Techniques for Better Data Analysis

Share this article:
There can be a downside to being "uniquely known," but in business, especially in e-business, that is rarely the case.


In fact, knowing your customers and prospects and being able to market to them on a more personal level than can be easily accomplished in the offline world is the holy grail of e-tailing. It is much easier said than done, however, as many online marketing professionals now know. It is not a simple matter to collect, analyze and effectively put to use all of the data generated from customer interactions and transactions with your Web site.


It is imperative that you know who your best customers are as well as which customers are costing you the most. The most productive use of an organization's resources is to focus on these two critical customer segments: growing the first and shrinking the latter.


Data mining and behavior analysis can be used to increase the likelihood of success in this endeavor. There is no magic formula or perfect tool that will spit out the answers to these questions. It will take astute business analysts using smart technology to sift through the reams of data, convert it into meaningful customer behaviors, recognize patterns of potential value, propose and implement a course of action, and then incorporate the results to refine the next go-round.


How do you start trying to improve your conversion from browser to buyer and then into loyal repeat customer? Try the following:


• Create a data capture strategy that makes you, not the customer, do the work. If the first thing you ask your customers to do is fill out a three-page survey about themselves, the next thing they will do is hit the unsubscribe button or click over to the next Web site. Data must be collected in bits and pieces where and when it makes sense for the customer. Build the back-end systems, or data warehouse, to collate data collected from a myriad of sources over a period of time. Also, focus on the most meaningful data instead of everything all at once.


• Be upfront about what you will do with the data. There are no secrets on the Internet, so you might just as well make explicit disclosure a benefit to the customer.


For example, profile and segment your customers and prospects. What do your best customers look and act like? Use data modeling to identify additional customers who could move up into this top-tier group and encourage others to act like them. What sources drove the highest conversion rates? What attracts consistently low-value customers? What areas of the site cause the most problems and, therefore, need the most customer service intervention, which is costly?


• Treat different customers differently based on their value, or potential value, to the organization. Determine which of your initiatives are actually going to benefit your best customers the most, and ruthlessly pare down the rest. Companies almost always have too many initiatives. Aggressively seek out projects to stop. You may even consider eliminating services or products that are attractive primarily to your least profitable customers.


• Monitor and measure the changes and use the results to make more informed decisions the next time. It won't be the reaction to one program that will provide the aha, but by studying and adapting over time and over lots of customers. Rigorously play "Beat the Control" to find the programs that drive incremental lift vs. expected results.


Improved profitability from data mining and personalization efforts results from a combination of the following factors:


• Reduced marketing costs due to better targeting.


• Increased customer loyalty and retention by building a dynamic learning relationship rather than just a series of transactions.


• Improved customer service through the avoidance and identification of problem areas, and the creation of a better shopping experience.


• Optimized allocation of all-too-scarce resources toward the most profitable activities and programs, and away from initiatives that did not contribute to the bottom line.


• Increased competitiveness due to reduced marketing campaign cycle time.


Just think about the power of knowing what needs to be done to increase, and maybe double, your visit-to-order conversion. With little or no additional marketing expenditure, your revenue and profits, and the likelihood of progressing down the yellow brick road of financial profitability, can improve exponentially.


• Evelyn Somers is senior vice president of marketing at SmarterKids.com Inc., Needham, MA. Reach her at esomers@smarterkids.com.

This material may not be published, broadcast, rewritten or redistributed in any form without prior authorization. Your use of this website constitutes acceptance of Haymarket Media's Privacy Policy and Terms & Conditions