Table of Content
1. Databases, Data Warehouse and Data-mining
1.2 What is a Database?
1.3 What is a Data Warehouse?
1.4 What is Data-mining?
1.5 What is Database Marketing?
1.5.1 Data Capture
1.5.2 Database Maintenance
1.5.3 Preserving the Privacy of the Individual Customer
2. Relationship Marketing and Customer Relationship Management
2.2 Defining Relationship Marketing
2.3 Defining Customer Relationship Management
2.4 Transactional Marketing vs. Relationship Marketing
2.5 Building/Forming Relationships
2.5.1 Customer Loyalty
2.5.2 Theory of Customer Lifetime Value
2.6 Implementing / Integrating CRM
3. Interconnection between Databases, Database-Marketing and (Customer) Relationship Marketing
3.2 The Need for Accuracy in CRM
3.3 Differences between CRM and Database Marketing
3.4 Privacy Issues of Data Collection
4.2 Deutsche Bank AG – Firm Portrait
4.2 Use of Databases and Analyse Tools
4.4 Database Marketing, Relationship Marketing or CRM
4.5 Theory Gaps
The evolution from transaction marketing to relationship marketing in recent years has resulted in a need for more rigorous databases and greater utilisation of current computerised tracking systems. Customer relationship management is a combination of people, processes and technology that seeks to understand organisations’ customers. It is an integrated approach to managing relationships by focussing on customer retention and relationship development. Organisations that successfully implement customer relationship management will reap the rewards in customer loyalty and long runs profitability. The impact of information technology in the relationship marketing context, such as database management systems, data warehouses, and data mining, is increasing, due to technological development which facilitates the storage and analysis of massive amounts of data. However, successful implementation is elusive to many organisations, mostly because they do not understand that customer relationship management requires company-wide, cross-functional, customer-focused business re-engineering. Furthermore, while organisations are collecting and analysing data, consumers are becoming increasingly concerned about the privacy of their personal information and information about their purchase behaviour.
The purpose of this project is to analyse the new marketing approaches, proposed by the literature, as shifts away from the traditionally dominant marketing mix model. The marketing approaches include database marketing, relationship marketing and customer relationship management, which carry with them additional analyse tools and theories for their effective implementation.
Firstly databases, database management systems, and data analyse tools such as data mining will be investigated to emphasise the necessity of these tools in database marketing, relationship marketing, and customer relationship management. The main issues dealt with consist of data, data collection, data warehouses, data marts and database marketing. It is intended to bring all these issues together to have a comprehensive overview considering modern information technology.
Secondly, relationship marketing and customer relationship management development and characteristics will be examined. The shift from transactional marketing to relational marketing will be described. Different viewpoints concerning the overall relationship marketing strategy will be shown. Moreover, linked and involved theories such as the customer lifetime value and customer loyalty programs will be brought in.
Thirdly, Chapter Three, will bring all information provided in Chapter One and Two together, focussing on the main cross over issues in databases and relationship marketing. The main issues discussed in this Chapter are the importance of accurate data used in customer relationship management, differences between database marketing and relationship marketing, and privacy issues of data collection.
Finally, in the fourth part of this paper, the contextualisation, an analysis of customer relationship management efforts in a worldwide operating organisation, Deutsche Bank AG, is evaluated and compared with literature. The differences in theory and practice will be drawn out. Discrepancies will then lead to gaps in theory and or practice which will be evaluated.
1. Databases, Data Warehouse and Data-mining
Knowledge of database technology increases in importance every day. Databases are used everywhere; they are key components of e-commerce and other Web-based applications. When a person purchases goods from his/her local supermarket, it is likely that a database will be accessed. The checkout assistant will run a bar code reader over each of the purchased items. This will be linked to database application program, which uses the bar code to find out the price of each the items from a products database. The program then reduces the number of such items in stock and rings the price up on the till (Connolly, Begg and Strachan, 1999). If the customer has a loyalty card, the purchased products will be sent to another database to store this information and to complement the profile of the customer affinities and preferences. Databases derive from the use manual file systems which were traditionally composed of a collection of file folders, each properly tagged and kept in a filing cabinet storing specific information (Rob and Coronel, 1997).
1.2 What is a Database?
The development of the internet, the use of personal computer and the fast technological evolution of electronics enabled companies and other institutions to gather significant amounts of information about their clients. The storage of this information is done by databases. A database can be seen as an electronic filing cabinet, a repository or container for a collection of computerised persistent data that is used by the application systems of some given organisations (Date, 2000).
