Nestlé’s Cost Reduction in the Coffee Business

Nestle is the world-wide leader in the coffee business. They offer coffees at virtually all price points. They invented instant coffee in the 1930s. After the buffets of the commodity markets over the last few years, the company has created a global push to reduce its costs and to increase the quantity and quality of the coffee it buys.




We have found four generic approaches to reducing costs.



• First, reduce the rate of cost of a cost input.

• Second, reduce the cost inputs that do not produce output.

• Third, reduce unique activities and components in processes and the product

• Fourth, spread fixed cost activities over additional product output



Nestle is using the first three of these approaches in its world-wide investment in cost management.



First, Nestle redesigned part of the process. Its scientists developed a new generation of Robusta and Arabica coffee plants for Mexico. The Robusta beans are relatively inexpensive and make up the bulk of the beans in instant coffee. The Arabica beans are more expensive, harder to grow and go to the higher end coffees. Today, Nestle has planted 100 thousand coffee trees in Mexico using its newly designed coffee trees. Once this experiment is complete, the company plans to distribute 220 million plants to coffee growers world-wide over the next ten years.



The use of these new plants will enable Nestle to reduce its rate of cost for the beans it buys. The new plant design increases yields so it eliminates some inputs that do not produce the output of coffee beans. Many long-term coffee farmers are using older trees, which yield fewer beans and lower quality beans. Many of these farmers are leaving the industry since they cannot compete. This magnifies the commodity price problem Nestle faces. Nestle’s new trees fit the region’s climate. They resist disease and allow for larger and easier harvests. These trees will make coffee beans more consistently and predictably available. Nestle will give these trees to the farmers without asking for a firm long-term contract or ownership of any part of the farm. But it should be obvious that Nestle will engender a great deal of farmer loyalty with this program.



Nestle also expects to reduce the rate of cost it pays for its beans with two other cost reduction initiatives. It will offer farming and investing advice to up to ten thousand farmers world-wide. As these farmers become more efficient, Nestle’s costs will drop. In addition, Nestle will also increase the amount of coffee it buys directly from the nearly 170 thousand growers who produce its coffees.



This kind of foresight and innovation suggests why Nestle commands its market leadership.

Cost Reduction by Redesigning the Product

Over the last several years, we have studied many examples of cost reduction initiatives to improve productivity and create economies of scale. In simplest terms, there are four actions that improve productivity and economies of scale. First, reduce the rate of cost you pay for an input. Second, reduce the inputs that do not produce output. Third, reduce unique activities or components in products and processes by redesigning the products and processes. Fourth, spread fixed cost activities over new product output. The cellular telephone carriers are introducing measures to reduce their costs by redesigning the product.

The wireless carriers use cellular towers to broadcast their signals. The cellular product design offers signals traveling long distances, primarily for voice and for relatively low data speeds. A cellular tower is expensive but capable of sending a signal for several miles.

This cellular technology worked well until the evolution of the smart phone. The growth of the smart phone has put very high demands on the cellular tower infrastructure because of the heavy data usage it brings to the market. Since 2010, data has taken over the majority of network traffic from voice communications. Now the carriers and, in particular, AT&T with its Apple iPhones, is having difficulty keeping up with the growth in demand.

AT&T today and, likely a few others in the future, has found a potential innovative solution, adding Wi Fi access points. These Wi Fi access points are ideal for heavy data traffic sent at high speeds over relatively short distances. Wi Fi access points transmit signals over a few hundred feet. The Wi Fi access points are smaller, easier and cheaper to install than are cellular towers. This low-cost approach appears to make sense in areas with high density of users. AT&T has placed them in New York’s Time Square and Rockefeller Center, in downtown Charlotte, North Carolina, in neighborhoods surrounding Chicago’s Wrigley Field and in San Francisco’s Embarcadero Center.

But there are some drawbacks to Wi Fi access points. Sometimes, a user has to take several steps to connect to a Wi Fi access point. Signals from the Wi Fi access points may interfere with one another, if signals come from multiple networks. Some smart phones do not have Wi Fi capability. These disadvantages have, so far, held back Verizon Wireless’s adoption of this apparently low-cost approach to providing service.

