Alltel Wireless has come a long way in strategic sourcing in a short period of time.

In just a few years, sourcing at the Little Rock, Ark. wireless services provider went from being highly decentralized with individual business units buying as they saw fit, to a centralized Procurement and Logistics Group (PLG) that leverages the company's $4.1 billion annual spend with suppliers.

The group, under the leadership of senior vice president Scott Searls, has instituted spend management, set strategies for more than 40 different commodities and implemented a scorecarding process for suppliers. It has also improved forecasting and worked very closely with engineering and marketing to make sure Alltel meets the needs of its 13 million wireless customers.

Centralization and strategic sourcing resulted in immediate savings and process improvements. During the first year of Alltel's strategic sourcing efforts in 2005, it cut transportation costs by 11%, reduced inventory of network equipment by 11% and improved order accuracy by 11%. PLG saved the company $435 million in materials costs in the first year and about $1 billion in two years.

The company also reduced its number of suppliers while improving its communications and relationships with strategic and preferred suppliers.

Searls says spend analysis was the linchpin for the strategic supply initiatives that the company instituted. "We started with the belief that anything we can measure, we can improve," he says. "We dug into a complete analysis of the company's spend."

That was no easy task since purchasing was done by about a half dozen different business units and it was not obvious what Alltel bought from a specific supplier.

"Every organization was acting independently," says Searls. "They were cost conscious, but they didn't appreciate that decisions they made might have a cost impact on other organizations."

It took a long time to figure out what Alltel was spending with each supplier. "We spent hours with engineering trying to understand what we bought," recalls Catherine Shackleford, former vice president business process for PLG. "It was frustrating because we didn't know what we bought from say an IBM."

She says Alltel had a lump sum number that it spent with suppliers. "Unless you went from invoice to invoice you wouldn't know what we bought from Fujitsu or EMC, or any other company," she says.

Alltel began its spend-analysis journey by categorizing its spend by supplier. "That's useful because in our industry most suppliers are reasonably specific to a commodity," says Searls. "So if you're buying something from Xerox it is probably a copier."

Searls says it was a painstaking process because buyers had to ask the business groups a lot of questions about spend with suppliers. After months of research, Alltel found its spend could be categorized in five areas: indirect spend, networking equipment, IT hardware and software, and access sourcing. (Access sourcing is sourcing for long distance connectivity, local access and for regulatory tariff services).

Alltel set up five supply groups to manage those categories. The groups include handsets, information technology, corporate supply, network supply and telecom services. These groups manage individual commodities that fall under the broader categories. For instance, the IT groups would cover software as well as hardware, including router, servers and storage devices. Each commodity was assigned a lead buyer.

With the spend-analysis data and an organization to manage suppliers and spend, Alltel found that it could strategically source. "In each commodity we had all the sources we were using and the amount of money we were spending," says Searls. That information helped the commodity groups develop strategies for each commodity which Alltel did not have.

"We did not have forward thinking strategies before," says Shackleford. "Purchasing was very reactive. A business stakeholder said 'I want to buy storage' and we said 'OK let's go out and source storage,' but there were no future plans, no strategy around storage or other areas," she says.

The data helped Alltel determine how many suppliers it needed for each commodity, what kind of support it needed and which suppliers it should use.

The spend-analysis information also gave the PLG group leverage with suppliers.

"The power in sourcing comes in two ways," says Searls. "You have power when you take business away and you have power when you give business to someone," he says. Decisions were made on which suppliers to keep for the various commodities.

While figuring out how much Alltel spent with suppliers in the past was important, so was forecasting how much it would spend in the future. Alltel buyers also work with marketing and engineering to forecast what Alltel will need to satisfy customers.

For instance, Alltel buyers will interview people in the IT department about how much hardware they will need in the coming year.

"If you ask someone a specific question they may say 'I have no idea what I'm going to spend on IT hardware next year,'" says Searls. "Then we ask 'well do you think it will be more than last year.' They say 'oh yeah it will be more than last year, 10–20% more, probably about 15%.' All of a sudden you know. Then you dig a little deeper. 'Do you think that is mostly servers or direct-access storage devices?' They say 'oh, it'll be storage.' All of a sudden you have it."

That information is enough to present to a supplier and leverage spend. "You don't have to say I'm going to buy 387 servers, but if you say between 300–400 servers, you can have a real conversation with that supplier," says Searls.

Another forecasting example is antennas that Alltel buys for its cell sites. Engineers may say they have no idea about how many antennas they'll need because they don't know how many cell sites they plan to build.

However, they know how many they built the previous year and they know they will build 15–20% more in the next year.

There are five antennas per site so if you know how many sites, you know how many antennas, says Searls.

Purchasing is also involved in handset forecasting. That is a bit trickier because cell handset models change so quickly. There are also various tiers of phones ranging from low-end inexpensive voice centric phones to high-end personal digital assistants. Alltel buys about 20 different phones from all the major handset manufacturers.

Purchasers meet once a quarter with marketing to figure out how many handsets in each tier and which suppliers it should buy from in each tier.

Spend analysis and forecasting future requirements to suppliers weren't the only initiatives that Alltel has instituted. Another key move was issuing scorecards to Alltel's strategic suppliers. Alltel has four categories of suppliers: strategic, preferred, allowed and restricted, but generally only issues scorecards to its 35 strategic suppliers.

About 67% of Alltel's spend is with strategic suppliers, which include networking equipment companies, long distance services providers and network software providers. "I'd like to say companies have earned their way to be a strategic supplier, but not everything you buy is strategic. Every supplier is not strategic," says Searls. "We buy about $1.4 billion in handsets from LG, Samsung, Motorola, Kyocera and Nokia. Are they strategic to us? Well, handsets are strategic because we put them in the hands of customers, but we generally consider handset suppliers to be preferred sources."

He says strategic suppliers are ones that are embedded in Alltel's operations and their failure would be costly to Alltel because switching business to another supplier would be very high. With handsets, it would be easy to move sourcing around. Networking equipment billing systems or anything that results in revenue generation is strategic.

"Strategic suppliers are also companies who we believe that if we spend time to invest in the relationship with them, we will get more out of it. They bring us a differential advantage, a competitive advantage," he says.

Searls says using scorecards with strategic suppliers is one way Alltel invests in supplier relationships. The idea of the scorecard is to improve communications with the strategic supplier as well as performance. The scorecard metrics can be different depending on the supplier. It's not always just about quality, delivery and cost.

"It is a mistake to try to apply the same scorecard to every commodity and every supplier relationship," says Searls. "You have to be willing to change what you measure and how you weight the criteria. You have to figure out what is important to the success of the relationship between Alltel and the supplier," he says. "Is it the accuracy of software, the system up-time? For others the question will be, 'Did the supplier deliver the thing on time at the right price?' "

Searls say what Alltel buyers look for is continuous improvement on scorecards. "You don't get upset with the exact score because it is not an exact science," he says. Alltel buyers want to see quarter-over-quarter improvement with suppliers.

"That's the measure of an effective scorecard," says Searls. "Whether a supplier's score is 48% or 78%, I don't get hung up on the number. I get hung up on the lack of progress. We want continuous improvement."

The true value of the scorecard is that it "gets the moose on the table so all the supplier management issues are out there" so Alltel knows where improvement is needed, says Searls.

The lead buyer for the commodity is responsible for the scorecard. Once a supplier is rated the scorecard is sent to the leader of the supplier who has profit and loss responsibility. Also on the scorecard are areas that Alltel wants to see improved in the next quarter.

"We try to create a forward agenda and not just focus on the history. We want an ongoing dialogue for improvement with suppliers," he says.