Consumer catalogers’ love affair with cooperative databases is still going strong. Not only are marketers continuing to use co-ops for a significant percentage of their prospecting, but they’re also turning to the databases for help enhancing the performance of their house files.
In turn, the five major cooperative databases are showing their love by showering the marketplace with a slew of new products and services.
With co-op databases, members in effect exchange their house file names for names from other catalogers in the co-op. The co-op providers take the member’s names, analyze data from them, then identify the best prospects available from other companies’ names in the co-op through various proprietary modeling methods.
Competition has increased among the co-ops, which has benefited the mailers, says Jerry Schillinger, president of Hinsdale, IL-based consultancy Schillinger Direct. “Competition is always good, and the co-ops are constantly trying to come up with new ways to harvest their databases to bring the mailers the best way they can.”
New tricks from older dogs
Broomfield, CO-based Abacus, a division of marketing services provider DoubleClick, has the largest of the co-op databases, the Abacus Alliance, with 65 million names; founded 14 years ago, it’s also one of the oldest. But that hasn’t stopped it from unveiling new products, says vice president, marketing data solutions Casey Carey. Most recently, in January, it introduced DataEdge Enhanced, which integrates Abacus’s models with DoubleClick’s DataEdge list processing. This service, which costs $70/M, creates what Carey calls a “smart merge.”
As an example of the service in action, Carey cites a merchant of limited-edition collectibles that wanted to improve prospecting response and generate new customers. Testing indicated that nonmatches to the Abacus Alliance on the mailer’s postmerge tape had low response and a low sales-per-piece rate. Using DataEdge Enhanced, the marketer could replace the low-performing names on the back end of the merge process with more-responsive modeled names from the Abacus Alliance. The suppressed names generated a 0.38% response rate, while the DataEdge names generated a 0.64% response rate.
Last summer Abacus launched Sequent, a prospect model that targets households with low to moderate spending levels — households that are often overlooked by mailers. A home decor cataloger using Sequent has benefited from a universe of almost 220,000 names, Carey says, 82% of which were unique.
Abacus shares its elder statesman status with another 14-year-old co-op database: Z-24, from Costa Mesa, CA-based Experian. As is the case with Abacus, Experian is continuously revamping and tinkering with Z-24, says Sally McMahon, marketing vice president for cooperative databases.
For instance, the company is leveraging Experian’s hygiene and matching technology for enhancements that, among other things, identify addresses that McMahon says other services can’t. The cost is $60/M.
Experian is also leveraging data from its October 2004 acquisition of Simmons Research, which provides information on what consumers buy, where they shop, their attitudes and lifestyles, and the media channels they use. Experian uses Simmons data as both a segmentation driver and as variables for more advanced targeting within the Z-24 database. Using the data, Z-24 unveiled what it calls Focused Models to find affinity groups of customers, such as, say, plus-size apparel buyers.
The newer kids on the block
Abacus and Z-24 work largely, though not exclusively, with models based on the recency/frequency/monetary (RFM) of catalog buyers as well as the specific catalog type. The newer cooperative databases, such as Prefer Network and I-Behavior, both founded in 1999, and NextAction, founded in 2002, have differentiated themselves by drilling deeper into the consumer data their members share and have launched variations of product aimed at getting SKU-level data. Although their methods vary, I-Behavior, NextAction, and Prefer Network, categorize buyer names by the product category purchase history as well as by the catalog category from which they’ve shopped.
Among their newer offerings, Harrison, NY-based I-Behavior in March launched I-Append. Sourced at the SKU level, this data enhancement product can supplement clients’ own RFM customer data for use in inhouse file modeling and selections. For example, suppose a multititle mailer wanted to transition customers between its gifts catalog and its gardening book. Using I-Append, the cataloger could discuss that John Doe, who had bought only tabletop gifts from the gifts catalog, had also bought flowers and outdoor patio furniture from another cataloger. By overlaying the additional data, I-Append “adds to what a marketer knows about the customer and may help convert that customer in ways the cataloger/retailer couldn’t before,” says I-Behavior president Lynn Wunderman. Costs for I-Append begin at $150,000; updates are released quarterly.
For its part, Westminster, CO-based NextAction plans to introduce several products and services within the next few months, just in time for the fall/holiday season, says Karen Crist, senior vice president, client services.
