Direct Marketing and Campaign Management - Assignment Example

Direct Marketing In Marketing there are generally two approaches to reaching prospective s, “Mass marketing” and “Direct marketing”. Mass marketing uses mass media like TV, Radio, and Newspaper etc indiscriminately to reach out to a larger set of its target audience while direct marketing takes a focused approach. Direct marketing studies customers’ characteristics and needs and thus selects certain customers as the target for promotion. Different channels for direct marketing include Direct mail, Telemarketing, Email Marketing, Door-to-Door Leaflet Marketing, Voicemail Marketing, Coupon Marketing, Direct response television marketing, Direct selling. Direct marketing campaigns involve two important issues which are: 1) Cost – As the cost of per target promotion in direct marketing tends to be substantially higher than that in case of mass marketing, marketers must send their message to a limited audience only, as sending the message to general audience at large with very few of them likely to respond positively or getting converted may increase the cost manifold and thus cause a lot of marketing dollar wastage. 2) Customer irritation – As direct marketing communication tends to be longer and more involving i.e. seek active participation of the customer, than traditional mass marketing communication, thus unnecessary and unsolicited message may lead to customer frustration, resultantly eroding brand equity. Thus in direct marketing it is important to come up with a list of the customers who would be most interested in the product offering and thus will be most likely to respond positively rather than non-responding or responding negatively by avoiding further communication. Thus three issues assume great importance in direct marketing: 1) When and how often should we send the mails, make calls etc to the customers? 2) How many customers / prospects need to be contacted i.e. how many customers should be targeted? 3) Which customers should be targeted for which campaign (campaign segment matching)? Data Mining “Data Mining is the process of discovering interesting knowledge from large amounts of data stored either in databases, data warehouses, or other information repositories. Simply stated, data mining refers to extracting or "mining" knowledge from large amounts of data.” (Han and Kamber, 2006). According to Fayyad et al (1996), Data mining is an important process of discovering novel, implicit, useful and comprehensive knowledge from large amount of data. For Direct marketing it translates in to a description of likely buyers or responders and thus Data mining helps in securing better results, for a direct marketing campaign than a traditional mass marketing campaign According to (Berry and Linoff , 2009) data mining can perform the following tasks: Classification, Estimation, Prediction, Affinity Grouping, Clustering and Description and profiling For example - To develop marketing and service –delivery strategy, Harrah’s entertainment Inc., the gambling and hotel chain uses Data mining to target customers on individual basis for direct marketing. Using its sophisticated proprietary technology to mine customer database, Harrah’s found that 26 percent of its gamblers generated 82 percent of its revenues reported by Loveman (2003). For further details of the example please refer Appendix 1 example 1. Data Fusion The data, which data mining uses for discovering patterns and predicting response rates from different segments of customers, classifying the total set of customers in to likely to respond and unlikely to respond etc, are available either internally i.e. within different departments of the organization or externally i.e. have to be solicited from agencies etc. But the data available are captured by separate departments for diverse operational objectives and not necessarily keeping data mining objectives in mind. The data may be saved using different database platforms like Sybase, MySQL, SQL server etc and thus the diverse datasets have to be fused or integrate to perform data mining task. Data fusion is the process of fusing or merging the different data sets so that they can be used jointly. For example, if one data set contains media exposure information and another data set contains product purchase information, the process of data fusion would try to produce one data set with both media exposure and purchase information for all observations. To fuse the data, all procedures rely on a set of common variables, such as demographic information. Data Mining in Direct marketing Using historical purchase data, a predictive response model with data mining techniques can be developed to predict the probability of a customer response to a promotion or an offer, thus “mailing the right offer to the right person at the right time” dramatically reduces the cost of printed media, catalogue, phone calls since only most likely to respond prospects will be targeted on the basis of demographics, lifestyle, preferences and past buying behaviour which will be mined through data mining. Data mining technology can also help in keeping an eye on changing customer behaviour which facilitates the effective planning of direct marketing campaigns. Direct marketing may involve various campaigns thus data mining may also help in measuring effectiveness of marketing campaigns for different products or customer segments at different times. Applications of Data mining technique in “Direct marketing” Data Mining may contribute to the success of Direct Marketing through 1) Optimizing Response for a fixed budget campaign 2) Optimizing campaign profitability 