.
Therefore, communication is needed between the external and internal sales department. In eight organizations, the account managers gather information and the internal sales department does the processing into quotes.
9.1.2.3 How quotes are created
All organizations used the MS Office suite except one.
17Quotes are stored as MS Word document. The text part of the quote is often made in MS Word. Almost one third of all organizations use MS Excel for small calculations to complete quote calculations. In ten organizations,
18an information system generates the quote. Seven organizations used standard templates
19, which are filled in and mutated. The remaining organizations make unique documents or have a hybrid solution, either automatic, through templates or a unique solution.
Templates 28%
Other 32%
Automatic
40%
Automatic Templates Other
Figure 9. The different methods for creating a quote.
9.1.2.4 From sales department to operations department
After the sales process the operations start. Although the operations department itself is outside of the focus of this research, the information exchange with the operations department is not. In industrial manufacturing this involves producing and delivering the product; in insurance companies this involves billing the premium and processing the claims; in publishing companies (Advertisement Company) this involves placing the advertisement. With the suppliers of fixed and mobile networks this subject was
17 This is ORG9.
18 Namely: ORG4, ORG5, ORG6, ORG9, ORG12, ORG14, ORG17, ORG18, ORG22 and, ORG25.
19 Namely: ORG7, ORG8, ORG10, ORG15, ORG16, ORG21 and, ORG24.
37 not discussed (ORG2, ORG3 and, ORG21). In addition, ORG13 had a similar sales process, which was divided into two separate sales stages. Several organizations automatically process this information exchange. These organizations can be divided into three groups:
1. The insurance companies (ORG6, ORG9 and, ORG18
20).
2. The energy providers (ORG4 and ORG22).
3. Others, namely ORG11, ORG12 and, ORG17.
All have in common that they make a standardized product. Furthermore, these organizations process the largest amount of quotes. It is interesting is to see that both ORG11 and ORG17 are product- and service-oriented. Logically, it is the number of quotes that determines the level of computerization.
The remaining organizations processes can also be grouped as follows:
• The industrial manufacturing organizations, which all process a small amount of quotes and orders
• The cleaning companies and ORG23, which all circulate the quote and/or other documents.
• Organizations with a tender process. This area is not completely investigated.
• Four organizations that cannot really be grouped.
The latter group can be divided into two publishing companies: ORG10 and ORG15, and two insurance companies: ORG14 and ORG16. All four organizations process the information for operations manually.
9.2 The collection of problems related to the software
In this section a common overview of the investigated customer-driven sales process is given, as well as the collection of problems. First, the results of Appendix XI will be discussed. This reflects the subject software within the interviews. Second, Appendix XII will be explained, reflecting problems and desired improvements in one collection. The latter appendix is most detailed. After this, a number of problems will be discussed in more detail. The next section will introduce a conceptual framework for further research.
This section reflects the step of induction. However, the collection of problems is too heterogeneous in nature to be generalized to the population. Although there is no real common problem, finding one problem was not the goal of this study.
9.2.1 Perspective from the software
Ten of the 25 organizations had problems with the user-friendliness of their software (see Appendix XI). Note that in five organizations this is only or partly based on the
20 No information systems were discussed in detail during the interview with ORG18. It can be assumed that these are processed automatically.
