Dig for Gold in Your Housefile

Evaluating customers based on accurate and timely data—especially behavioral and demographic data—has given catalogers who understand and leverage the value of a marketing database the ability to achieve significant performance gains.

But in terms of understanding customers and marketing to them as wisely as possible, this technique is just a beginning.

Indexing transactional activity (that is, matching all activity to the appropriate customer) creates an even higher level of marketing database utility—a truly customer-centric view of behavioral data. With this kind of data you can improve customer segmentation simply because the data is more accurately rolled up, easier to access and “cleaner” due to the intensive scrutiny it gets during construction of a new database.

This better-knowledge effect typically enables a new view of data from which you can improve housefile performance by 1 percent to 3 percent—and that’s before applying modeling or other advanced segmentation tactics. Generally, gains of this sort, along with productivity advances that accompany easier data access and name selection, deliver profit improvements that more than offset the cost of a new database.

But real performance improvements come when you can assess customer behavior through a multi-dimensional, dynamic view of transactional activity. What does this mean? Traditionally, forms of recency, frequency and monetary (RFM) values or statistical modeling get better as the data used to build a segmentation tool become cleaner.

And the very depth and breadth of the data—now properly rolled up and easily accessible in a database—creates the opportunity for further analysis. New analytical tools and data-modeling technologies, coupled with lower cost and increasingly powerful computers, offer opportunities to improve profitability through data mining.

The key is to look at customer transactional and demographic data differently and more often than you have in the past. Incremental gains of 10 percent to 25 percent can be achieved by looking at predictive data more dynamically than traditional segmentation methodologies. Where do you look and what should you consider if you want to achieve these kinds of gains? The following five techniques can help uncover significant marketing opportunities.

Tip #1: Consider your segmentation criteria and methodology each time you mail.

In cataloging it’s common to see the same segmentation strategy followed every year. Marketers may update transactional activity after each mailing, which changes the values for RFM and product purchase information. But how often do you revisit the stratification capability of the variables used in segmentation? Is the top-to-bottom-segment ratio still as great as it was on last season’s campaign? Should new behavioral characteristics that would help differentiate between good- and poor-performing groups of names be considered?

Today, you can rebuild segmentation and name selection with each mailing to take advantage of newly discovered predictive characteristics. Dynamic segmentation of this sort reveals profit opportunities that easily exceed the cost of analysis and implementation.

Tip #2: Look forward with your segmentation to the offer you’re about to make.

As a cataloger, you typically consider pagination, product mix, product price and major product category variations when you do roll-up marketing forecasts. That’s all right, but there’s money left on the table

if a segmentation scheme fails to account for the next offer’s specifics.

Ask profit-driving questions before settling into a familiar marketing pattern. Are you actually using the contents of the next offer to drive name selection outside the bounds of a standard housefile-segmentation methodology? Should you re-rank names for selection because of shifts in product mix? Should you re-set monetary breaks for new and higher-price-point offers?

Many offers contain only subtle variations on merchandising, in which case the effect is minimal on the characteristics that determine segmentation. However, as the merchandise or offer mix changes, or as you introduce new product categories, failing to modify segmentation parameters may mean you miss your best new buyers.

Tip #3: Don’t lock into a single methodology for segmentation.

Regression modeling is one way to analyze and act on a more expansive view of transactional-predictive behavioral data. Performance gains of 10 percent to 15 percent are common when using data obtained through this technique compared with traditional RFM methods.

Those are great numbers, but even with substantial data covering your whole customer population, a model is limited. It defines predictive behavior with only the most significant predictive characteristics included in the equation. This means the model essentially is a best-fit equation defining expected performance. This can leave groups of customers under-represented in a housefile-scoring effort, resulting in low scores for groups that include potentially high-performing names.

So, is regression modeling for your housefile outdated and ineffective? Not at all. Model scores, when coupled with other predictive characteristics, can provide a more discriminating segmentation scheme coming into and out of a segmentation formula based on, say, offer, seasonality, recency and/or frequency. The key is to use all available tools in developing the most discriminating segmentation methodology you can.

It’s not an either/or question; the best solutions consider all available analytical tools and customer intelligence when it comes to leveraging segmentation methods. For example, some catalogers using regression modeling as a housefile-segmentation tool have incorporated additional behavioral characteristics to define their segmentation to improve performance. These same catalogers have achieved additional gains beyond modeling of up to 16 percent.

Tip #4: Create segmentation from list interactions, not just “hits” files.

Comparing a housefile to rental files to help identify external buying activity can be beneficial. Housefile names that have little activity (e.g., inquiry names, giftees) or activity considered too old (e.g., 60-month singles), but that match active rental file names, can be profitable. Typically, catalogers segment these names into two groups: those with and without current outside activity. Most of us limit our segmentation to a hit against any kind of activity, without ranking the hit’s value. Ranking the type of hit by the characteristics of the list (e.g., cooperative database, catalog response list, subscriber list, primary merchandise list category) provides more data for dynamic segmentation. You’ll find that segmentation ranking results roughly follow the same order of performance as the response generated from that list category when mailed for new-name acquisition.

This strategy reveals segments of hits that shouldn’t be mailed, as well as best hits on a housefile segment with a sub-standard performance history. This strategy’s benefits can be substantial when acquisition lists aren’t performing or when universes that continue to perform well are so small they fail to support name flows that meet your business goals.

Tip #5: Build promotion history into your segmentation strategy.

While maintaining promotion history raises the maintenance costs of an outsourced database, the precision of the contact strategy you can develop and the resulting performance gains far outweigh the costs.

Also, having this data available is key to analyzing lifetime customer value. Try capturing promotional velocity as well (i.e., the time between promotions). The ultimate contact strategy pits promotional velocity against order velocity to optimize customer profitability. Generally, catalogers with promotion history available to them can generate long-term gains of 5 percent to 10 percent based on their increased ability to leverage lifetime-value equations and contact strategies across groups of customers.

To act on these opportunities requires a lot of analysis and some roll-up methodology. But as you get started, let Pareto’s principle be your guide: Identify opportunities that get you 80 percent of the gain for 20 percent of the effort.

Mike Talbott is vice president of business development for Catalog Marketing Services (CMS), which provides merge/purge and database services, and data analysis and marketing consulting services. He can be reached at (651) 636-6265 or by e-mail at miket@netcms.com.