AT&T is leading this cost-saving innovation experiment. Their network strains force it to be creative and experimental. AT&T saves costs by redesigning the product itself using a less expensive technology with some shortcomings. If the AT&T experiment proves both cost effective and acceptable to cellular customers, every other wireless carrier will be forced to adopt it. And since a Wi Fi access point is largely a fixed cost, the wireless carriers with the highest density of membership within the Wi Fi area will have the lowest cost per unit. In most areas of the country that is likely to be either Verizon or AT&T. They will end up getting a unit cost advantage over their smaller competitors…if this works.

Constrictions in Components Supply Support Higher Prices

Years ago we were doing some work in the roofing business. In one study, we were working on the asphalt shingle roofing manufacturing business. At the time, this was a terrible business. Returns were low, growth rates were modest, at best, and there was a good deal of overcapacity in the industry. Then the industry caught a break. A shortage in asphalt developed. This shortage of asphalt rolled through the asphalt shingle plants and restricted their output. Immediately, prices jumped, returns became attractive and industry participants breathed a sigh of relief. Unfortunately, this asphalt shortage did not last very long. The industry shortly returned to its previous hostile condition. (See the Perspective, “What Ends Hostility?” on StrategyStreet.com.)

A shortage in any component, or labor, will restrict industry capacity and tend to raise prices. A labor shortage is, in part, responsible for some of the high prices in mining today. Miners work in areas that are often hard to reach. They also are skilled employees. The run-up in commodity prices, especially those related to ores such as silver, gold and copper, has increased the demand for these skilled miners. In addition, the mining industry faces competition for skilled workers from the oil and natural gas industries, which are also growing.

Mining companies are now going to great lengths to attract and retain these skilled workers. Some of these miners are now earning 25% more in compensation than they were a year ago. Some companies are flying workers to and from remote mines. For example, BHP Billiton plans to fly 500 workers from Brisbane, about 500 miles away, to a coal mine site that they are opening and then fly them back home after a couple of weeks.

If this commodity boom continues, the industry’s total capacity will be determined more by labor availability than by its more traditional measures of capacity. (See “Audio Tip #117: Capacity Constraints and Pricing” on StrategyStreet.com.)

Costs - The Problem with Weak Constraints

Here are two random observations of the results that any manager can expect to face when there is little to no constraint on the level of costs in an organization.

The first comes from the Heritage Foundation. This foundation analyzed the percentage of jobs gained or lost since January of 2008 through July 2010, a time of recession. The foundation measured job growth in the federal government, state government, local government and the private sector. The private sector was under extreme constraints as revenues flattened or shrank. This sector lost 6.8% of its jobs. Local government was under pressure from the fall-off in property tax receipts. This sector lost a little less than 1% of its jobs. State governments suffered from falling income tax revenues as the recession flattened consumers and commercial tax payers. It lost one tenth of one percent of its jobs. Then there is the federal government, who operated without constraints by creating debt. In just the two and a half year period, total federal government employment increased by 10%. Shocking.

The second observation is also a great source of concern. This data tracks the performance of public schools, K through 12, from 1970 to 2010, forty difficult years. Voters of all kinds have tended to support public education. This support shows up in both spending on the public school sector and in its employment. Since 1970, the real spending, that is after adjusting for inflation, on public K through 12 education has increased by 150%. (See “Audio Tip 195: Economies of Scale and Their Measurement” on StrategyStreet.com.) The total employment has increased by about 100%.

Did we get any more for that additional spending? Enrollment increased by about 5%, after having fallen for the first twenty years of the measures. So, productivity, as measured by employment divided by enrollment, declined a great deal. But perhaps there was more benefit in the quality of the education? It turns out that hasn’t happened either. The scores for science, math and reading have not moved at all, despite the increase in spending.

In both of these examples, we seem to be spending without accountability. (See “Audio Tip 198: Diseconomies of Scale” on StrategyStreet.com.) As much as you can criticize the budgeting system of most businesses, results like these are highly unlikely to occur over a period of time in business systems because there would be quick accountability with this kind of loss in productivity. If that accountability did not come from within the business then, surely, competition would call the profligate business to account.

Previews of Coming Attractions in Public Services

A generation ago, public servants earned less than equivalent employees in the private sector. This is no longer the case. Many reports today suggest that public servants earn 25% or more greater compensation than equivalent private sector employees. While a percentage of the workforce employed in private industry union shops has steadily declined for more than thirty years, unionization in the public sector has grown rapidly. This is important because of the inflexibility of many unions in changing work rules and compensation when confronted with economic realities such as tightening budgets.

