Final database marketing class

Good job Marketing 455 students! The students presented the final case, a Harvard one, entitled, "Pilgrim Bank" which is a study in how to determine the most profitable customers for a mythical bank. Students analyze the data in stages and at first it looks like online banking does not make a customer profitable but then we need to take age and income factors into account and when doing so, online banking is a factor in profitability. Students run various regression models and come up with the solution with the most explantory power. The students did well and were able to understand and analyze this advanced case. Their SPSS skills have improved and they were able to get to the heart of the case with some legwork. We also discussed what other types of data Pilgrim Bank might want to use to better understand its customers, both internal and external and how the Bank can manage customer relationshps for maximized lifetime value. Database Marketer extraordinary, Rob Jackson, an interactive marketing advisory board member, came to class and helped provide insight to our discussion. Thanks, Rob!

Database class update

I am so proud of my database marketing class. We have spent the last several weeks working on some non-statistical techniques of data analysis such as RFM, neural networking, decision trees and cluster techniques. We have also introduced the class to data mining as a discipline and the SPSS Clementine data mining software. The students were so familiar with SPSS that it took a while to get used to Clementine. Although Clementine has an 'easier to use' graphical interface, the students liked the control offered through the traditional SPSS interface. Clementine is more of a 'black box' environment, although it makes certain functions easier than in traditional SPSS. Some of the functions that are easier include neural networking, creating lifts and gains charts and comparting multiple models. Clementine also includes advanced graphics capabilities. I like being able to get a graphic representation of how large a cluster grouping is and the relative importance of different variables in creating the clusters. The students handed in their data mining assignment today. It was great seeing them play with the program and the data and get a real life taste of data mining.

Database marketing used to support sales efforts

I haven't had much time since returning from the Online Marketing Summit to blog about my experiences. There was a lot of information relating to interactive marketing. From a database perspective, on e of the most interesting presentations was from Adam Needles at Silverpop. Adam talked about something that is called 'demand generation' which really is using database and web technology to manage the lead generation process for the sales force. I know when I was in B2B sales we always complained about the quality of leads from marketing. Just a few years ago, leads for sales from marketing were confined to purchased lists and trade show leads, many of which were unqualified. Telemarketing to pre-qualify helped improve the quality of the lead. The problem is that companies want that experienced sales rep to be given an actionable lead. Research shows that only about 5% of the sales process today really involves direct contact with the salesperson. Most of the sales process occurs before the sales rep gets involved, as is reflected in changes in B2B spending. There was a 19% decrease in spending for B2B marketing spend in tradeshow, print and digital but an uptick in content marketing through digital channels. B2B buyers now spend most of their time researching products online and leveraging the groundswell of social media to make their decisions. The successful B2B marketer today seeks to capture the B2B buyer's attention during this phase by drawing the buyer into the web site and once, there, keeping the buyer with meaningful content on products and industry issues and concerns. Everyone wants content today, ways to do things better and more effectively. Once a lead is captured at the web site, the B2B marketer's job it to nurture that lead. Now, through the internet and database technology, we can tell if a prospect that has registered at the site has downloaded a whitepaper, email a follow up to see if there are questions, send related content and nurture that relationship to the point where it is worth the salesperson's time to contact and try to develop a sale. The B2B marketer needs to make sure that the company is providing thoughtful content through twitter, blogs and other social media sites so the B2B buyer will want to visit the web site. Search engine optimization plays a critical role also in attracting that buyer during the evluation phase, before a sales rep is involved. From a database perspective, marketing can track leads and report the success of its program in a way that until recently was difficult to do. Leads can be segmented. Some can be discarded, some sent to the sales force right away and others nurtured through the process and marketing can report on its demand generation process just as sales has for years reported on its sales process. There are software tools to help automate this process. So although the power has shifted and the B2B buyer is more in charge of the process, the B2B marketer has more data to work with to try to nurture and report on customer relationships. Look for more and better automation technologies in the future, to integrate marketing and sales.

Segmentation and buliding the database

The last week and a half we have been talking about putting the marketing database together. We take internal information, external information and modelled data and create a database that really works for marketing applications. Internal information is transaction oriented, typically and includes information about what the customber has purchased and when. But transaction information only tells part of the story. We might not even have the customer's name if the product was shipped to a different person than a purchaser (b2b) or not have information about the household if just one person in the house buys our product (b2c). We need to enhance the database to get value and we talked about buying data from an outside data source to supplement our internal data. Typically, we send our raw data files to an outside source and they do a merge/purge and cleanse our data and append to our data records. We can then begin the process of modelling our data to identify our best customers. We talked about RFM (Recency, Frequency and Monetary Value) as a way of segmentation but that the RFM approach is only a starting point. Marriott uses a CAP model which includes capacity and propensity to purchase. The different types of segmentation that are discussed in the book are a good starting point for data anlysis but more sophisticated companies eventually develop their own proprietary segementation schemes.

Database Marketing Gets Serious!

The students successfully completed the first assignment with descriptive statistics, cross tabs and chi squared analyes. They learned that categorical data puts data in groups (think cross tabs) and that metric or continuous data is more suited to other types of statistical analysis such as t-tests, correlation and regression. In marketing, we like variation and the null hypothesis is usually what we don't want to have happen, no difference in counts for the chi square, no difference in means for a t-test and no linear relationship for correlation and regression. We are looking for a p value of under .05 for significance level. We had fun looking at live data and used the President's approval rating as a dependent variable in a simple regression. There was a regression relationship (F stat is <.05), 83% of the variance (Adjusted R sqared) in approval rating is explained by ONE significant variable (p < .05), unemployment rate. Interpreting the regression, the President's approval rating drops by about 6.6% for every one percentage point increase in the unemployment rate. The class is working now on creating a model to predict performance in our IM classes. They are looking to create a parsimonious model, that means as few variables with the most explanatory power, to explain an particular hypothesis.

