Recognized for over a decade as one of the leading authorities on CRM strategies for business, Don Peppers is an acclaimed author and a founding partner of Peppers and Rogers Group, the leading customer-centered management consulting firm. Don's vision, perspective and thoughtful analysis of global business practices has earned him some significant citations by internationally recognized entities. The Chartered Institute of Marketing cited Don among their inaugural listing of the 50 "most influential thinkers in marketing and business today."

Don's expertise and clear concise way of thinking places him in high demand as both a speaker and a management advisor with Fortune 500 executives and entrepreneurs seeking to identify their most valuable customers, increase customer satisfaction, and improve ROI. He is a popular voice among editors and the media, both online and in print, and is the co-author, with Martha Rogers, Ph.D., of a series of international best sellers that have collectively sold over a million copies in 14 languages.

The One-to-One Future (1993) was named by Inc. magazine's editor, George Gendron as "one of the two or three most important business books of all time," and is considered by many as the bible of the CRM revolution. The authors have published the first CRM textbook for university use in graduate level courses, Managing Customer Relationships (April 2004). In addition, the authors have recently published their newest business book, Return On Customer.

Predictive Analytics Insight were fortunate enough to grab this exclusive interview with the CRM guru.

PAI: What is the must-have data for predictive analytics?
DP: Initially, there’s what I call first order data and this clearly means customer transaction data. But it’s important to mix this with all other non-purchase data gleaned from the interaction. In other words, not just purchase details, but enquiries, web tracks, and service issues. Customer interactions of any kind could be relevant to customer motivation. In many ways, attitudinal data is just as important as transactional information as it logs issues like customer satisfaction, complaints and problems.

PAI: Can you give an example of this non-purchase data?
DP: Any proactive complaint, comment or compliments is data gold dust for firms. It is always worth adding in a field at the end of a customer interaction to gather attitudinal data such as rating customer satisfaction, and finding out whether things could be done differently. In general, the more open ended the question, the better.

PAI: Does any other data help build a more complete picture of customers?
DP: Definitely. It’s important to factor in information from customer satisfaction surveys. Often firms carry out this research, but then fail to factor the findings into the way they handle their customer relationships. As a result, they are unable to predict future spending patterns. This situation occurs because the department that measures customer satisfaction is often in a completely different department to financial records, with the marketers falling somewhere in between.

PAI: What about integrating all of these data streams?
DP: It’s clear that you have to have an integrated set of data. It’s only by mixing transactional data with attitudinal data that a company is able to gauge a deep understanding of their customers to predict future behaviour. This data it can also be used to work out customer problems – for example a customer’s failure to buy could be explained by the fact that a complaint has gone unanswered.

PAI: Should firms focus on internally-generated data?
DP: To build up a really good view of customers, it is also necessary to build up a profile from third-party, independent information, such as data gathered by post code, and psycho-demographic scores. This can provide an important external benchmark. Firms can also access a wealth of syndicated data based on inferences such as an individual’s propensity to act depending on post code.

PAI: How do you pull these data streams together?
DP: Once you have the data, it’s then a matter of making sure that it is used to full effect. This often relies on how effectively firms are able to integrate the different data streams. And more often than not, this hinges on the structure of the organisation as much as the structure of the data itself. Data is often collected within companies with a specific purpose without regard to how it can be united to form a clearer picture. This is how data silos spring up, completely independent of one and other

PAI: What do most firms get wrong in handling their data?
DP: While many firms are able to get an idea of their performance in a historic sense, most of them come unstuck when the try to create a model for future outcomes. The firm is likely to have a lot of data on transactions and past events but is unlikely to have any data suggesting future happenings. One of the main problems is that firms have misunderstood the way in which customers create value. As well as being aware of how customers buy things currently, they also need to be aware of how they can influence the customer’s buying intentions in the future. In this way, the customer’s lifetime value goes up and down depending on the buying process.

PAI: Why should consumer-facing firms change the way they operate?
DP: Rather than focusing on short term buying patterns, firms need to consider the customer’s value to the company over the lifetime of their relationship. At present, most companies tend to focus on short-term results and current earnings. They overlook the importance of giving the customer a favourable view of their business and ensuring that they return to the business in the future. In this way they can work out whether the customer will be more or less inclined to buy.

PAI: Why is there so much customer churn?
DP: If a company operates an aggressive marketing policy they are more likely to recruit new customers today, but those same customers are less likely to be around in the future. There is a massive trade off between short-term sales and long-term value from the customer. We are seeing a return to the age where the customer is king. Companies need to be much more aware of how their relationship with them can impact on the long-term customer value.

PAI: Are there bridges to build?
DP: The concept of trust is very important because too many companies abuse their customers whether this is by exploiting personal information about them or by putting different prices on products depending on which channel you choose to buy them. They are relying on the fact that their customers don’t get to find out. Happily more and more firms are realising their folly, and are making their processes more transparent. By doing this they are able to develop a more long term and value-driven relationship with their customers.

PAI: How can predictive analytics help?
DP: Predictive analytics gives firms the ability to look beyond the short-term focus on earnings for that quarter and concentrate on building a long term, highly trusted relationship with their customers that will ultimately deliver more shareholder value. It is not within their interest to focus on current earnings at the expense of the future.

PAI: Will this help firms’ bottom line?
DP: Predictive analytics will provide currency and ammunition when firms go to the analysts and the city. It can also provide firms with data that proves the long-term effectiveness of investment programmes. In sheer value terms, the methodology can help firms identify customers that are likely to churn, a hugely profitable exercise, as some are likely to be much more valuable than others. Armed with this knowledge it is much easier for firms to boost their retention efforts surrounding these special customers as this will contribute directly to the bottom line.

PAI: How do you automate processes to ensure your firm responds in the right way to a customer’s data profile?
DP: Before you automate these changes, you have to understand them. It’s important to watch the warning light go off, then observe customer reactions because each customer is likely to behave in a different way. Firms must work on the principles of mass customisation which involved breaking up a product management or service delivery model into different modules. In this way firms are able to digitally tailor their modules to match a greater variety of outcomes. For example, one credit card company has two million different permutations of the same product, each one designed to meet a different customer need.

PAI: Why else should firms move to predictive analytics?
DP: All firms are looking to boost revenues, increase their competitive advantage and reduce costs – predictive analytics enables firms to deliver on all three. Predictive analytics also enables firms to mitigate the risk of being defrauded by their customers as it become much easier to pin point those who are likely to default.