3) Reaching the people most influenced by the message 4) Reaching the right segment at right time through right channel Techniques of Data Mining in Direct Marketing a) Response Modelling Direct marketing campaigns usually have response rates in single digits (Berry and Linoff, 2009). Response models are used to employ to provide an actual estimate of likely response to a given campaign. The models rank prospects on the basis of likelihood of response. Thus targeting or calling the people who rank highest in the list dramatically increases the effectiveness of the campaign and reduces the cost. For example - Using data mining, the Body Shop was able to reduce the number of catalogues it sent out in 2001 by 50 % after identifying a new 120,000 customer from its web store, and customer database. And revenue per catalogue increased 20 percent (Haag, Cunning and McCubbrey, 2003). It uses predictive analysis which is one type of data mining. Predictive analysis consists of forecasting and propensity analysis. Further details of the example can be found out in Appendix 1 example 2 b) Direct Market scoring modelling On the basis of some information available about the customers or prospects the scoring models can predict the likelihood of some desired or undesired behaviour of the customers and prospects. Malthouse (2003) says, A scoring model assigns every observation in a database a score indicating how likely someone is to engage in a particular behaviour. For example American Express continually mines a gigantic pool of computerized data on its 30 million credit card holders to create highly personalized marketing campaigns reported by Laudon & Laudon (2000). For further details of the example please refer Appendix 1 Example 3 c) Customer Life Time Value Analysis Model CLV analysis is a popular concept used for forecasting the value of a customer to a company. This analysis helps a company in classifying its customers in to groups of customers worth keeping and the customers worth firing in fact a company can rank its customers and determine how much cost can be incurred in retaining a particular customer looking at the customer’s life time value for the company. For example - Hallmark Greeting Card Company measured the effectiveness of direct marketing campaigns and customized them for all its Gold Crown customers. With SAS, Hallmark delivered information to business users in 10 minutes rather than the two days it once took, helping the company increase the number of customized direct marketing campaigns it can produce, as reported by SAS on its website. For detailed example please refer Appendix -1 Example 4 Challenges Some of the common challenges faced by Data Mining in Direct marketing are: 1) Shaw et al (2001) state that, Methods for data mining on a specific context are still unstructured and based mostly on judgment. This process need to be more structured in order to improve the productivity of data mining projects in direct marketing 2) One of the direct marketing key challenges is to research on which data mining methods to use under what circumstances. This means answering key questions before data analysis can proceed including; which customers should be selected as target for direct marketing? What data mining method should be used for targets selection? (Bose and Chen, 2009). 3) One of the important issues in data mining for direct marketing is how to protect the trade secrecy of corporations and privacy of customers contained in the data sets collected and used for the purpose of data mining. 4) The lack of marketing managers’ involvement to develop and use these models to support their decisions show the urgent need for systematic, simple and well-integrated framework to help them get more involved and capable to use data mining with little help from data analysis experts (Martínez- López, and Casillas, 2008). Conclusion: Overall, it can be seen that “Direct marketing” as a discipline is expanding and is enabling companies to increase customer retention, cross-selling and up-selling opportunities and also the chances of exploiting them, all of which ultimately contribute to higher overall revenues. At the same time, the pace of change can be challenging and the demands that “Direct marketing” places on its enablers, especially IT support, need to be carefully managed. This calls for tight integration of data mining and campaign management technologies. In this situation, marketers can invoke statistical models from within the campaign management application, score customer segments on the fly, and then can quickly create campaigns targeted to customer segments with the greatest potential. Appendix 1) Harrah’s entertainment, a major casino operator found that 26 percent of its gamblers generated 82 percent of its revenues by closely examining the data bases. Managers were also surprised to find the best customer that were not limousine high rollers but middle aged and senior adults who were former teachers, doctors, bankers and mechanists. It is found that this group responded better for $60 of casino chips offer than to a free room, two steak meal and $30 worth of chips offer. It deeply mines information into fine customer segments. It gathers increasingly specific information about customer’s preferences – then appeal to those interests. Reward customers for spending more, which increases their life time value. 