It wasn't supposed to take a year. Many software vendors will tell you it shouldn't take nearly that long, either. Hartman says he and his team ran into a major snag that added an additional seven months to the software's roll-out—something that he calls dirty data.

"The main issue that we're seeing that I think companies will face is what the industry calls ETL: extract, transform and load," he says. "The vendors want to show you the analytics and the dashboarding and all that stuff because everybody really likes shiny things, and that's what they are. What they don't show you is the greasy stuff that makes the shiny stuff work, and that's the ETL."

The extract, transform and load function is often handled by a service provider such as a data warehouser, which basically "cleans up" the data so spend-analysis programs can process it and give back useful information. Without proper data, a spend-analysis program is useless.

"It's going to take you a long time to collect your spend data in a way that your classification and analytics software folks can make sense of it," says Hartman.

Four months after procuring a spend-analysis program, Boston Scientific gave the program its first run. Hartman and his team weren't happy with the results. The data Hartman's team tried to funnel through their spend-analysis program was basically too "dirty" to get much out of it.

He lists a few of the dilemmas Boston Scientific faced in cleaning up their data: multiple names for the same vendor, redundant material numbers, fragmented Pcard and T&E card information, PO and AP spend information, divisions not in the company's central ERP system, and multicurrency problems. All these different problems made for a particularly tricky ETL challenge to Hartman and his team before they could get any quality information out of their spend-analysis program.

"You put all of that into a pot and stir it around and at the end of the day, your software provider needs a file that they can understand," he says. "It's not simply an IT exercise, because you have to understand where you want to go, what you want to see and how you want to see it. That will dictate the data that you need to extract and how you will need to normalize it."

At its simplest, making sure you're ready for a spend-analysis program is a question of doing your homework before you dive in.

"You need to know the answer before you start asking the questions on the data. You have to do it backwards," Hartman says. One example he cites is measuring compliance. How do you quantify compliance? What materials and services will you count? What do you do if you're buying from a vendor both compliant and non-compliant items or services? If this information isn't stored on your ERP system, how do you begin to measure that at all?

Hartman adds that many of his suppliers assured him they had ETL tools that would help Boston Scientific clean up and compile its data, but the company still had difficulties. Thankfully, Hartman says, Boston Scientific's spend-analysis software vendor stepped up and helped the company with its data difficulties. "The vendor was pretty up front, and gave us the file structures, then sat down with us to go through them—they were very good about it," he says. "We went into this thing with our eyes wide open and despite that, we still didn't think through all of the complexities that we needed to." Every buyer needs to be cognizant of these hidden time costs.

Although it snagged Boston Scientific's spend-analysis roll-out, highly detailed and organized data is exactly what appeals to many buyers, including Pat Walker, vice president of finance and top procurement executive at Walker Machinery, in Belle, W.V. His company's raw spend information was about as low-tech as it could get: paper purchase orders.

"We knew how much money we were spending and who we were spending it with, but not what exactly we were buying," he says. "All that information was locked into the paper orders. We needed information like quantity, price and SKU." This dearth of detailed information made it difficult to compare suppliers to see if Walker Machinery was getting the best possible deal. "If we knew exactly what we were buying, then we could take that activity and all that detail, give it to suppliers and ask what it'd cost to do business with them."

Hartman: “You need to know the answer before you start asking questions on the data.”

"Definitely be more conservative than software vendors," he says. "When you start talking about software productivity dollars, it can be a tough sell. If I was on the other side of the table, I'd be very skeptical and question everything."

For his business case, Walker made a comparison between the paper-based system and the potential with spend-analysis software. "We came from a traditional paper-based PO system, about as freeform as you could get," he says. When his company made a homegrown electronic PO system, it wasn't much better—just a digital version of the paper method.

"Six Sigma really highlighted how handicapped we were in capturing supplier spend," he says. After the company's switch to spend-analysis software, Walker Machinery now has a much easier time identifying high-volume suppliers, a powerful negotiation tool. "It wasn't plug-and-play," says Walker, but was a welcome improvement.

One tip in building a good business case is to determine your needs by looking at where your current capabilities are lacking, says NPI consultant Jeff Muscarella. "Look at the existing modules in your ERP system and the organizational standards in your company—where's the gap? Build your business case and determine the return on closing that gap." One possibility says Muscarella, especially at smaller companies, is to be resourceful and "fill in the gaps with your own solutions," though he cautions this isn't always successful.

When it's not, there are plenty of commercial options to choose from, as you'll see on page 45. The key to deciding which works for you: a detailed understanding of your own business process—and a keen eye for those hidden costs.