What might you expect to happen in such an environment? Growth of private sector companies offering the same services, or better, for less money. These private companies operate under the price umbrellas of the public sector. That is certainly happening today, even in the most unlikely of places. A little company in Maryland has grown into the country’s fifth largest library system, measured by number of branches. This small company, Library Systems and Services, Inc., runs fourteen library service systems operating 63 branches. It has $35 million in annual revenue and 800 employees. It ranks behind Los Angeles County, New York City, Chicago and the city of Los Angeles in the size of its branch system.

The company is finding it relatively easy to succeed by cutting overhead and replacing unionized employees with non-union employees willing to do the jobs for less. In a recent $4 million contract, the company pledged to save $1 million a year using its cost reduction techniques. (See “Audio Tip 187: The Components of Productivity” on StrategyStreet.com.)

Nor does the company need to reduce hours and services in order to succeed. The company has found that the operating policies of public libraries often serve to protect job security and ensure high rates of pay. (See “Audio Tip 182: Productivity as a Measure of Physical Costs” on StrategyStreet.com.) Of course, not all people are happy with the success of this company. In particular, the company’s most recent contract came in for severe criticism from the Service Employees’ International Union. That union has 87 members in libraries recently transferred to Library Systems and Services.

As the cost of public employee pay and pensions becomes less bearable in the future, we can expect to see a good deal more of companies like Library Systems and Services. These private companies should also be good investments. Their first need is not to generate greater revenues, though I am sure they will try that. Instead, they need only reduce costs. That should be relatively easy, due to the price umbrella held up by current public sector management of citizen services.

Service Levels Go Up, Not Down, in Hostility

A market in overcapacity is hostile. Surprisingly, in a hostile market service levels to customers go up, not down. (See “Video #36: Probable Priorities for Innovation in Hostile Markets” on StrategyStreet.com.) The airline industry is an example. The airline industry has been hostile virtually from the day it was deregulated in 1978 until today. During that time, the industry has made great strides in reducing its costs and increasing its service levels at the same time.

Here are some interesting statistics that bear out this contention. These statistics compare the airline industry in 1969 to that of 2009. In 1969, 172 million passengers flew U.S. Airlines. By 2009, that number had grown to 770 million. In 1969, there were 5.4 million flights. By 2009, the flight numbers had risen to 10.1 million. Service levels, as measured by number of flights and number of passengers, clearly have risen over the last forty years. During that time, safety clearly improved. Fatal accidents per 100,000 departures were 1.3 in 1969 and .1 by 2009. Pricing dropped as well, because costs dropped. In 2009, it cost a passenger 14 cents to fly one mile. The comparable number in 1969, using 2009 dollars, was 34 cents. Today you can get to more places faster by airliner than you could in 1969. Service levels have risen.

Naturally, those of us who fly would complain that service levels in terms of comfort have fallen drastically. Meals used to be free and there used to be ample space for knees and luggage. Those days seem to have passed...or have they?

The airlines have learned time and again that customers will not pay for onboard meals and more leg room. However, those customers who are willing to pay for more comfort can fly in economy plus or business class or first class. The prices for these services today are much lower than they were several years ago. So no matter how you slice it, service levels have risen in the industry when you look at the service levels for which customers are willing to pay.

The same holds true in every hostile industry. (See “Video #37: Performance Innovation Tradeoffs in Hostility” on StrategyStreet.com.)

Dis-Economies of Scale

McKinsey research has found that only 10% of cost reduction programs sustain their results after three years. The problem seems to be overhead. Companies have exploited manufacturing efficiencies to reduce the cost of goods sold as a percentage of revenues by nearly 3% over the past decade. On the other hand, sales, general and administrative costs have remained about the same. The performance of sales, general and administrative SG&A costs is an example of dis-economies of scale.

