VTech Acheives Supply Chain Excellence
Forecasting retailer purchases should be a simple mathematical computation for consumer goods manufacturers if they know what the customer is going to sell, what their stocking policies are and what their current inventory position is. That is the supply chain philosophy of Nick Delany, president of VTech Communications Inc.VTech is the largest player in the U.S. cordless phone market, with group sales of more than $1.4 billion in the last fiscal year. With headquarters in Hong Kong and manufacturing facilities in mainland China, VTech currently has a presence in 11 countries and approximately 27,000 employees. Its U.S. telecom division based in Beaverton, Ore. markets VTech and AT&T branded wireline corded/cordless telephones through a large distribution network comprised of leading national retailers, regional retailers, distributors and resellers. Collectively, both brands are featured in excess of 25,000 retail store fronts in the United States. In the last fiscal year, the global telecom division contributed about 42.9 percent of the group's annual sales.
Founded in 1976, the company's mission is to be the most cost effective designer and manufacturer of innovative, high quality consumer electronics products and to distribute them to markets worldwide in the most efficient manner. Today, VTech is able to stay true to this mission because -- almost a decade ago -- it pioneered the effective use of a data warehouse that includes demand signals. In fact, the company had a demand signal repository of sorts in place before the term was even coined in 2003.
Today, this enterprise data warehouse, which contains weekly, store-level, point-of-sale (POS) data by SKU from 90 percent of its customers, including Wal-Mart, Target, Sears, K-Mart, Best Buy, RadioShack, OfficeMax and Costco, empowers nearly every department -- from supply chain to sales to research and development and more.
Here is the story of how this forward-thinking company leveraged its demand signal expertise to become a customer-focused organization, which touts supply chain as a core competency.
VTech's supply chain transformation began in 2000 when it set in motion a corporate initiative to establish supply chain management as a core competency. The challenges were many as inventory turned three times per year; forecast accuracy measured 55 percent relative error; on-time order shipment performance equaled 55 percent; and manufacturing lead time totalled 140 days on average. The company also lacked a formal supply chain organization to manage the value chain. Plus, an internally-focused sales and operations planning (S&OP) process drove forecasts that were almost always wrong, while manufacturing proceeded in a completely different direction. There was no question that VTech was in need of a supply chain intervention.
"Excellence in supply chain is a significant profit impacting process -- it effectively determines if the business can deliver sustainable growth in profits," says Kedar Kulkarni, director of supply chain, VTech Communications Inc. "There were two very specific goals we wanted to accomplish at the beginning: reduce inventory and improve customer service levels."
In 2000, VTech decided to centralize all demand planning and inventory management. "We implemented a supply chain-led, cross-functional review of account plans and supply plans with the sales and product marketing departments -- essentially the traditional internal S&OP process," says Kulkarni.
VTech then linked its supply chain directly with its customers' logistics organizations to streamline planning with greater downstream visibility. It also adjusted the global supply chain to operate on a weekly cycle rather than a monthly one while simultaneously implementing high-mix lean manufacturing to support more granular planning.
These changes are what effectively transformed VTech's supply chain from an inside-out push model to an outside-in, pull-driven process. The company saw an immediate improvement in relative forecast error from 55 percent to about 25 percent with an absolute error of 60 percent.
"However, additional improvements refused to accrue as we seemed to have hit a wall," admits Kulkarni.
At this point, VTech began formal CPFR engagements with a few of its top retailers in 2001. This phase involved abandoning spreadsheet-based runrate forecasts and introducing standard EDI-based forecasts directly from retailers. This enabled VTech to better examine the quality of retail replenishment plans and understand that forecast error would likely continue to be high unless order forecasts from retailers were normalized against retailer on-hand inventories.
The implementation of rapid prototypes for POS-driven replenishment planning came in 2002. After reporting dramatic improvements in forecast accuracy and fill rates, the company began shopping for a demand planning solution and chose Oracle Demantra in 2003 due to its statistical forecasting ability and customizable data model. The implementation of a single instance of enterprise resource planning (ERP) and a data warehouse (also provided by Oracle) quickly followed in 2004. A business intelligence engine from SAP Business Objects was then deployed to highlight performance exceptions.