A very similar definition but one more focused on the purpose of using the database as an instrument to support the organisational objectives is given by Beyon-Davis (1996), who states that a database is a structured repository for data. The overall purpose of such a repository is to maintain data for some set of organisational objectives. Most databases are built to retain the data required for the running of day-to-day business or activities of an organisation.
Kroenke (2003) adds in his definition, ‘a database is a self-describing collection of related records’. The term related record describes logically linked information. Back in the supermarket, the purchased items are linked with the price and number in stock, these pieces of information are related. Self describing means that a description of the structure of the database is contained within the database itself. That’s why any user can always determine the contents of a database just by looking inside it. Whilst Kroenke’s (2003) and Beyon-Davis’ (1996) definitions only have the term collection of data in common, Connolly, Begg and Strachan (1999) define a database as a shared collection of logically related data (and description of this data), designed to meet the information needs of an organisation and combine both definitions. Taking all definitions into account, a database can be understood as an electronic filing cabinet, to collect and share related data. The database is self-describing and designed to meet the information needs in an organisation.
As shown, the use of a database is beneficial to the use of a manual filing system. If the database is used without a management system, the advantage of using this database decreases. This is because local databases are not as effective as centralised ones. Figure 1-1 shows the difference between local databases (or so called file systems) and a database system.
Figure 1-1 DBMS vs. File System (Rob and Coronel, 1997)
illustration not visible in this excerpt
As shown, the database management system (DBMS) allows different groups to access the stored data, where in the file system each group possesses its own data without access to data from other departments. The DBMS mostly is a computer which routes all the information from and to the central database (Rob and Coronel, 1997). Date (2000) states that the DBMS’s overall purpose is to store information and to allow users to retrieve, update, and demand the information.
The most common and important benefits using a DBMS are listed below (Atzeni et al., 1999):
- The data can be shared. Normally, files are owned by people or departments that use them. But, the database belongs to the entire organisation and can be shared by all authorised users (Connolly, Begg and Strachan (1999).
- Redundancy of data can be reduced. There is no space wasted by storing the same information in different files because the file only exists one time on the accessible database (Burnett, 1997).
- Data Consistency. The risk, that inconsistencies may occur are reduced by eliminating or controlling redundancy. If a data item is stored only once in a database, any update to its value has performed only once and the new value is immediately available to all users (Connolly, Begg and Strachan (1999).
- Reduction of maintenance. The need for program maintenance is reduces because there is only one database used (Burnett, 1997).
- Database Systems are persistent. That is they have a lifespan that is not limited to single executions of the programs that use them (Atzeni et al., 1999).
- Increase of data security. The data is secured by a protection system which allows users to access only with a valid authorisation. In contrast, typically local data is not password secured (Burnett, 1997; Connolly, Begg and Strachan (1999).
On the other hand, disadvantages can be seen in the complexity and the size of the DBMS software. Another disadvantage is the initial cost for hardware and software to build up a DBMS (Connolly, Begg and Strachan, 1999). Burnett (1997) states that the centralisation of resources increases the vulnerability of the system, because all users and applications rely on the availability of the DBMS and a failure of any component can immediately stop all operations.
The use of a single database in each office or department is seen as ineffective, cost and time wasting. Organisations therefore use a database management system which provides a centralised data storage for shared files. Hereby the system reduces redundancy and provides various advantages. After the initial implementation, the system allows retrieving and manipulating data in an accurate and fast way (Rob and Coronel, 1997).
1.3 What is a Data Warehouse?
Most organisations possess massive amounts of data but do not have enough information. This discrepancy portrays quite accurately the situation in many organisations. Managers are often frustrated by their inability to access or use the data and information they need (Atzeni et al., 1999).
There are two reasons for the increasing gap between data and information. The first reason is the fragmented way in which organisations have developed information systems-and their supporting databases- for many years. Instead of using a database system with one central database, many organisations use different databases, which are based on a variety of hardware and software platforms, which make it extremely (if not impossible) for managers to locate and use accurate information. The second reason is that most systems are developed to support operational processing, with little or no thought given to the information or analytical tools needed for decision making (Hoffer, Prescott and McFadden, 2002). To get needed information, organisations use data warehouses which consolidate and integrate information from many internal end external sources and arrange them in a meaningful format for an accurate decision making process (Martin, 1997).
Kantardzic (2003) defines a data warehouse as a collection of integrated, subject orientated database, designed to support the decision-support functions, where each unit of data is relevant to some moment in time. A very similar but slightly different definition is given by Immon (1992) who describes a data warehouse as a subject-orientated, integrated, non-volatile, time variant data store to support management decisions. Based on these definitions, a data warehouse can be described as an organisation’s repository of data, set up to support strategic decision making.