A few definitions are in order. Economies of scale is the phenomenon where unit costs decline as the number of units sold increases. (See “Audio Tip #195: Economies of Scale and Their Measurement” on StrategyStreet.com) This happens because part of the cost structure is fixed, so it grows at a fraction of the rate of growth of the unit volume the company sells. Dis-economies of scale occur where units of costs increase at a rate that is greater than the increase in the units of output. (See “Audio Tip #198: Dis-Economies of Scale” on StrategyStreet.com) In other words, there appear to be no significant fixed costs in the company’s cost structure when dis-economies of scale occur. At the other end of the cost spectrum, you see super-economies of scale. Super-economies occur with costs decline, even as the number of units sold increases. (See “Audio Tip #199: Super-Economies of Scale” on StrategyStreet.com) Usually the super-economies occur due to changes in technology, when the company replaces many employees with technology capital investments.

Now let’s return to the problem of SG&A costs. In 1998, the SG&A costs for the S&P 500 companies were roughly 22% of sales. By 2008, SG&A costs still commanded 22% of the sales dollar. During those ten years, the physical units of output the S&P 500 companies produced certainly increased. On the other hand, the SG&A costs remained the same as a percentage of sales. Are none of the SG&A costs fixed? If there were some fixed costs in SG&A, the companies should have been able to increase the units sold without a proportional increase in numbers of people in the SG&A functions, creating economies of sale. But they did not produce economies of scale as measured as a percentage of revenue. What happened then? Either the number of people employed in the SG&A functions grew with unit sales or the companies paid the average SG&A employee at a higher rate than sales grew. In either case, you have dis-economies of scale operating in SG&A.

Over the years, we have been involved in many cost reduction efforts. We have seen that it is hard to sustain the results of a cost reduction effort over a long period of time. The McKinsey study serves as ample testimony to this fact. What may be helpful is to tie physical units of costs, for example numbers of full time equivalent employees, to physical measures of output, such as customer orders. If the ratio of physical units of cost to physical units of output goes down, the company has an excellent chance of creating economies of scale.

Here We Go Again

The leader of the United Auto Workers is retiring. He is leaving a union under siege. By 2009, UAW membership was about half of the level of 1995. The union has hemorrhaged members as the big three domestic automobile producers have shrunk in market share, lost billions of dollars, and closed plants.

The departing leader of the UAW claims that the industry’s difficulties never rested with the union and its rich contracts. In his view, the crisis that led to the bankruptcies of GM and Chrysler and the near bankruptcy of Ford was strictly the result of an unexpected spike in gas prices and a recession that resulted from the mortgage crisis. He believes that the fault lay not with the union and not with the industry. Following this belief, he is encouraging his successor to begin clawing back the cost-cutting concessions that the union has granted the Detroit big three domestic automobile manufacturers now that these companies are moving toward profitable operations.

The problem is that these concessions did not do enough, at least from the results they seem to have produced. The concessions really got underway in 2003, as the union reduced its wages and benefits and transferred retiree healthcare costs from the automakers to an independent trust. Despite these concessions, union membership fell parabolically from 2003 to 2009, right along with the profits in the big three. In the meantime, German and Asian manufacturers continued to be profitable. These profits included profits in U.S. domestic manufacturing facilities as well. (See the Symptom & Implication, “Some industry leaders have lower returns than the smaller competitors” on StrategyStreet.com.)

The union is heading back to trouble and will take its unionized facilities with them. In an earlier blog (See Blog HERE), we described the hourly cost differences in wage rates between a unionized and non-unionized domestic facility. These cost differences are unsustainable in the longer term. No one can expect that an automobile plant with $73 dollar an hour labor will be profitable enough to compete with another domestic plant producing similar automobiles at $48 an hour. Despite recent troubles, the Asian manufacturers still command a premium price over their big three competitors for their products. So, Toyota and Honda get a higher price and produce with a lower costs. (See “Video #1: The Two Best Consultants in the World” on StrategyStreet.com.) Tell me how GM, Chrysler and Ford can produce an equivalent or better car with these economic conditions. The claw-backs will only make things worse.

The Tables Have Turned

Just a few years ago, Dell Computer was the darling of the PC industry. Hewlett-Packard was an also-ran. Today the tables have turned. Over the last year, Hewlett-Packard’s market share has jumped from 18.5% to 20% of the global PC shipments. Dell’s market share has fallen from 15.7% to 13.7%. The change in market share is customers saying that HP offers a better value proposition. (See the Perspective, “The Two Best Consultants in the World” on StrategyStreet.com.)

