Snyder’s of Hanover: Twisted Logic

It took Snyder’s of Hanover more than 94 years to become the world’s second-largest pretzel maker.

But now the company, a family-owned and -operated firm headquartered on the outskirts of the Pennsylvania Dutch Country, is in the middle of a three-year project that will revamp the way it manages, analyzes and collects financial data, a system that for years has left its finance department tied up in knots.

The problem: Like clockwork, the company’s financial analyst, Lois Stambaugh, would spend every working hour of each month’s final week collecting Excel spreadsheets from more than 50 department heads scattered worldwide.

Then she’d consolidate and re-enter all the data into another Microsoft Excel spreadsheet. The consolidated spreadsheet would be the company’s Bible, the document that would serve as the de facto monthly profit-and-loss statement for a company that sells more than 78 million bags of pretzels, chips and organic snack items each year.

It got even worse at the end of each fiscal quarter, when three months of data had to be collected and consolidated, but that paled in comparison to the year-end profit-and-loss statement and the annual budgeting process. That’s because any corrections or additions at any point during these report-closing periods would require Stambaugh to return the edited spreadsheets to the department heads and then wait for them to be resubmitted to her for further consolidation.

Information about last-minute sales or returns—or even an unforeseen expense created when, say, a delivery truck broke down on a route—had to be resubmitted on yet another revised spreadsheet which, in turn, was re-entered into the master document.

It was the paper-based system from hell.

“We weren’t getting the information we needed efficiently or at a detailed level,” says controller Sean Grim. “And we didn’t have this information available to us immediately and at our fingertips.”

Data such as the weekly or daily sales figures from individual delivery routes across the country couldn’t be accessed. The numbers that management needed to make adjustments—be it in headcount or manufacturing to meet higher demand—only came in at the end of the month. They couldn’t ramp up production quickly enough when necessary. They couldn’t identify or eliminate unprofitable routes or product lines immediately. They had to wait for weeks or months to get the relevant numbers.

Not only did Snyder’s waste a lot of time gathering this financial data, it was occasionally plagued by errors, small and large, created by intensive re-keying and editing of the data.

Another problem that Snyder’s hopes to rectify with the new system, once it’s in place: its inability to get sales reports on how individual products are faring each month and each year. Right now, chief executive officer Michael Warehime and his executive team can track the gross profits of business units. But they can’t track performance of each of the company’s 4,500-plus products that are spread across more than 150 product lines.

“We don’t have the detail to know whether a certain flavor or style of pretzel is actually making money or losing money,” says Grim. “We only see it at a high level within a larger category group. That’s something we want to change.”

The varied brands that Snyder’s produces include items such as Old Tyme Pretzels, Homestyle Pretzels, Sourdough Hard Pretzels, Jalapeño Pieces, Honey BBQ Pretzel Pieces, Butter Snaps, Pretzel Rods, milk-chocolate and white-fudge Mini Dips, cheese minis, oat-bran sticks and many, many others that can be found on grocery-store shelves or in vending machines.

Information about sales is especially crucial to a company like Snyder’s because of its extensive distribution system. Along with its snack foods, Snyder’s also delivers products from other vendors, such as Tasty Baking Co.’s Tastykakes.

So Snyder’s turned to Conshohocken, Pa.-based software vendor Satori Group, then only a year old, to put an end to its reliance on paper reports and off-the-cuff spreadsheets for reporting financial results.

In doing so, Synder’s had two goals in mind: Cut labor costs involved in creating financial reports and obtain information on a timely basis to identify and eliminate money-losing brands. After all, it wanted to make a stab at displacing Rold Gold, a division of PepsiCo’s Frito-Lay, as the market leader.

At first, in November 2002, Satori Group chief executive officer David Libesman met with Grim and the Snyder’s staff and pitched his company’s proCube software. In two days, Libesman’s team created from scratch a profit-and-loss statement for one of Snyder’s divisions. “There was a ‘Wow’ factor,” Libesman says now. “They were used to dealing with a lot of pain extracting [sales and expense] data from their system. We were also able to show them that they could continue using Excel spreadsheets as they had before, but proCube would collect, consolidate and organize the data for them in a manageable way with far more detail.”

Now that the first phase is complete, Snyder’s gets companywide profit-and-loss statements generated by the proCube system and Lois Stambaugh spends no more than a day or two each quarter preparing the financial results. That portion of the project took 90 days to complete.