Database Marketing: A New Semester

I am going to blog here about database marketing and on our NING social network for Marketing Technology. We began with a simple exercise called PC's unlmited to familiarize or refamiliarize the class with descriptive statistics in SPSS, how to create a new variable, perform cross tabs and other basic tools. Smart students will go back to this handout throughout the course of the semester for looking up tasks like how to create deciles and quintiles, etc. Students were asked to recreate the information listed and then to provide a marketing interpretation of the graph or table. For example, if most of the company's customers are men does it mean the firm should focus on men as customers or branch out to serving more females? We then had a review of direct marekting and linked direct marketing to database marketing through the use of customer information. It is the database that underlies most direct marketing campaigns that allows DM practitioners to measure, adjust and re-launch their campaigns. Finally, we concluded the two week period with a brief statistics review, focusing on mean, standard deviation, correlation and other descriptive tools like histograms and bar charts. An important concept here is categorical versus continuous data. Categorical data cannot be used to calculate means but can be used in a cross tabs. With continuous data and two groups, a t-test can be conducted.

Some final thoughts on customer service and Internet Marketing

Today was the last day of class in Internet Marketing 370 at Northern Illinois. We said goodbye to our graduating seniors and three students recieved the certificate in Interactive Marketing. Good luck to them and to all those graduation this semester. It occurred to me that talking about customer service at the end of class as we did could either give the impression the topic was important or that it was not important enough to cover early in the semester. For the record, customber service can make or break an internet company, or any company seeking to use technology for better, differentiated, customer service. The Rogers and Peppers marketing model of IDIC (Identify, Differentiate, Interact and Customize/Personalize) is, I believe, replacing older persuasion models such as AIDA (Attention, Interest, Desire and Action). Older models are meant for the mass marketing world. In the interactive and direct marketing world, it is important to differentiate individual customers or groups of customers and treat different customers differently. Differentiation is meaningful for targeting and attracting the right customers but becomes critical with customer service. In class we looked at the British grocery chain Tesco's and how the company uses segmentation to send out multiple versions of its coupon offers depending upon customer needs in implementing the Tesco' loyalty program. Retention is important because of the cost of acquisition, the potential negative word-of-mouth from an unhappy customer and also because of long-term customer profitability. This discussion on customer service is a nice transition to database marketing next semester, where we will begin to see how databases can be used for targeted, segmented acquisition, service and retention. Stay tuned after a short break. Class begins again on January 12 for the Interactive Program at NIU's DeKalb campus. We will be teaching database marketing, marketing technology, IMC and direct marketing.

Getting a job in Interactive Marketing

Kelli Hanson talked about places to look for a job in interactive marketing. She and I strongly believe in networking through professional associations, such as CADM, BMA, AMA and CIMA. Joining the groups these organizations have on linkedin is usually free and an easy way to start getting involved. Since 80% of jobs are not posted or advertised, it is important to network as well as look on these job sites. Still, there are hundreds of jobs posted online in this field. It is up to the student seeking a job to figure out what type of job he or she wants and then to target that position and demonstrate his or her value to potential employers. Develop specialized skills that might be needed to get that job. There is no substitution for networking and hard work. Be prepared to send out at least 200 resumes and to be focused but also flexible.Dates to watch out for: February 18: Lynn Hazan, recruiter, 5pm, February 26, Okner Symposium, 9-4. Both events are required for 455 and 470 and will take place in Barsema Hall.

Legal and ethical issues in marketing

We learned this week that there are a number of laws that regulate internet and interactive marketing. I asked the students to consider a scenario where a company obtained a list of mobile phone numbers in its target market and whether the company could use the list for text messages or phone calls promoting its services. We identified a number of laws that related to this issue. CAN-SPAM, requires express prior permission before sending the message and applies not only to email but to text messages that reference a domain name (3122222455@wirelesscarrier.com) and consumers cannot be required to incur any costs for the authorization. The TCPA (Telephone Consumer Protection Act) also prohibits the use of automated equipment to place calls to a mobile number without prior consent. In addition, the Mobile Marketing Association as published Guidelines which do not have the force of law, but also suggest that the consumer should have prior notice and consent that the text message is coming. The Do-Not-Call List would also likely apply if the company chose to call these numbers. If the phone number were obtained from children under 13, the Children's Online Privacy Protection Act (COPPA) would apply. This simple scenario indicates that internet marketing law is complex and evolving. From an ethical point of view, both guidelines and laws point to the simple principle of permission. Getting permission is an easy way to know that dealings with the customer are both ethically and legally correct. In addition to protecting children, we also protect consumer healthcare (HIPPA) and financial (GLBA) information. However, there is no general right to privacy of personal data in the United States as is required by the European Union, for example. We also discussed copyright law and the fact that peer-to-peer sharing of music is illegal, even though facilitated by technology. We will have to see how the music industry evolves and adapts to these technology challenges.

Dreamweaver and web design

We looked at the Dreamweaver program over the past few sessions. Dreamweaver creates cleaner HTML code as opposed to word and has a number of other advanced features. The ability to create a separate Dreamweaver 'site' has the advantage of making sure everything needed for the web site is in one place: Cascading Style Sheets knowns as CSS (preformatting for your web page), photos, images and HTML pages. In lab, we recommend putting that site on the network drive and/or backing it up on a portable drive as well. I have been able to solve some problems with website displays by reminding students to put everything in their www folder that the web site needs, including CSS's and images. Also, make sure in Dreamweaver you just have the image name and not a specific location on the desktop or network drive, otherwise the image will not display. Remember to use tables and not frames for your web page for good web design. Enjoy your web site design!