Data warehouses function as stores for historical data of an organisation. The data in a warehouse are never updated but used only to respond queries from end users who are generally decision makers. In most cases, companies have several local or departmental data warehouses often called data marts. A data mart is a data warehouse that has been designed to meet the needs of a specific group of users (marketers, controller, etc.). A data warehouse includes time dependent categories of data such as old detail data, current (new) data, lightly summarized data, highly summarized data and metadata (directory or guide)(Kantardzic, 2003).
There are different methods used to prepare data in a data warehouse, such as Simple Transformations. These transformations are the building blocks of all other more complex transformations. Another method is Cleansing and Scrubbing, which ensures consistent formatting and usage of fields, or related groups of fields within a data matrix. A field is a single entry like the age, name, gender, and so on. Integration is the third method and it is the process of taking operational data from one or more sources and mapping it field by field onto a new data structure in the data warehouse. Finally, the method of Aggregation and Summarisation describes the simple addition of values along one or more data dimensions. Aggregation refers to the different business elements into a common total. For example, a company collects data out of the sales systems from their different branches. Firstly, the figures have to be transferred (transformed) from each sales system into the data warehouse. After cleansing and scrubbing (it is easier, if all sales systems use the same format) the data has to be integrated in the right, needed structure (sales, products, time, etc.). Summarisation, then adds up all daily sales to produce monthly sales. Aggregation is adding daily product sales and monthly service sales, to get the combined monthly total. Figure 1-2 shows the warehouse process to complement the explanation of this process.
Figure 1-2 The Warehouse Process (Connolly, Begg and Strachan, 1999)
illustration not visible in this excerpt
Information stored in a data warehouse, either new or old, then can be used to analyse the company’s current situation, customer preferences, competitors’ behaviour etc. and thereby to support decision makers. The special preparation of data in a data warehouse, by transformations for example, are the main reason why managers as well as analysts prefer a data warehouse as source for different presentation reasons such as data-mining or other types of analysing processes (Connolly, Begg and Strachan, 1999).
1.4 What is Data-mining?
With the growing use of computer systems there is a great amount of data being generated by such systems. Business, scientific institutions and government agencies have all dedicated enormous resources to collecting and storing data. In reality only a small amount of data will ever be used because the volumes are simply too large to manage or too complicated to be analysed effectively. Hidden, useful knowledge in this data has to be extracted, because this knowledge leads the owner to a competitive market participant. Data-mining is a process to extract or to discover this information in complex data stores through either automatic or manual methods. It describes the search for new, valuable and nontrivial information in large volumes of data.
Data-mining activities can be put in one of two categories, either in the Predictive data-mining, which produces the model of the system described by a given data set, or the Descriptive data-mining, which produces new, nontrivial information based on the available data set. Predictive data-mining has the goal to produce a model. This can be used to perform classifications, predictions, estimations and similar tasks. The goal of the descriptive data-mining is to gain an understanding of the analyzed system (through the model) by uncovering pattern and relationships in large data sets, which can be interpreted by humans (Kantardzic, 2003). To achieve these goals, several data-mining techniques are used. Kantardzic (2003) argues, that the techniques can be summarised as follows (see Appendix 1):
1. Classification – discovery of predictive learning function that classifies a data item into one of several predefined classes.
2. Regression – discovery of a predictive learning function, which maps a data item to a real-value prediction variable.
3. Clustering – a common descriptive task in which one seeks to identify a finite set of categories or clusters to describe the data.
4. Summarisation – an additional descriptive task that involves methods for finding a compact description for a set (or subset) of data.
5. Dependency Modelling – finding a local model that describes significant dependencies between variables or between the values of a feature in a data or in a part of a data set.
6. Change and Deviation Detection – discovering the most significant changes in the data set.
The choice of an appropriate technique depends on the nature of the data to be analysed, as well as on the size of the data set. Data-mining can be performed against either the data marts or the organisation’s data warehouse. Each of these techniques enables an organisation to find or ‘uncover’ information in collected data sets. This relatively new analyse tool delivers new and unpredictable ways of viewing and analysing data sets (Hoffer, Prescott and McFadden, 2002).
The application of data mining and the use of a data mining technique are illustrated by the following examples. Wal-Mart, after having mined its sales information by using the dependency model, found a link between the sale of babies’ nappies and beer. There was no existing, rational and obvious link between the two, until a data mining analysis illustrated that men are the primary purchasers. The beer purchase was an impulsive buy, while the nappies were the reason for the shopping. Therefore, Wal-Mart placed, as a result of the analysis, the beer close to the nappies, which caused beer sales to climb significantly (Brabazon, 1997).