With a strong applications environment in place, the company switched its focus to implement predictive trade promotion planning in 2006 to leverage the rich POS data and promotional history it had collected. In 2008, everything came together when VTech leaned on a variety of solutions to create a S&OP process around a systematic monthly review of demand and supply plans.
"The key transformation in our supply chain processes has been a strategic shift from an inward-facing organization to a customer-focused one," says Kulkarni. "Our overall vision is to create a seamless integration with our retailers, distribution partners and end consumers. We have seen a dramatic improvement in our ability to sense and shape demand and turn shelf-level insights into timely and profitable supply responses. This ability has led to a significant improvement in our customer service levels while driving down operations costs."
Doing Data Differently
What makes VTech's systems infrastructure unique is the fact that all of its applications, including its ERP system and the corresponding operational planning and transactional data, are inherently integrated into one enterprise data warehouse rather than being fed into or stored in an ERP system. This leads to a single source of truth.
"This forced integration leads to world-class communication," says Delany, meaning that insights based on nine years of retailer POS and promotional history are now ingrained into VTech's planning processes in sales, supply chain, pricing and product management. This allows for rich modeling of promotions, new product launches, product seasonality and retailer in-stock levels.
"We have a culture where POS data is considered the Holy Grail in all planning. Weekly retailer POS and inventory trends are viewed across the company and even at our senior management and board meetings," says Kulkarni. "In fact, the weekly POS report is the most widely circulated and read report in the company."
This one invaluable report, which contains a series of charts that compare current POS information at any degree of desired granularity to historical information for the prior year (or any year for that matter), has undoubtedly helped VTech to forecast and adapt to changing consumer trends in a floundering economy.
With weekly POS data on hand, VTech can modify its operations in the middle of the cycle should red flags be raised. For example, a leading national retailer, during the course of collaboration with VTech, insisted on forecasting its business with reduced seasonality and consistently under-forecasted POS to justify its lower purchases budget. Due to predictive models in house, VTech was able to quickly identify the inconsistency in those forecasts. In an S&OP meeting, VTech decided as a company that its data was compelling and brought in the incremental inventory anyway. When the customer's inventory levels dipped due to consistent overselling, VTech got the call it had been waiting for. The customer was in desperate need of additional inventory, and VTech just happened to have the right mix waiting to ship.
"This validated our predictive modeling," says Kulkarni. "Without the ability to sense and respond to POS and shelf inventory, we would not have had the confidence to bring in this inventory. We would have probably lost the business to competition and risked being labeled an unreliable supplier."
Delany says that VTech can also now predict Black Friday and Christmas sales within 5 percent forecast error, which helps the company avoid spending money liquidating extra inventory.
What's more, within the first 90 days of a new product launch, VTech can keep a new product in stock on any retail shelf 90 percent to 95 percent of the time on average.
"We've taken the excitement out of forecasting. We know exactly what is going to happen," says Delany.
And operational improvements aren't expected to slow down anytime soon. Next, VTech plans to integrate even more granular downstream data into its S&OP process, looking at retailer distribution centers and stores to reduce out of stocks and using store demographics to improve product and pricing decisions. The company also plans to model multiple, adaptive supply chain networks to support corporate growth and diversification goals.
"Our vision is to further extend downstream data into a company-wide data asset," closes Kulkarni.
Success by the Numbers
VTech's total supply chain transformation resulted in:
- Improved POS forecast accuracy to 12% absolute error 12 weeks out
- Reduced finished goods inventory days by 30% and dollars by 50%
- Reduced revenue leakage in excess of 5% of net sales
- A more than doubled market share
- Improved on-time and complete orders by more than 50%
Supply Chain Checklist
VTech's Director of Supply Chain Kedar Kulkarni is charged to lead continuous improvement in demand sensing and shaping. Here, he shares tips that you can apply to supply chain initiatives:
- Take an outside-in view of the entire value chain and get visibility as far down the channel as possible to drive demand sensing
- Create a continuous buzz and culture around point-of-sale/downstream data with visual reporting
- Conduct S&OP meetings with customers directly instead of guessing what they might order
- Model your upstream supply chain to make it the most responsive within the constraints imposed by the network
- Drive a single source of truth across functions to enable a meaningful S&OP
- Model risk in your supply chain design to ensure multiple demand scenarios can be met effectively at lowest cost/risk