Adding the "P" (Product Category) to RFM (1,079 words)

There are few catalogers who don't know what RFM stands for. It is the technique of capturing customer purchase history by the three most important variables:

R = Recency -- date of customer's last purchase; F = Frequency of purchase -- number of times a customer bought; M = Monetary -- lifetime dollars spent on a catalog by a customer.

RFM is the catalog industry standard for segmenting a catalog's customer or buyer file. RFM is not applicable to new-customer acquisition—only customer list marketing.

A term that is less familiar is RFMP, in which the "P" stands for Product Category. While simple RFM segmentation can ensure you pass your break-even point, adding this fourth variable into your efforts can help you greatly surpass it.

Using Straight RFM

As a catalog grows its customer list to a point where it can be segmented and divided into cells, it can apply simple segmentation techniques to customers to be mailed. For example, rather than thinking "a customer is a customer is a customer," a cataloger can use RFM to answer such questions as:

• To whom should the next campaign be mailed?

• How often should the customer list be mailed?

• Which segments should receive a second or a third catalog mailing during a season?

• Where do you draw the line in determining whether or not to mail?

Determining the optimum number of mailings to your customers is not always easy, even with RFM. Generally, catalogers tend to under-mail their customer list. But there is a right number of catalogs that customers will tolerate before complaining. Victoria's Secret mails its good customers weekly—sometimes more often—and customers complain that they get too many catalogs. There are several creative options in changing the look of your catalog when considering re-mailing within a season.

• Change the cover and keep the inside 32 pages or 48 pages constant

• Change the outer eight pages of the catalog, keeping the inside pages constant

• Re-shuffle or re-paginate the entire catalog to make it look different.

• Add a wrap or dot-whack with a specific action or message, e.g., "We Miss You," "It's Not Too Late," etc.

You must understand that customers are not naïve. The better customers are, the more they will know if you are re-mailing them with just a cover change. If in doubt, ask your best customers how often they wish to receive the catalog. Or, use RFMP in order to send customers just the catalogs that most interest them.

Adding the "P" To RFM - A Mini-Case Study

Most catalog databases, with their management systems, track product category with each customer purchase. Many track this all the way to the item or even SKU (stock keeping unit) level. Some catalogers also use product category as a predictor of future buying propensity.

One of those catalog companies that uses product category in their mailing efforts is Fingerhut Corp. As early as the late 1970s and early 1980s, Fingerhut was light years ahead of other catalogers in database management and data mining. Perhaps it was the fact that Fingerhut started out as a solo direct marketer, rather than a cataloger, that gave it the product line orientation. It has always been a broad, multi-line direct marketer of diverse categories including:

• Women's and men's apparel

• Children's toys and games

• Kitchen products

• Tools, electronics and sporting goods

• Linens and decorative home furnishings.

Fingerhut ascertained that the initial product purchase category could be an excellent predictor of what future products customers purchase. For example, customers who buy women's apparel will be more receptive to future apparel offerings.

Fingerhut's major growth came by increasing the number of mailings offered to customers from 20 to more than 100 per year. The significant point is that all customers didn't receive all 100 mailings. Product category of past purchase (carefully tracked and maintained in Fingerhut's database) was used to determine which mailings customers received.

For example, if the first purchase was women's apparel, customers would received six women's apparel promotions during the coming year, and would be tested on complementary product categories such as kitchen, decorative home furnishings, linens, cosmetics, etc. If customers buy linens, they then receive four linen mailings during the next year, and so forth.

The theory was sound and led to Fingerhut's astounding growth during that time: Buying one product category leads to further purchases in that and related product categories. If a man purchases men's clothing as the initial purchase, Fingerhut will test tools, electronics and sporting goods offers in future mailings to see which interest him.

The Challenge in Using the P-Word

During much of the 1970s Fingerhut mailed more solo direct mail pieces than catalogs. For a general merchant like Fingerhut, cataloging presented some challenges because of the broader, multi-product offering in its catalogs. Two things changed: more niche-driven catalogs and the development of the "multi-mailer," a promotion of multi-products (up to 30 or more) that was cut-sheeted and mailed with a letter in an envelope. The multi-mailer was half catalog and half solo. Fingerhut continues to use the multi-mailer today, while it has broadened its more targeted catalog efforts.

Adding "P" or product category to RFM is highly applicable to many catalogers. Business-to-business catalogs work almost exclusively in product categories. They organize their catalogs by product category; their selling efforts (catalog, telemarketing or the Internet) are often driven by product category.

There are few consumer catalogs in which product category has not become more significant in the past five to 10 years. By studying product category purchase history, you may discover that:

1. Initial product category purchased is a predictor of future product purchases.

2. Future lifetime value (LTV) can be predicted by initial product category purchased.

3. Product category can be used along with RFM as a segmentation tool in determining future mailing, bounce-back promotions, special Internet offers, etc.

4. Internet promotional opportunities are ripe with opportunity based on product category.

The bottom line: Product category may be a hidden key in your database that unlocks a future revenue and profit stream. Track it and test it as part of a new RFMP database tool.

Jack Schmid is president of J. Schmid & Associates, a catalog consulting firm in Shawnee Mission, KS. He can be reached at (913) 385-0220.

P IS FOR PRODUCT: Fingerhut's multi-mailers helped the company take advantage of targeting by product category.