BI Is in Fashion at Elie Tahari

By Paula Klein, TechWeb, published May 23, 2011

In the world of high fashion, Elie Tahari competes with its cutting-edge clothing, chic shoes and stylish runway shows. In the world of IT, the company competes by using business intelligence and data analytics.

Based on its size and organizational structure, the $500 million, privately held business may seem like an unlikely IT leader. While it has a global presence and more than 600 boutiques worldwide, the company has no CIO or CTO and runs on tight budgets with lots of high-touch personal interactions.

In fact, this was the perfect environment for standardizing reports across dispersed lines of business, creating shared data repositories and offering real-time transaction data. In 2005, the company built its first data warehouse with the help of consultants and vendors and over the next several years, it rolled out applications to every part of the business — from order entry to sales to customer service. Modules for reporting, logistics, analytics, budgeting and planning were added and acceptance grew. The efforts were driven from the business side and were based on their priorities, says Nihad Aytaman, director of business applications. “We didn’t do a big-bang implementation and that reduced resistance. We took baby steps and made [reports] very visual so that users wanted more.”

“Store managers could see patterns and trends and plan their inventory better; buyers could check order status immediately,” says Aytaman. ROI is measured by increased sales, fewer returns, faster shipments and more-satisfied customers, he says.

As the business gained familiarity with the system, it trained BI “super-users” in each department to learn about new tools and lead their efforts. As a result, only one corporate BI manager is needed to maintain the systems and to address more complex issues. “My message to small businesses is that BI is within reach,” Aytaman says.

Last year, the company began implementing predictive analytic tools based on a statistical engine that culls two years of product history. It’s not designed for the short cycles of seasonal fashion lines, but does help on the annual side of the business. The analytics engine is tied to a planning tool that can forecast demand so that buyers adjust purchasing accordingly, he says. “Our risk exposure is minimized, and we’re not overproducing.” The data is more accurate and more actionable, he says.

Aytaman notes that “hammers don’t replace carpenters,” meaning that it is not the BI tools themselves that help the business; it’s how the results are used. “Lots of companies still use BI for reporting only, but if you just have data and dashboards and no one uses [them], it’s not worthwhile.”

Later this year, he will apply BI techniques to social networks to analyze “chatter about the company,” but unstructured data is much harder to capture and measure, he says. “That’s not an exact science,” so we’ll have to see how it works, he adds.