Next up, Snyder’s will get profit-and-loss statements at the product level, not just a companywide picture. “Right now, to get the detail we need at the brand level, we’re in the middle of a companywide business-process review so that we can figure out exactly what data we want and how we’re going to collect it to complete the second phase,” says Dave Thomas, director of information technology for Snyder’s.

Once the complete profit-and-loss system is in place, the Snyder’s team will be able to compare not only the number of Garlic Bread Nibblers produced, but also how many were sold in a specific week or month, and which particular stores and delivery routes are best servicing the garlic-loving end customers.

This information will help the company better manage production and distribution of its products. It can be sure that better-selling products get to stores faster and in higher quantities.

In a project that helped lay the groundwork for this initiative, Synder’s in 1999 implemented J.D. Edwards enterprise software to manage its manufacturing and distribution data. This system now feeds manufacturing data, such as the number of units of Garlic Bread Nibblers baked during a production run, to the proCube software.

To improve financial reporting, department heads across the country will be able to enter daily sales and returns information into Excel spreadsheets, which will then be fed back into the proCube software for consolidation. Management can then extract the data they want—such as how many bags of pumpernickel-and-onion pretzels were sold in California last week—and project sales and budgeting needs for that unit for the next quarter or year.

“They use Excel and there’s nothing wrong with using Excel,” Libesman says. “But they were using it as a two-dimensional tool. It’s very difficult to create meaningful data from disparate, disconnected spreadsheets when you’re trying to do linkages and interact with multiple databases.”

For the profit-and-loss analytical system, Satori implemented a data warehouse that runs on a Microsoft SQL server. The warehouse serves as a repository of data collected from disparate sources. The data is then consolidated and organized before it’s pulled into the proCube software for reporting and analysis.

Looking ahead, Snyder’s plans to develop a corporate portal to allow department heads and executive managers to view sales results and distribution plans with a single click. It will also help executives size up cost-cutting opportunities on demand instead of waiting until the end of the quarter or year before ramping up production of a popular pretzel brand or paring distribution routes where necessary.

The Satori implementation has cost between $200,000 and $300,000 for the first two phases and will likely top out at somewhere between $400,000 and $500,000 once it’s completed in late 2004 or early 2005, according to Grim, the controller.

At this juncture, Snyder’s real challenge is determining how it is going to use the technology to improve profitability.

Much has changed in the pretzel industry since family patriarch Harvey “Gramp Harry” Warehime founded the Hanover Pretzel Company in 1909, baking his pretzels and selling them to markets in Hanover, Pa., and other nearby towns.

Today, Snyder’s of Hanover has roughly 40 distribution centers throughout the U.S. and Europe and annual sales in excess of $200 million. In addition to chips and other snack foods, it sells more than 30 different types of pretzels.

In 2002, U.S.-based pretzel makers sold $1.28 billion worth of pretzels, according to market-research firm Business Trend Analysts Inc. in Commack, N.Y., making the salty snacks the third-most-popular choice in the $22.5-billion snack food industry, trailing only potato chips ($6 billion) and tortilla chips ($4.5 billion).

Consumers are buying more pretzels than ever before. In 1997, the average American ate 2.8 pounds of pretzels. Last year, individual consumption averaged 3.4 pounds and is expected to reach 4.2 pounds a year by 2005, according to Business Trend Analysts.

To capture these consumer dollars, Snyder’s had no choice but to invest in a software system that would change its business model from one dependent on forecasts to one that’s more demand driven.

“This software gives them an analytic tool they didn’t have before,” says Eric Austvold, an analyst at AMR Research Inc. “They’re making a push to become more intimate with their customers and figure out which product lines are stars and which are dogs.”

Grim says company executives believe there are specific brands that have been money-losers for quite some time but the company hasn’t had the data to verify its assumptions. “I’m sure once we get this brand-level system up and running, we’re going to get more information than we ever thought we’d need,” he says. “It’s going to change our whole philosophy.”

Does that mean Snyder’s could end up killing off some of its offerings?

“With between 150 and 200 different brands, we’re past the point of estimating. Some people may not be very happy with the answers it’s going to give us,” Grim says.

To make a serious run at industry leader Rold Gold, Snyder’s knows too well that it needs a more modern and detailed view of its financials and day-to-day operations. It’s not enough to simply make a good-tasting pretzel.