The Mellon Bank (USA) used the collected information on existing credit-card customers to characterise their behaviour and to predict what they will do next. They used data mining to create a credit-card attrition model. As a result, they were able to predict which customers are likely stop using Mellon’s credit-card in the next few months. Based on these results, the bank was able to retain these customers using specific marketing actions (Kantardzic, 2003). Both examples show the effective application of data mining and its new way of analysing big data sets. Data mining is one of the fastest growing fields in the computer industry. The wide range of methodologies and techniques, which can be used to solve problem sets is one of the biggest strengths using this analyse tool. It is used to discover knowledge in huge data sets by creating and exploiting models which analyse the data (Rob and Coronel, 1997).
Huge quantities of data are collected in organisations today and many managers have difficulty obtaining the information they need for decision making. Two major problems caused this problem. Firstly, data are often heterogeneous and inconsistent as a result of the different approaches to the use and storage of data. Secondly most systems in organisations are developed to satisfy operational objectives (stock management) with little or no thought given to the information needs of managers. There has to be a difference made between operational and informational systems within an organisation. Organisational Systems are used to run the business on a current basis (check out at till) and informational systems are designed to support managerial decision making. Database management systems are used to run operational systems whilst data warehouses are created to fulfil the informational system requirements (Hoffer, Prescott and McFadden, 2002). The purpose of a data warehouse is to consolidate and integrate subject-orientated data from a variety of sources and to format those data in a context for making accurate business decisions. A data-mart is a data warehouse whose data are limited to fulfil the needs for decision-making of a specific user group (Connolly, Begg and Strachan, 1999). To discover knowledge in collected sets of data, stored either in a data warehouse or in a data-mart, analysing tools like data mining, derived from traditional statistics, are used. This process enables organisations to use this knowledge to enlarge their customer base knowledge and to complement customer preferences. Thereby organisations have the ability to offer their clients products and/or services which more accurately meet their needs. The process, which describes this approach, is known as database marketing.
1.5 What is Database Marketing?
Database marketing is a relatively new discipline within the marketing framework. Database marketing derived from a long chain of successful developments in marketing and selling products and services from the earliest advertising, through mass marketing, direct marketing, coupons, catalogues, and frequent-flyer clubs, to the beginning of real database marketing today (Hughes, 1996).
However, direct mail is really the foundation of database marketing. Direct mail professionals pioneered working with lists, containing huge numbers of customers their preferences and purchase history. When direct mail people began working with computers, they noticed that the evolving hard- and software available to them, enables a better, faster, and less expensive way of direct marketing. The database marketing was born.
The concept is based on the lifetime value of customers. The first sale is not the primary target of database marketing, but it is essential to add the customer to a marketing database and to build up a lifetime relationship (Jackson and Wang, 1996). Many companies already have an active database marketing programme and realised, that database marketing is a powerful approach to acquire and keep customers, by creating relationships with potential customers.
To define database marketing, Fletcher and Peters (1995), describe database marketing as the process, which firstly allows marketers to identify and collect information from the business, secondly to use that information to prioritise, select and segment marketing activities, thirdly to produce a more tailored and personalised message and finally to measure every effort made, money spent and evaluate what was gained. The aim of database marketing is to build a profitable individual relationship with each customer whereby the relationship should make the customer feel, that he/she is well recognised and receives personal service and attention only (Schoenbacher, et al., 1997).
Figure 1-3 Database Marketing Loop
illustration not visible in this excerptDatabase marketing is the process, where data, which is stored and updated in databases, is used, to fulfil a strategic marketing plan. The database represents the heart of the marketing decision-making process (Jackson and Wang, 1996). The database provides unique information for developing the strategic plan for a data-driven marketing program. The process is a loop of repeating actions as shown in Figure 1-3. The loop shows that the process starts again and again after updating the data in the database each time. Hughes (1996) states, that with the accumulated knowledge in the database, organisations are able to start a dialogue with their customers. Thereby the organisations provide the customer with products and information, and the customers provide sales, loyalty and even more information. The organisations carefully listen to their customers and respond to their ideas and wishes. Customers appreciate the recognition and respond with more sales, more loyalty and more information. This loop goes on for a lifetime and can make the customers resistance against offers from competitors.
However, Roberts (1995) argues that different disciplines of database marketing and a commitment to follow those disciplines on a day-to-day basis, as a company’s strategic plan, are necessary to gain success on a long term basis. These disciplines are, the data capture of relevant information, database maintenance in a format that permits effective retrieval and analysis, executing marketing programmes using the data and the privacy of individual information.
