INDUSTRY	Food & beverage
COMPANY SIZE	$3.9 billion
ERP/GENERAL LEDGER	SAP
BUDGETING, FORECASTING & ANALYTICS	Applix TM1
SALES VOLUME REPORTING	Cognos
FINANCIAL CONSOLIDATION & REPORTING	Applix TM1

For far too many companies, major business decisions are made by trial and error. Whether the organization launches a new product or where it focuses marketing efforts or how much it spends on customer service depends primarily on the management team's guesswork. What if a business operating in this way were given access to solid data to support these choices? Cadbury Schweppes Americas Beverages has made scenario planning a priority, and the company is reaping big rewards from that approach.

BPM: What led you to look for better scenario planning capabilities?

Bill Primerano: Back in 1998, when we were part of the Snapple Beverage Group, all of our complex financial modeling and forecasting was done in Excel. Everything was one spreadsheet linked to another spreadsheet. If you got a bug in one of them, it ran rampant throughout all your schedules. At the time, Snapple had 13 different legal entities trying to do forecasts on all the different combinations of customers and products that we had at the time. It was just a nightmare to do it in Excel.

That was what led us to start looking for a business intelligence tool. We needed a tool that would let us change the dynamics -- from which customer to which SKU or which product they were buying -- to do what-if scenarios. When it came time to do long-range plans or investment modeling, we needed the capability to say, "What if our business were to grow 8 percent or 10 percent? What would this look like?" And to do it on segments of the business, so we could look at Snapple versus Mistic versus Royal Crown Cola. We needed something that we could use as a data repository. We decided on Applix TM1.

BPM: Could you describe some of the analyses that you're able to do now that weren't possible with Excel?

Primerano: You can do what-if scenarios in Excel, but it's going to take you longer to break those formulas down to all the different segments and apply only certain characteristics of the brands. One of the analytics we do is the impact of a price increase. What if we wanted to do a price increase only on 16-ounce products? We'd have to gather all the 16-ounce products on an Excel schedule and then look up all their current pricing, then apply a dollar amount increase or a percentage. We'd be spending hours just setting up the Excel schedule with all of the SKU data, the customer data. Then what's our selling price? What's our cost of goods? If we increase the price, we may have a loss in volume, so we'd have to look at what portion of the volume is at risk at what time frame. Then we'd back that out and determine the variable costs associated with that lost volume. Are we going to forgo paying any trade dollars out?

When you look at the impact of taking a 25-cent price increase, you can't just say, "I'm going to take all my volume from one time frame to another time frame and add 25 cents." You have to look at all the other costs that are coming out of the system. What are some of the trade dollars that you may have to put into the system because not every customer's going to agree to that price increase? We're doing things like that faster and easier.

BPM: How much faster does a BPM system make the collection of this sort of data?

Primerano: Days. It's a matter of being able to get the information into a format that's usable. Depending on your skill level within Excel and the types of data sets that you need to get from various systems, it could take you days to put a schedule together. We've got SAP feeding into TM1. We've got data from the Cognos volume forecast in TM1. We've got the ERP system of our Snapple Distribution Company, which is Award, feeding into TM1. We've got one place to go to get all of our data, as opposed to having to go to four different systems. Tie all that together, convert it so now it's all on an equal platform. That's how we've gone from days to within an hour to have a complex schedule going.

BPM: What are some of the benefits from saving time?

Primerano: It helps us improve our margins; it helps us with accountability. We're testing unlimited possibilities of products and production facilities and customers, and it gives us the insight to strategically position our brands in the marketplace. We've got the exposure to the category size of our products by different package sizes. Snapple's more than just a 16-ounce bottle. You also have gallons, 64-ounce, 32-ounce, cans, which go into different areas of the marketplace.

Not only do we manufacture, but we also own distributors. Distribution centers are our face to the customer. When I produce a bottle of Snapple and sell it to distributor XYZ down the road, I don't know who their customers are. If I sell the Snapple Distribution Company 10 pallets of Snapple, I'm able to see where they end up. I can see how much ends up in the pizzerias, how much of our product is sold through delis, how much goes into the school systems.

By having something like TM1 in place, I can then capture that sale to the end customer and determine the route to market of our business. Do we derive more profit by selling into the grocery stores or the chain stores? Do we get more profit by selling into the delis and the schools? Do we get more profit by selling to the club stores? That helps us get more insight into the drivers of our business.

BPM: How does the finance team use this analysis capability to support senior management, marketing, and the business units?

Primerano: They'll come to us with an idea, and we'll go into TM1 and start doing what-if scenarios. For example, if we're looking at a specific customer deal, we'll pull information on the brand's performance in prior periods. That's our do-nothing scenario. Then we can start layering in what-if scenarios. If we were able to take on this particular new customer or get this much leverage from this other existing customer, it's incremental volume. And what would that cost be? Then we layer in the additional trade dollars that we may have to give to leverage that volume call. And then we just do all of our scenarios. We say: Well, what if we were to decrease the price by 25 cents for this customer? What would that add in incremental volume? What would that take as far as cost of production? Will that put a capacity constraint to grow this volume at this level this fast? Are there going to be any additional charges incurred because of line time at the production facilities? All the way down to: Are our dollars working better for us on the operating profit line?

BPM: And then you would go back to the business unit manager or the marketing department with your analysis?

Primerano: We have a team -- a gatekeeper team that's made up of senior leaders of finance, the operations group, the marketing group, sales -- and we present that financial analysis to them, as well as the strategy behind the proposition. Whether it's taking a pricing action to a specific customer to gain top-line and bottom-line growth or it's coming up with a completely new product, we present how it can add value and profitably grow the business.

If it's a proposition that's not going to grow shareholder value, we're not going to support it. But we wouldn't know that unless we were able to do what-if scenarios.

BPM: What would you say to an organization that doesn't do this kind of analysis?

Primerano: When we were at the BPM Summit last year, everybody was talking about, "We spend more time gathering data and mining through different systems to pull this data together, and that leaves us very little time to analyze it." Because everything is now in one system, we've got a single source of the truth -- which is something everybody talks about, but we really do have that with this application. By having that one source of the truth, you're able to get into things faster.

Whether you do what-if scenarios or not, just the functionality of being able to forecast with a greater level of accuracy is worth the investment. You can take your specific pricing tables and your exact costs and put them into the system. You can then match up your customers and your products against those selling dollars and those costs that are associated to that exact customer, to the exact product they're buying from the production facility that makes it. If you have that information at hand and spin it around, you're going to be 10 times better at growing your business and understanding the drivers of your business than somebody who's in Excel that just uses an average rate.

If I see that on average it costs us $12 to produce a case, and on average we sell it for $20, that's not really helping me if the case actually costs $8 to produce on the East Coast and $15 to produce on the West Coast. If we change so that more of our volume is sold on the East Coast, we're going to grow our business. If we grow our volume in this geographic location, it's going to be cheaper for us to make it, it costs less in shipping, and there is more of a demand here that we haven't tapped into yet because we haven't focused on these markets.

Taking Excel and using an average, which is what a lot of companies do, does not give you that exposure. That's the functionality that we're getting out of our system today. We're able to understand our trends. That provides a big benefit to senior management.