Benchtracking Purchasing Performance - A Management Tool
Joseph J. Schoen, Jr., C.P.M.
Joseph J. Schoen, Jr., C.P.M., Manager, Planning and Process Improvement, BASF Corporation, Mount Olive, NJ 07828, 201/426-3320.
82nd Annual International Conference Proceedings - 1997
Need For Action: Because of today's competitive business environment, a purchasing organization must assure itself and its internal customer base that the costs of procured products and services are cost competitive and contribute to the overall corporate competitive position. Meeting this requirement presents real challenges since determining the competitive cost position is extremely difficult, time consuming, and expensive. Internal benchmarking is virtually impossible because of understandable constraints. Benchmarking by third parties may provide an answer but is expensive and only documents a position at a given point in time. The "Benchtracking" approach described here provides a tool to directionally track competitive costs when firm figures are impossible to obtain. Benchtracking's objective is to determine the comparative direction or trend, not the absolute position. Once this analysis is completed, measures are taken to confirm and continue the procurement strategies or take required corrective actions.
Background: For a business where purchased raw materials and services represent a significant part of the total cost of production, purchasing and business management must understand in depth this critical part of its competitive position. For example, at BASF Corporation raw materials represent about a third of the cost of sales, so its importance is understandable. In the past, some purchasing organizations have reported some discounts as cost savings when these were really available to all buyers as industry standards and do not represent real competitive advantages. The demand for hard data and an understanding of the real relative competitive position provides either a credibility problem or opportunity for the purchasing organization depending on the outcome.
Once the objective of understanding an organization's competitive position is clear, the challenge to determine it becomes obvious. Exact knowledge of a firm's raw material cost structure relative to its competitors is virtually impossible to obtain. The most obvious difficulty is the unwillingness of competitors to share their costs because of the competitive, legal and confidentiality problems encountered. Legal constraints and basic business practices also prevent competitors from sharing sensitive pricing and other cost data.
Recognizing that benchmarking by the organization itself is virtually impossible, obtaining a benchmarking analysis and report developed by an experienced third party firms may provide an answer. However, these studies are expensive and must be repeated periodically since they really represent a competitive position only at a given point in time. Additionally, we have found that even some respected benchmarking firms have been unable to determine the competitive position of several major "core" raw materials that represent major product cost contributions. As expected, our competitors are apparently as wary as are we of sharing this very sensitive data.
Outline of approach: To address this dilemma and challenge, we first outlined our objective to develop an analytical tool. First, we needed to define a "Marketbasket" of purchased products and services that would represent our overall purchased dollar volume. This basket should mirror the breath or variety of products purchased since BASF buys products at multiple stages or levels of production ranging from basic chemical building blocks (e.g. benzene or natural gas) to products well down the manufacturing chain (e.g. resins and pigments). The marketbasket should also capture the major dollar items that are procured. Hopefully the "80/20" rule would prevail in our selection. The next step was the identification of published or reference prices that could be lined up against the basket items. These reference prices must track market movements. Here the key is to capture trends, not necessarily the actual net prices. Next, a base period and definitions of measuring frequency were defined. Finally, an analytical model consisting of a database and spreadsheet were developed to run the calculations and provide a summary of trends and indicators.
The approach described in this paper has been used for nearly a year. Internally it is now known as the "Purchasing Performance Report" and has been well received by our internal customer base, financial management and by Purchasing itself.
Product selection for the Marketbasket: We prepared a list of purchased products to select a key portion of our purchased raw materials. The preliminary list of core products represented about eighty percent of our total raw material expenditures. To finalize the list, we reviewed it with key business representatives and our financial management. This refined the list as representative of our purchases and inclusive of the products that act as a barometer of what happens in the marketplace. These fifty products has become known as our "Key Raw Materials". We noted that our analysis includes only purchases from third parties (i.e. excludes purchases from our parent company and affiliated subsidiary companies). These transactions are viewed internally as transfers between worldwide businesses.
Reference price selection: This part of the project presented the biggest challenge. The fifty (50) products selected for the marketbasket are sourced from a wide variety of markets within the broad chemical industry. They range from very basic products (e.g. natural gas, benzene and ammonia), to intermediates like glycerine and chemical intermediates and finally to further downstream products like resins and pigments. Each of these products belongs to a different market with distinctive pricing and supply/demand balances. We reviewed each product to determine suitable publications or reference sources that would track real market trends and pricing. Custom manufactured items were tracked via raw materials and an index to reflect the other costs. Potential reporting sources examined included trade publications, magazines and reports, government data and reports, and finally some multi-client reports written by respected industry experts. After initial review, we reviewed the pricing history and our personal experience to confirm that the suggested source and data actually represent what actually happens in the marketplace.
The final list of reference sources includes approximately fifteen different reports or publications. These range from publications like Purchasing Magazine, US Government FERC reports, state regulatory summaries, and recognized chemical industry consultant reports. As an example of a reference, the CPI Edition of Purchasing Magazine publishes market transaction on a monthly basis of seventy (70) different chemicals. We now gather the selected comparative prices monthly.
