Guess Gets IT Right

Founded by four brothers from Marseilles, Guess Inc. has become one of the great American success stories. In 1981, after it opened a clothing store in Beverly Hills and introduced a line of European-inspired jeans, the Marciano brothers convinced Bloomingdale's to take 30 pairs of their jeans on consignment. When the jeans sold out within hours, the Guess phenomenon ignited.
During three decades of colorful corporate history -- including highly publicized struggles for control of the company as well as edgy, controversial ad campaigns that launched the careers of several supermodels -- Guess grew and diversified, expanding its product line beyond denims to a wide selection of apparel, accessories and footwear. Today, its major brands include the upscale, youth-oriented GUESS?, the even more upscale, women's fashion-oriented GUESS by MARCIANO, and the affordable but aspirational youth brand G by GUESS. Still led by two of the Marciano brothers even after going public in 1996, the company posted revenue exceeding $2.1 billion and net income of $242 million for fiscal 2010. In recent years, Guess has been one of the most consistently profitable of U.S. apparel companies, and its volume has more than tripled since 2003.
The company has diversified not only its product mix but also its business model. Although the Marcianos started out in retail, Guess was primarily a wholesaler until the 1990s, when it began adding stores in both high-end malls and outlet malls. Today, 85 percent of its North American revenues are direct-to-consumer, derived from 433 stores and a thriving e-commerce business.
Diversification the name of the game
During the past five years, the balance has swung back toward wholesale as Guess bought back many of its international licenses and expanded its presence in Europe and Asia, where operations are still largely wholesale even after hundreds of new store openings. Between 2005 and 2010, European revenues grew from $154 million to $705 million and Asian revenues grew from $14 million to $132 million. Close to half of Guess' total non-licensing revenues are now international, as are more than half its operating earnings. International revenues will continue to grow in the future -- and will probably shift the mix further toward retail again as the company opens new stores in northern and eastern Europe as well as in China.
"Diversification has really helped us," says Mike Relich, the company's CIO, noting that one country's economic strength may make up for another's weakness, and that diversity in products, brands and business models also helps hedge the company's bets.
On the other hand, accommodating widely differing business models has made Relich's job extraordinarily complex. In Korea, for example, department stores operate on a consignment basis, though this is not the rule anywhere else. Guess owns its North American stores, but in other countries it has a mix of company-owned, franchised and joint-venture retail stores. Stores may buy Guess products directly from the company, from distributors, or from licensees -- and the mix varies from country to country. Achieving visibility into, and control over, the company's operations is a huge challenge because of all these variations, and the challenge has been compounded by the company's rapid growth.
In an ideal world, a global company such as Guess might want all its divisions to follow the same processes and use the same systems. In the real world, such uniformity isn't always practical. The European and Asian licensees that Guess reacquired had developed independently, and the company had to make strategic decisions about how best to integrate their operations with the North American operations.

Strategy first, IT integration second
One of Guess' key decisions was to create a core global product line, primarily consisting of denim apparel, and then to augment this line with a set of localized products for each market.
A second strategic decision was to centralize the oversight of sourcing and reduce the number of vendors. The most reliable vendors have seen their orders increased and have been treated more like partners; in return, they are expected to maintain high standards for ontime delivery, pricing and quality. "We've worked to create a vendor scorecard, so now it's not some big black box," Relich says. "We sit down with them and show them their on-time rates and their defect rates. It's created an incentive for the vendors."
These two decisions defined the way the different regions would work together and gave Relich a basis for deciding how to integrate the company's IT systems.
Designers in the United States and Europe found they had real problems developing a core product line with separate style numbers and different color palettes. Even if they could develop a common product, they wouldn't have a common language to talk about it with suppliers. Clearly, some integration was called for. On the other hand, ripping and replacing all the company's IT systems was a daunting, costly and risky proposition. Relich says, "When we bought the [European] license back, Europe had a whole different set of systems, and at first we had discussions about, 'Should we put in a common ERP?' But the company was growing like mad, and everyone was focused on trying to handle the growth of the company, so were we really going to change the ERP system at that point? It would have been disruptive."
Relich developed a three-layer technical strategy that would help designers collaborate on the core product line, keep the sourcing operation informed and also help top management understand and control the company. In the first layer, a common product life cycle management (PLM) system would be used by the design centers in Florence and the United States and by the sourcing office in Hong Kong. In the second layer, regional transaction processing systems -- ordering, shipping, financial and so forth -- could remain separate because the regions did not need to collaborate in these processes and also because their business models were so different. Finally, in the third layer, all historical data would be exported into a common data warehouse and used to create companywide reports for use by top management.