2) The Body Shop International, UK based company prefer to send their catalogue only to those who will place orders for a product and so it uses predictive analysis which is one type of data mining. Predictive analysis consists of forecasting and propensity analysis. Forecasting finds trends and then predicts the future based on those trends. Propensity analysis uses statistical methods like regression analysis and clustering in combination with neural networks to determine the likelihood that a particular consumer will respond to an offer, or buy a product or service. Using data mining, the Body Shop was able to cut in half the number of catalogues it sent out in 2001 after identifying a new 120,000 customer from its web store, and customer database. And, not only that, revenue per catalogue increased 20 percent. 3) Hallmark’s treasure trove of detailed customer information and insights from predictive modeling and data mining are used to customize the content and timing of direct marketing vehicles. “SAS helps us understand what each individual’s buying profile is – what they like to buy and how they would like to be communicated with and rewarded for using Hallmark products,’’ says Cathy Richmond, Consumer Understanding and Insight Manager. Combining this detailed information with SAS predictive modeling and data mining allows Hallmark to customize its direct market­ing campaigns so each customer receives the right offer at the right time. Another insight the company gleaned with help from SAS was when to adver­tise and for what holidays. “We realized we were overspending in January and at Easter. We had traditionally invested in advertising earlier in the year and at Easter, but we weren’t doing anything at Father’s Day,’’ explains Jay Dittmann, Vice President of Marketing Strategy. Everything was based on historical numbers. The company experimented with more Father’s Day advertising and measured the results with SAS. “It was a huge suc­cess,’’ Dittmann says. Timing of targeted offers is also critical. “If somebody always shops late, we don’t want to contact them too early because they’re not even thinking of the holidays at that time,” says Vaczi. Likewise, the company likes to run tests – for instance, offering special incentives to see which shoppers will come into stores earlier in the holiday season. “With SAS, we’ve got the tools to do that and that helps with our sales plans and goals,” Vaczi says. SAS software allows power users to create predictive models and use data mining techniques, while less technical users can use SAS BI to easily query information without going through IT. Benefits With SAS, Hallmark delivers information to business users in 10 minutes rather than the two days it once took, helping the company increase the number of customized direct marketing campaigns it can produce. 4) American Express continually mines a gigantic pool of computerized data on its 30 million credit card holders to create highly personalized marketing campaigns (Laudon & Laudon, 2000). If, for example, a customer purchases a dress at Saks Fifth Avenue department store, American Express might include in her next bill an offer of discount on a pair of shoes purchased at the same store and charged on her American Express Card. The two goals are to increase the customer’s use of the American Express card and to expand the presence of American Express at Saks. 5) Credit Lyonnais had to apply results of sophisticated marketing research to entire customer base, SAS provided solutions for broad marketing strategies in areas including data mining, attrition, prediction and direct marketing and as reported by their executive“The SAS implementation enabled us to obtain very good results: We reduced our departure rate by nearly 40 percent.”Crédit Lyonnais also relies on SAS solutions to launch direct marketing campaigns by product and by customer, especially via e-mail programs. The latter use the affinity matrix, potential indicators and segmentation – a set of tools that clearly contribute to improving cross-selling and the number of services a customer receives. In addition, with SAS, Crédit Lyonnais has more than doubled the number of marketing campaigns and obtained better profitability, with a ratio 10 times higher on certain highly targeted e-mail campaigns. (Source SAS Website) References: 1. Berry, J., and Linoff, S. (2009) Data Mining Techniques, 2nd edition. 2. Bose, I., and Chen, X. (2009) ‘Quantitative models for direct marketing: A review from systems perspective’, European Journal of Operational Research, vol. 195, no. 1, pp. 116. 3. Fayyad, U. M., Piatetsky-Shapiro, G., Smyth, P., and Uthurusamy, R, (Eds.). (1996) Advances in Knowledge Discovery and Data Mining. MIT Press, Mento Park. 4. Han, J. and Kamber, M. (2006) Data Mining: Concepts and Techniques, San Francisco, U.S.A, Morgan Kaufman Publishers. 5. Haag, S., Cummings, M., and McCubbrey , J. (2003) Management Information Systems for the Information Age, 4th edition. 6. Laudon, K ., C., and Laudon, J., P. ( 2000) Management Information System: Organization and Technology in the Networked Enterprise, 6th edition : Prentice Hall. 7. Loveman, Gary. (2003) Diamonds in the data mine. Harvard Business Review, May, pp. 3-7 8. Malthouse, E. , D. , Iacobucci, B. Calder, eds. Scoring models Kellogg on Integrated Marketing, John Wiley & Sons : New York, pp. 227-249. 9. Martínez-López, F.J. and Casillas, J. (2008) ‘Marketing Intelligent Systems for consumer behaviour modelling by a descriptive induction approach based on Genetic Fuzzy Systems’, Industrial Marketing Management, vol. 38, no. 7, pp. 714-731 10. Shaw, M.J., Subramaniam, C., Tan, G.W. and Welge, M.E. (2001), ‘Knowledge management and data mining for marketing’, Decision Support Systems, vol. 31, no. 1, pp. 127-137. 11. Thelen, Shawn, Mottner Sandra, and Berman . (2004) ‘Barry, Data Mining: On the trail to marketing gold’, Harvard Business Review, Nov, pp. 25-32. Read More