It was a disappointing start to the fall season at the nation's retailers, with nearly 60% of the industry missing sales expectations for July. Blame blistering weather and great car deals for the poor showing at big names like Federated Department Stores, Gap, and Nordstrom, not tapped-out consumers. But regardless of the cause, the pressure will be on for stores to move fall inventory without, um, losing their shirts.

Maybe it's time for the industry to take a closer look at retail revenue-management software, technology to help optimize pricing for general sales, markdowns, and promotions, to keep inventory moving and gross margins where they should be. Many of the early converts to price-optimization software have been supermarket chains, which typically carry tens of thousands of items across hundreds of categories and must constantly readjust prices to meet supply and demand changes and competition. But department store and apparel retailers--including The Children's Place and Sears, electronics outlets Best Buy and RadioShack, and drugstore chains Duane Reade and Longs Drugs--also are trying to maximize their merchandising profits by using the technology.

Children's Place's strategy relies on measuring the risk in taking or not taking markdowns and missing the opportunity to move products, says Richard Flaks, senior VP of planning, allocation, and IT. "There are literally millions and millions of permutations imaginable to determine the correct price and effective time to take a markdown," he says. "I just don't have a team big enough to crunch the numbers."

Convenience-store chain 7-Eleven Inc. just joined the small set of retailers that have ventured in this direction, beginning with a four-month test last year in one of its Denver stores of a software suite called Customer Demand Solutions from Khimetrics Inc. In one case, the software helped the store determine that it would make a better profit selling a certain beverage for a discount price of $1.50 rather than offering customers two drinks for $3, which all its competitors were doing under the supplier's promotional guidelines. The result: It sold 10% more of the product than it normally does.

Only 5% to 6% of retailers use price-optimization technology, says Scott Langdoc, VP of retail research at AMR Research. Analysts and vendors--which include i2 Technologies, Khimetrics, Manugistics, ProfitLogic (recently acquired by Oracle), and SAP--tout a 1% to 3% boost in overall sales and in some cases a 10% rise in gross margins for companies that deploy the technology. In December 2002, discount department store ShopKo Stores Inc. shocked many on Wall Street when it reported a profit of 3 cents per share on gross-margin improvements, despite a 3.2% decline in third-quarter same-store sales, and its use of markdown-management software by Spotlight Solutions, later acquired by ProfitLogic, was publicly credited as a key factor in the results. (Shopko has fallen on harder times since then and is in the process of being sold to a Minneapolis investment firm.) Langdoc has seen gross margins improve by as much as 16%.

THE UPSHOT Helping companies nail the prices, markdowns, and promotions that will appeal to customers adds up to bigger gross margins. So much promise, so little pickup. Revenue-mangement technology has been around for years, but complex interfaces, high costs, and merchandising managers' fears of losing control have held back adoption. That may be changing, as more competition and reduced inventory cycle times demand that retailers pay more attention to efficiently moving products. For more on 7-Eleven's IT initiatives, see "King Of Convenience" in Optimize

Despite evidence of successes, retailers have hesitated to embrace the software, which has been around for at least a decade in one form or another. High deployment costs in a historically IT-averse industry, the problems germane to extracting and integrating data from multiple retail systems, and first-generation interfaces that only a Ph.D. in computational science could love all had a hand in that. Yet in many ways, retail-revenue-management software is finally coming into its own. Interfaces are easier to use, and the numbers-crunching capabilities, such as Bayesian analysis that uses the knowledge of prior events to predict future events, occur automatically behind the scenes. And today, the average retailer pays between a half million and $1 million for an in-house deployment.

Still, retailers also have been reluctant to trust the technology. "A lot of [pricing] is gut feel," says Kosin Huang, program manager at market-research firm the Yankee Group. Retailers base prices not only on tangibles, such as competitors' pricing or an item's sales performance, but also by observing and talking with customers as they shop (see story, "Trust Issues: Leading With The Heart And The Head").

Challenging times call for new approaches, however. Increased competition from rapidly growing discount chains--Wal-Mart Stores Inc. is opening 530 stores this year, including 160 SuperCenter expansions--and reduced cycle times for inventory in stores are helping to drive the change. "Today, you have new [apparel] products on shelves for about six weeks, compared with 15 weeks about two years ago," says Gladys Lau, director of Oracle's retail division.

Retailers are expected to spend $218 million in the $1 billion market for price-management and profit-optimization applications in 2007, according to a new Yankee Group report. That's up from $70 million last year in a market with overall sales of $291 million, the report says. "Many companies look to deploy markdown optimization as a method to correct their buying mistakes," Flaks says.

Most retailers have the raw data they need to do pricing. Point-of-sale systems generate huge amounts of raw numbers that are stored in data warehouses and can be turned into pricing intelligence. "For a large retailer like Wal-Mart, where they have more than 400 stores, you're talking about terabytes of data," says Alan Montgomery, an associate professor at the Tepper School of Business at Carnegie Mellon University, who has studied pricing decision-support systems for the retail industry.

But it's a huge task to get that data ready to use. Retailers struggle to clean up and integrate transaction data stored in data warehouses with inventory and enterprise-resource-planning systems that house information on in-stock items, purchasing, and financials to form a complete picture of when a promotion makes sense to clear out unsold merchandise, for instance, or to predict profits based on wholesale costs. Companies must develop business rules for the software algorithms that sort out hundreds of options and scenarios for purchases, prices, promotions, and markdowns.