Technology: Eliminating duplication and fostering collaboration
Dassault Systèmes' ENOVIA solution was selected for PLM, along with the ENOVIA Apparel Accelerator for Sourcing and Production. Because all style numbers are assigned in ENOVIA, there are no longer any duplicate styles. "It's a cost-effective, low-risk way to ensure the commonality of data," Relich explains. Once the style data is pushed into the ERP systems, the regions execute their transactions independently, "but at least we call it the same thing." At the end of the process, the common style numbers and data definitions allow the data to be recombined in the data warehouse for reporting.
The North American and Hong Kong offices have already adopted the PLM system; the European designers are using it only to assign style numbers but are getting ready to implement it fully as a design collaboration tool -- including attaching comments, screenprints and even videos of fittings. Relich says PLM will make the design process less error-prone, compared with sharing information via e-mail. "Any user in the world can go into the PLM and see the current version of a style. No matter how many changes are made, you can't end up manufacturing the wrong thing."
To improve accuracy still further, the company is working on bringing vendors into the PLM system. Relich says, "If you wanted to get a mill sheet or information on fabric or trim, you'd send an e-mail with Word templates -- but no one wanted to enter the information into the PLM; everyone would wait till the last moment because [the style] might be dropped. Now, you have a function in the PLM to request fabric information; the vendor clicks on a link, logs on, and enters it directly." After a successful pilot project, vendors are now being trained to enter fabric, trim and costing information.
Costing, once performed in ERP, is being moved to PLM to facilitate what-if analysis. The PLM system can produce cost estimates without all style attributes having been entered, allowing designers to gauge costs at an earlier point in the design; if the costing exercise shows that a style is too expensive, it can be dropped without much wasted effort. Costing in PLM also makes it easy for designers to play with multiple versions of a style and decide which ones to keep based on cost estimates. Relich says, "The idea is to make PLM the playground and not enter all the data elements into the ERP until they're sure what they're going to adopt. It's really helped with productivity."
When transactional systems are shared, as they are with some Asian countries, they are implemented on a single instance of Oracle (for financial systems), Jesta (for ERP), Manhattan Associates PkMS (for warehouse management) or the in-house e-commerce platform. Akamai's content delivery network is used to provide geographically dispersed users with adequate performance from a centrally located system.
The advantage of using a single instance of each solution, Relich says, is that data can be consolidated without synchronizing. "If you have multiple instances, you need to download trial balances and then upload again. Now, you can go in and push a button -- you don't have to synchronize." Even if the platform is consolidated, the data's geographic source can be identified so it is sent on to appropriate places -- for example, pick tickets are sent to the local warehouses equipped to handle them and e-commerce users are directed to the correct local versions of the Guess website.
Some other systems, though they don't need to be consolidated, are being standardized anyway -- not to enable collaboration but to share best practices. In some cases, European or Asian systems are being imported to U.S. operations. In the case of retail systems, however, the American systems are being exported. "We've got a lot of decent retail experience in the United States," Relich says, noting that Guess is primarily a retail company in North America and primarily wholesale elsewhere. This summer, the company will pilot its point-of-sale system in a European setting; if the pilot goes well, Relich hopes to roll out the entire retail suite in Europe, including MICROS-Retail for point-of-sale; JDA Software for retail merchandising, allocation, and planning; and MicroStrategy for business intelligence.
Converting Asian stores to the U.S. retail suite is more difficult because of the difference in alphabets ("It's hard to translate everything into Asian characters, so why do it?") but simply making the data consistent solves many of the problems that once stood in the way of consolidated reporting.

Going mobile
Relich hasn't spent all his time integrating Guess' disparate operations, however. The company is also pioneering advanced-technology initiatives of many kinds. Some of these make creative use of mobile devices, for example:
> Business intelligence reports are pushed daily to regional managers' BlackBerries, allowing them to see at a glance what they sell each day, as well as traffic, conversion rates, margins, year-to-year comparisons, and planned-to-actual comparisons. "Before, they would have to log onto the PC and print out reports," Relich says. "Now they just look at the BlackBerry and get all that."
> A new iPhone application will allow customers to create closets of their favorite outfits and then purchase them via iPhone. The company's 250,000 Facebook fans will be able to post these outfits on their Facebook pages and show them to friends. (Guess already keeps Facebook fans engaged by letting them know about new styles and special events.)
> A mobile POS application will allow store personnel to pre-process customers if lines get long, scanning items while customers are still waiting in line. Customers will still have to go to the checkout counter to wrap and pay for the items.
Assortment planning, loyalty programs, traffic counting and more
Another new initiative is assortment planning, using daVinci's assortment planning solution. "We were doing assortment planning with Excel," Relich explains, "but we wanted to automate it and make it consistent across divisions. We did a pilot in the retail stores and it worked well, so we're rolling it out to stores, first here and then in Europe."
The ShopperTrak solution has been introduced in the United States to count retail traffic for each time period. Guess has found this data useful in understanding sales fluctuations ("Before, if the comps were down, we might beat up the store manager") and for hiring the right number of personnel to meet expected store volume.
The company's loyalty programs, based on MICROS-Retail, include a total of 3 million customers, which Guess can now segment and market to appropriately. Loyal customers are given rewards in the form of discounts on future purchases; in return, Guess can perform market basket analysis to find out what each customer is likely to buy next. The loyalty program is integrated with the point-of-sale system and the e-commerce system.
Finally, Guess uses the hosted platform TradeCard for processes from purchase order to payment. Vendors can use TradeCard to submit invoices electronically and have chargebacks handled automatically. Guess uses the system to offer early-payment programs to vendors, taking discounts if they can pay in 10 days. "Before, vendors would be calling production people to ask what was wrong. Now, they can find out why their invoice was rejected and when they'll get paid.
TradeCard has eliminated the letter of credit for small suppliers, allowing them to receive credit based on Guess' creditworthiness. "The cost of capital to these manufacturers becomes lower," says Relich, "and we can reflect that into our margins. Not only do they save money, but it's an opportunity for them to get earlier payments. A lot of the vendors are using it -- and we've integrated it with Oracle and with Jesta. There are a lot of benefits for everyone."
Masha Zager is a New York City-based free-lance writer who specializes in business and technology.
systems at a glance
* Assortment Planning: daVinci
* Business Intelligence: MicroStrategy
* Content Delivery: Akamai
* ERP: Jesta
* Financial Systems: Oracle
* PLM and sourcing: Dassault Systemes ENOVIA
* Point-of-Sale: MICROS-Retail
* Purchase Order to Payment Processing: TradeCard
* Retail Merchandising, Allocation and Planning: JDA
* Store Traffic: ShopperTrak
* Warehouse Management: Manhattan Associates PkMS
This ad will auto-close in 10 seconds