HP has strenuously reduced its costs at the same time it has gained market share. HP, at one time, was in the middle of the PC pack in costs. Today, its PC operating margin is 4.6%, better than Dell’s operating margin. The superior cost performance of HP is the competitors acknowledging that HP has a lower overall cost.

HP has gained much of its market in the retail side of the business, especially with lower-end products. Most Standard Leaders don’t like lower-end products because they offer low margins and not much reputation (see StrategyStreet/Diagnose/Products and Services/Innovation for Customer Cost Reduction/Price Point Bias).

HP is gaining some of its new market share with very low pricing, a characteristic that used to belong to Dell. Wal-Mart’s recent experience with HP is indicative of the company’s current aggressiveness. Wal-Mart wanted to sell a personal computer for less than the price of the hot Netbook products. Netbooks are mini laptops costing less than $500. Wal-Mart did the consumer research to produce specifications for the proposed full-sized laptop product and passed those to some PC companies. Several responded positively. Dell offered a product for just under $400. Toshiba and Acer offered a product priced just below $350. HP beat all the competitors, though, with a $298 machine.

Dell is allowing Hewlett-Packard to take market share with its low prices because it sees little profit in the low-priced machines. But there is a problem here. HP has already proven itself capable of using its large size to streamline logistics and extract lower purchases of costs from its suppliers. It has smartly redesigned its products to reduce their costs. How is it, then, that Dell, or anyone else, believes that they will be better off by ceding market share to Hewlett-Packard on the basis of low prices? Will Hewlett-Packard become weaker? Will the PC companies losing share become stronger? This is another example of companies in a Leader’s Trap.

You Mean I Have to Pay for This?

Those of us who fly a lot have noticed how few people have a meal on an airplane anymore. In flight food was attractive when we got it for free, much less so when we have to pay for it.

The WiFi industry is learning a similar lesson. A couple of WiFi suppliers to the airline industry are trying to figure out how to charge for their services. (The WiFi suppliers control the pricing so that the airlines can not give it away, as they have tended to do in the past with other benefits…though that trend certainly has ended.) The WiFi suppliers have found that when the service is free, many customers use it. But when they charge for the service, even at the low price of $1.00, usage drops drastically off.

The companies are trying price schemes that are tied to the length of the trip. One plan charges $12.95 for the service when the flight lasts longer than three hours and $9.95 for flights from ninety minutes to three hours. If the flight is shorter than ninety minutes, the price is $5.95.

The WiFi sellers need about 10% of travelers to pay for internet access in order for the service to be profitable. That will be difficult. A few flights have seen usage in the 12% to 15% range, but they tend to be on longer flights. A large percentage of U.S. domestic flights last less than two hours.

The WiFi suppliers, though, are coming up with a different approach to pricing that is much more likely to succeed. One approach will be to sell packages of service. A business traveler might buy a package of five flights for a fixed price. Once the price has been paid and is out of the customer’s mind, it is more likely that the service will appear attractive while the traveler is on the airplane. In addition, it should be much cheaper to sell a package of five, ten or fifteen flights than it is to sell one unit of service on each flight. In another approach, the WiFi companies are planning to negotiate directly with companies to sell their services in bulk to the companies for use by their employees. This approach has even better cost savings than the package sales. It is a much more efficient way to go. Both of these approaches change the price by altering the basis of the charge for a unit of sale.

Visit www.StrategyStreet.com

Big Cost Differences in an Industry - Part 2

McKinsey and Company has undertaken a detailed examination of Productivity in the pharmaceutical industry. This extensive study offers us an opportunity to see common patterns in cost management and productivity improvement. We described these patterns in Part 1 of this blog. These common patterns of Productivity improvement include four major concepts.