1.5.1 Data Capture
Data, collected by companies, comes from many sources. Sources can be described by type or by source. Information which refers to consumers age, gender, geography, income, type of residence, size and type of family, education, occupation, registered car ownership by brand and year, etc. are called demographic information. Psychographic information or behavioural information is also collected, such as consumer’s lifestyle, means what people buy, wear, their hobbies, how they choose to spend their leisure time, their sports, cultural interests, interests in general, etc. (Holtmann and Mann, 1992).
Sources can mainly be divided in internal and external ones. External data is defined as that obtained as a compiled list from outside the company. This could include outside lists, census data, and so on. Most data, and data this paper deals with, is sourced internally. Internal data sources will vary considerably between sectors, depending on whether you have direct contact with customers as a natural way of doing business or indirect contact (Tapp, 2000).
The engine of any database marketing system is a computerised database, or a data warehouse, that holds information about each customer and prospect. Which data to capture and how to do so are the key questions for the database marketing system. The decision should be taken against the lifetime profitability of the individual customer and then implementing long-term programmes that will maintain the customer relationship and increase the lifetime value (Roberts, 1995). Data capture should be on-line if that is remotely feasible. If it is not, a process must be developed that ensures timeliness and enforces quality control.
The quality of the information in the database cannot exceed the quality of the data that are input (Roberts, 1995). However, there are still important steps to be taken, such as cleansing and scrubbing or aggregation and summarisation before the data can drive programmes and analysis.
1.5.2 Database Maintenance
Maintaining the database is broader than then technical housekeeping details that keeps the system functioning and requires a set of policies and procedures. There are several characteristics of the database marketing system, which determine its effectiveness in the organisational context. The first characteristic is the locus of ownership and control. To avoid organisational barriers, caused through the view of data as property of each organisational department or unit, to database marketing success, it is necessary to bring all parties to the table to establish policies to ensure that all voices are heard and all needs are met in a system with reasonable priorities. Control should be kept by a data administrator who is responsible for implementing the policies and overseeing the functioning of the system itself (Roberts, 1995).
Access and security are obtained by the data administrator, who controls access to the marketing database and implements procedures that protect the integrity of the database and ensures the privacy of individual records. The transfer of data via Local Area Networks (LANs), Wireless Local Area Networks (WLANs), Wide Area Networks (WANs), glass fibre-optic or telephone lines, etc., requires great sensitivity to security issues as well (Roberts 1995).
The quality of data in a database is described by the term integrity. Quality stands for up-to-date data, which is cleaned, means there are no double entries of data (Roberts, 1995). Again, an advantage is a database system, which enables to store all files on one or more servers. Once a centralised system is given, every authorized person has access to all the stored files. A database system avoids widespread data on individual PC’s, which is not accessible and mostly not up-to-date (Connolly, Begg and Strachan, 1999). All data must not only be accurate it must also be available to those who need when they need it (Rob and Coronel, 1997).
To prevent conflicts over the ownership, maintenance and use of the database, Roberts (1995) recommends establishing a policy committee composed of top marketing managers and an operating committee of middle managers. For example, the data maintenance requires marketers, firstly to establish policies and procedures that define the marketing database as the official repository of information about customers and prospects and secondly to ensure the integrity of data whilst ensuring to provide the appropriate level of access to all users in a timely and effective manner.
More precise segmentation and more focused communication target the most profitable customers are the key goals of the database marketing function with a view to engaging in more one-to-one relationship marketing. Managers must ensure, that all marketing communications and promotions originate from and are executed by the marketing database (Roberts, 1995).
1.5.3 Preserving the Privacy of the Individual Customer
‘Consumer concerns about information privacy and physical privacy have led to further governmental intervention which has dictated an increase in legislation’ (Mitchell, 2003). The main concern, expressed by customers is the issue of data being swapped or sold. An example of unprofessional data handling is given by Nash (1998), where a woman from Ohio listed her preferences for certain products in exchange for coupons and free samples, which seemed like a good idea. That is, until she began receiving obscene letters at her house from a prison inmate who knew intimate details about her life, which he acquired when the prison was hired to enter data into a customer database. Therefore, data privacy should be stringently and sensitively managed to maintain customer’s confidence with the firm and create an atmosphere of trust to strengthen and expand customer relationship (Roberts, 1995).
- Arbeit zitieren
- Mathias Lüdtke-Handjery (Autor), 2004, How can databases improve sales?, München, GRIN Verlag, https://www.grin.com/document/39181