Core Knowledge Reports, launching this month, provide analysis of members’ data and reports on the basic metrics of their business compared with those of the marketplace on a whole. The reports provide information to members on market share, seasonality, demographic profiles, merchandise patterns, and channel habits.
“We’ve always analyzed the client’s data when developing model recommendations,” Crist says. “We decided to package the information to allow clients to leverage it for other parts of their business.” The standard reports are free; NextAction will provide a custom quote for any customized analysis.
To find incremental names, NextAction revamped its database to incorporate merchandise-level data as variables in the modeling process. Also launching this month, its Merchandise Variable Models identify not only the merchandise a household purchases but the frequency and level at which they buy certain products as well. “We weigh the interaction among the merchandise purchased to better picture a buyer’s lifestyle,” Crist says, adding that the merchandise variables can also be applied to other NextAction models, such as its merchandise and activation models.
Not to be outdone, St. Paul, MN-based CMS Direct, parent company of the Prefer Network, last October launched a merchandise reporting service called Merchandise Intelligence that shows marketers what customers are buying, at a product level, from other companies as well as from them, says chief information officer Andy Cossette. The service enables clients to understand what consumers are purchasing across the entire co-op database. “Why would you spend $250,000 to produce and mail catalogs without spending $1,500 to understand what your customers want to buy?” he asks.
The service provides access to yearly trends within a cataloger’s merchandise categories and helps mailers understand which categories possess affinities with respect to potential spin-offs and category expansion.
Another product from CMS Direct, Proselect, is able to find low-performing names in traditionally strong lists and high-performing names in traditionally weak lists, to help marketers expand their prospecting universe, Cossette says. Launched this past September, ProSelect costs $7,500, and scoring is $15/M.
As if five consumer co-operative databases weren’t enough for catalogers to keep track of, Longmont, CO-based Wiland Direct is launching a consumer co-op database in September with 25 million names and more than 300 catalogers participating.
B-to-b merchants, we didn’t forget about you: Next month we’ll look at what’s new among the business-to-business cooperative databases.
|12-month file size||Total number of participants||Level of data on databases||Cost for prospect names||Cost for optimization|
| CMS DIRECT
|25.0+ million||500+ (catalogs only)||SKU||$70/M||$40/M|
|65.0 million||1,900||Transaction, product segment, merchandise category, household||$70/M||$35-$45/M|
|40.5 million||706 catalog titles||Transaction, order||$60/M||$20-$40/M|
|I-BEHAVIOR||31.0 million||1,109||Customer, order item/SKU||$67/M||$37/M|
|NEXTACTION CORP.||50.0 million||950||Order, product||$55/M||$40/M|
|Source: Schillinger Direct|
How to make the most of the database co-ops
To get the biggest bang for your buck when working with co-op databases, keep in mind these tips from Jim Coogan, president of Sante Fe, NM-based consultancy Catalog Marketing Economics:
Share your results The modelers need to know your results so that they can tweak the models that are below breakeven, Coogan says. They will build the best strategy for running the models in rounds, comparing recent results with past results and with your controls.
Educate the database about your business What makes your catalog unique? How does your product differ from that sold by other catalogs in your same broad merchandise category? What is your seasonality? What mailing lists have failed in the past? What rental lists always work? What plans do you have to change your merchandise mix in the upcoming season? The more your database knows about your business, the better your models will respond.
Even if you have a list broker or a circulation consultant managing your list acquisition, direct contact between cataloger and co-op database is vital, Coogan says. After all, no one knows your business better than you.
Be clear on your financial goals for prospecting Share your breakeven sales per catalog so that the database professionals can understand which models are working as well as your financial expectations.
Know your modeling goals Often co-op databases assume you want the largest potential list universe, but you may instead want to find the models with the highest possible return and then gradually expand into less responsive segments. Determine whether your priority is quantity or quality, and be certain to communicate that with your team at the co-op.
Set up control models — and stick with them Controls are vital to measuring whether you are getting better results over time, for measuring the relative performance of new models, and for gauging seasonality. They also let you know if your co-op list universe is growing or shrinking.
Track your unique universe of names coming from the database, the overlap of names with other databases, and the overlap of names with outside prospecting lists. Measure how effective the database is at finding new names within your core market and at finding fringe names you’d likely never reach through traditional rental lists. Know the overlap with other databases; don’t use multiple databases unless they give you separate universes of unique names.