1. Reduce the rate of cost for the Input

2. Reduce the Inputs not producing Output

3. Reduce unique Intermediate Cost Drivers(ICDs) in products and processes

4. Spread fixed cost ICDs over new Output

McKinsey undertook a detailed analysis of 1900 pharmaceutical production lines at 150 plants located all around the world. The Firm measured the Productivity levels among the plants and companies in their study and found dramatic differences in Productivity among these 150 plants. (See the Perspective, “What Makes Returns High?” on StrategyStreet.com.) The top quartile of the companies involved in this study had manufacturing utilization rates that were more than twice those of the bottom quartile companies. McKinsey estimated that, if the average drug-maker would match the total labor productivity of the top players in the industry, it would realize an improvement of 5 to 6 percentage points in operating earnings, quite an improvement

Here are some of the cost management reasons for the superior performance of the best performing companies and how these cost management efforts fall into the four categories of cost reduction above:

* Measure of I/ICD and Concept #2: McKinsey found that the top performers used non-production labor extraordinarily efficiently. For example, the quality control employees for the top performers reviewed an average of 110 batches a year, while those in the bottom quartile did less than 5. This finding is a good example of both an Input, the quality control employee, and an Intermediate Cost Driver, a quality control batch. A company using this measure would reduce Inputs not producing Output through reporting to the employees of this measure of their efficiency. (Concept 2)

* Concepts #2 and #3: The high performers use standardized ways of measuring and controlling equipment, reducing line stoppages and waste. This is an example of two of the patterns. (Concept 2) This approach reduces Inputs not producing Output by eliminating unplanned downtime. It also reduces the unique ICDs in processes by standardizing processes. (Concept 3)

* Concepts #2 and #3: The top performers were more likely to use lean management tools to plan and schedule activities, so they released a higher percentage of their products to market without reworking. This approach reduced Inputs not producing Output by improving the accuracy of production forecasts. (Concept 2) It also reduced unique ICDs in the process by reducing the rework activities through a reduction in errors. (Concept 3)

* Concept #2: The top quartile players reached final delivery in half as much time as the average manufacturer, and more than five times faster than those in the bottom quartile. The top performers reduced Inputs not producing Output by speeding the process.

* Concept #3: The best performers eliminated unnecessary complexity from their production planning activities by using fixed, repeatable, short duration production schedules in order to avoid forced changes in production plans. These companies reduced unique ICDs in their processes by reducing the movements of Inputs.

* Concept #4: McKinsey found that small plants were substantially less productive than larger plants. However, the very largest plants were not the most productive. This illustrates the spreading of fixed cost ICDs over additional product Output. It also warns us of the limit of that concept when there are multiple products emerging from the plant.

Every drug maker that McKinsey studied had launched a lean, or Six Sigma, project in the recent past. Yet relatively few of these companies were effective in reducing their comparative costs.

For further explanation of these cost reduction patterns, and for over 600 cost reduction concepts, illustrated by 2400 examples of these concepts in action, please visit www.strategystreet.com/improve/costs. These concepts will help you improve your company’s productivity.

Big Cost Differences in an Industry - Part 1

The objective of every cost management system is to improve the Productivity of the Input (I) resources used to produce the final Output (O) product. These resources include the Inputs of People, Purchases and Capital. The Output is the unit of product sold. A simple measure of Productivity is Inputs divided by Outputs (P=I/O).

There are several stages in producing a unit of Output, so relatively few employees (one of the key Inputs) actually produce a unit of Output. Instead, most employees produce something else as an intermediate end product on the way to the final product. We call these intermediate end products Intermediate Cost Drivers. So, to increase Productivity, you would like to improve the ratio of Inputs needed to produce Intermediate Cost Drivers (I/ICD).

You would also like to reduce the activities, or Intermediate Cost Drivers, you require to produce a unit of final product Output. In ratio form, this means you want to reduce the ratio of Intermediate Cost Drivers in the Output (ICDs/O).

The measure of Productivity then expands into two factors:

Productivity = Inputs/Output = Inputs/Intermediate Cost Drivers x Intermediate Cost Driver

Or

P = I/ICD x ICD/O = I/O

We have studied several thousand cost reduction efforts from the last 25 years. We believe that you can categorize all cost reduction efforts into one of four major concepts:

1. Reduce the rate of cost for the Input

2. Reduce the Inputs not producing Output

3. Reduce unique Intermediate Cost Drivers(ICDs) in products and processes

4. Spread fixed cost ICDs over new Output

The current issue of The McKinsey Quarterly describes an interesting study that McKinsey has undertaken in the pharmaceutical industry. The study’s conclusions offer us the opportunity to categorize their findings into one of these four cost reduction concepts.

In Part 2, we will describe some of McKinsey’s findings and then tie the findings to the four major productivity improvement concepts above.