Base period selection: We reviewed price history and finally chose 1993 as our base year. This was done since it represented a timeframe in the industry where pricing and demand were relatively balanced and not strong influenced by long or tight markets. We did not want to start off with major market pricing shifts that could adversely prejudice the subsequent price developments and analysis. An earlier base period was rejected because it had the lingering influences of the previous recession.
Reporting Frequency: We reviewed the frequency that data should be collected and reported. We compared the benefits of various report frequencies versus the workload required to support these frequencies. Looking at annual, quarterly and finally monthly reports, we ultimately determined that a monthly basis would be best despite the workload it demanded. Various market pricing changes indicated that the monthly reporting would reflect the broadest base of monthly and quarterly prices. Reporting more frequent than a month wasn't found to be meaningful. The monthly report also coincides with business group report frequency. Finally, the monthly report keeps interest alive within the purchasing teams since other reports like cost savings are published monthly.
The key business factor that favored the monthly report is the need to quickly spot trends and respond promptly as market conditions warrant. While the monthly reporting initially involved a great amount of work, our staff and buying teams became familiar with the new tasks and recognized the importance of this new tool and its application.
Development of database and spreadsheet: This part of the project was very time-consuming but relatively easy to organize since the parameters of products, reference prices, base period and reporting frequency had been determined. An Excel spreadsheet was selected as the organization's standard. Base period data of BASF actual prices and the industry reference prices were loaded and set up with 1993 as 1.0 or "100" basis. Prices of subsequent periods were loaded and then compared back to the base period. The key measurement was to compare how prices change for both BASF and reference sources and then compare them back to the base period. Once the data is loaded, the comparative analysis is relatively easy to accomplish.
Published reports and graphs: Now that this tool has been in place, it is interesting to view the results. The key reporting products are several graphs which depict interesting performance and tracking developments. These are prepared monthly for key financial and business management groups as part of Purchasing's monthly reporting. Several theoretical results of these graphs using a fictitious Company A are included below. Data for further analysis is also available as needed, but for management reporting the graphs have been more useful.
For illustration purposes only, a theoretical case was developed for Company A. The first graph compares prices changes versus a base period. Assume that a 1993 base period was one of relative stability and that 1994 and early 1995 saw price increases for both industry and Company A. Later assume that prices declined. The results would look as follows:
(graphic not available in text-only version of this paper)
The second graph compares Company A's theoretical performance to the industry reference prices. Here the industry price changes are set to equal 100% and Company A's price changes relative to the industry are compared.
(graphic not available in text-only version of this paper)
The third graph compares Company A to the reference prices for the percentage change from the prior year. This relates to Company A's target of holding their price increases below the reference prices when prices are increasing or seeking a larger price decrease when prices are falling. This objective recognizes that market forces in effect will influence the overall trend, but Purchasing efforts should lessen increases or provide a greater opportunity when market prices fall. Note that the bars compare Company A to the reference index along with a target.
(graphic not available in text-only version of this paper)
Current use of the index: At BASF, we now collect the reference prices monthly and compare these to our own transactional prices. These are entered into the spreadsheet which also includes volumes of the individual items. The individual prices are extended by the purchase volumes and summarized to give an overall company view. As mentioned before, the Purchasing Performance Index is reported monthly and serves as a tracking device to measure our progress against objectives to minimize the effects of price increases or take advantage of price decreases. Further analysis of raw materials used by individual business groups helps them understand their margins and profitability expectations.
Within Purchasing, the data and reports are further broken down to team levels and used to assess the broad performance of individual Purchasing teams. Here we are really only looking for overall broad direction. As a tool, we are careful to avoid using the report as a "hammer," and prefer to think of it as a "flashlight." Many of the items are influenced by many mitigating factors. The tool is valuable for the light it sheds to better understand our position and provide how and where it needs to be strengthened.
For our business groups, we have also broken out the purchase price performance for their individual units and shared this data with them. This approach has been useful in their own analysis of their margins and overall business results.
Additionally, the analysis has pointed out some areas where our costs were out of line with the overall trend. Further review pointed out that specific locations or specific logistical positions may be what is causing the problem. This has been the first step in correcting the situation.
Leenders, Michiel R. and Harold E. Fearon, Purchasing and Materials Management, Tenth Edition, Richard D. Irwin, Inc., Homewood, IL., 1993, 479-481.
Lewis, Cheryl, "Chemical Transaction Prices CPI Edition," Purchasing, March 7, 1996, 32C19-32C20.
Moltlok, Tracy and Carolyn Pye, "Resource List - Forecasting Tools", Purchasing Today, June 1996, 32- 33.
Morgan, Jim, "Benchmarking is Not an Instant Success." Purchasing, May 23, 1996, 42-44.
Steinway, Diane, "One Purchaser's Quest to Benchmark." Purchasing Today, February 1996, 52-54.
Stork, Ken, "Benchmarking - What's in it For Me", Purchasing, June 20, 1996, 19.