Today, as in times past, the food business relies on analytical testing. From quality control monitoring of key chemical components impacting food flavor, texture and functionality; to performing analyses to satisfy regulatory requirements; to supporting R&D product development efforts; analytical testing performs a crucial function. That's not news. What is new is the question many food-industry firms currently ponder: "Should analytical testing be outsourced?"
In today's business environment, it's no simple task for food companies to make the impressive profits needed to turn Wall Street investors' heads. Operating in an intensely competitive, overly regulated and mature industry, food companies have resorted to downsizing and outsourcing essential services as ways to improve profits (at least for the short-term).
Ironically, at a time when more real, honest-to-goodness, new product innovation is needed, food companies are cutting big in areas in which they most desperately need to be more productive -- R&D and peripheral support groups, such as analytical services.
Quaker Oats Company and Keebler Company, two Chicago-area food businesses that have reduced their in-house analytical staffs in recent months, have increasingly outsourced analytical work while downsizing and restructuring.
Outsourcing routine analytical demands has worked well for Kellogg Company, Battle Creek, MI. The nation's leading cereal maker recently decided to outsource even more analytical work in addition to its nutritional-testing efforts, which have been outsourced for some time.
Expensive maintenance
Maintaining state-of-the-art analytical food chemistry labs is, indeed, expensive. The list of continually escalating costs includes: lab technician and chemist salaries and fringe benefits, as well as chemical reagents, glassware, instrumentation, instrumentation repair and maintenance, and chemical-waste disposal. Construction costs -- including lab furniture, fume hoods and basic start-up -- also are significant. But contract labs also must deal with these same cost issues, and guess who will be paying their bills? So no matter who does the work -- in-house or outsourced -- analytical testing doesn't come cheap.
"Sometimes, food companies have a hot, but temporary, need for a particular analytical test, and instrumentation is purchased, heavily used for a short time, but then sits around unused for months on end," says Wayne Ellefson, senior account executive at Madison, Wisc.-based Covance, Inc. (formerly Corning Hazleton Labs). "Outsourcing analytical work when this situation occurs is a good way for food companies to save money. Since contract labs tend to operate instrumentation day in and day out, they have efficiencies of scale going for them that food manufacturers might not have. Operating efficiencies can lead to considerable cost savings."
Of course, judicious resource management by in-house lab managers can produce similar cost savings. So, the more efficiently an in-house analytical lab is run, the less likely are the chances a food company will save much money through outsourcing.
As in any business, efficiently managing labor and equipment is critical to profitability. Autosamplers for gas chromatographs (GCs) and high performance liquid chromatographs (HPLCs) can operate unattended overnight to reduce labor costs. HPLC solvent recyclers can save on solvent and waste-disposal costs. New instrument techniques such as supercritical fluid extraction and accelerated solvent extraction might reduce labor and chemical reagent costs, depending on the particular application.
Rather than outsourcing all analytical services, a more prudent approach is: first maximize the in-house analytical lab's operating efficiencies, then outsource some work that can't be cost-effectively performed in-house.
Quantity vs. quality
When evaluating the cost-effectiveness of in-house analytical testing, quality of results - such as result accuracy and business decisions made based on those results -- is even more important than quantity of analytical data generated. A food quality control lab might be analyzing dozens of sour-cream samples daily for fat content to ensure maintenance of production specifications. Because the processing techniques are honed to ensure accuracy, chances are few that out-of-spec samples will be discovered. On the other hand, a research analytical lab might crank up its GC/MS system only once a month, run a few samples, and resolve a critical off-flavor problem in a new product line, leading to changes in product formulation, processing parameters, or packaging alterations that prevent future off-flavor problems.
If in-house analytical labs generate data of little use, then the question isn't whether to perform the analysis in-house or outsource it; a more appropriate query is whether to perform it at all. In-house labs probably are generating analytical data used for various purposes. Examples include: nutritional testing to satisfy the Nutritional Labeling and Education Act of 1990; analytical studies intrinsic to new product development efforts; examination of competitor product formulations (to determine how competitors are undercutting pricing or getting a particular end result); resolution of quality control complaint samples; and pesticide testing. Which test types are most appropriate to conduct in-house and which should be outsourced?
Reputation to protect
Are outside lab results less accurate than those produced in-house? To stay in business, contract labs must turn out reasonably accurate results. If they lose their reputation for accuracy, they won't remain up-and-running. Occasionally, however, pressure to crank out volumes of test results to generate revenues can lead to sloppy work and worse. In November 1991, one contract lab was slapped with state criminal and civil charges for allegedly falsifying test reports for antibiotic tests run on milk samples at six Wisconsin cheese factories. The lab pleaded no contest to six counts of filing a false report with intent to mislead. Such cases are rare. Most contract analytical lab errors are probably caused by lack of experience testing the client's specific type of food product. Nonetheless, the 1991 incident emphasizes the need to carefully scrutinize a contract lab's reputation before hiring.
"We outsource our analytical overload," says an analytical services manager for one major Chicago food marketer. "We outsource the common tests we know for sure contract labs can do.
"But, there are several issues involved that you need to carefully consider before outsourcing," warns the lab manager. "Straightforward cost accounting is easy to do and looks great on paper. But there are many intangibles to consider."
It's important to recognize the distinction between a contract lab operating as a business-profit center and an in-house analytical lab serving a support function to R&D and quality control.
"To a contract lab, the analytical data it generates are just numbers," says the lab manager. "But an analytical result is more important than a mere number to us. We use data to make important business decisions, and we take the time to do the tests right. This may be one reason why, in our experience, in-house lab results are usually more accurate."
But another reason exists as to why in-house lab results might lean toward accuracy more than contract lab results: In-house labs usually have considerably more experience with their own particular product types and are aware of matrix errors that might arise during testing. Since they often possess years of experience running a particular sample type, in-house chemists can spot a bad answer more easily than can contract-lab chemists. They also know which corrections and modifications to make in test procedures.
"We have 16 different test parameters for moisture determination, depending on the type of product being tested," the lab manager says. "For example, for some types of products, you can simply heat the sample in an oven for several hours and determine the moisture content as weight loss. But for other types of samples that are more sensitive to thermal degradation, you need to heat samples at lower temperatures, removing water vapor with a vacuum. We don't really want to share our test methods -- methods that have taken us years to perfect -- with an outside lab that can use them to test competitor products."
But, for certain types of testing, contract labs might have the experiential upper hand. Julie Hill, chemistry division laboratory manager at The National Food Laboratory (NFL), a division of the National Food Processors Association, says she sees more and more outsourcing for pesticide analyses. "That's one area of testing that requires significant expertise and capital investment," Hill says. "It's difficult for a food company to justify the personnel and instrumentation expenses for pesticide testing, unless they're running several samples daily. And most food companies simply don't have that type of pesticide workload."
Hill sees frequent workload cycles even with pesticide testing. Several years ago, one NFL client performed pesticide testing in-house and then decided to contract it out to save money. After a few years, a pesticide crisis occurred in which some of the food company's products were contaminated with an illegal pesticide. The firm had so many samples to test, company executives decided to bring it back in-house. Now the crisis has passed, and once again, the company is negotiating with NFL to outsource the testing.
The big interest in nutrition testing, Hill says, is the change from gravimetric solvent extraction methods for fat to more accurate fat determinations by saponifying the fat, methylating resulting fatty acids, and then analyzing the methyl esters by gas chromatography. Another area she says needs more work is vitamin A testing: "Everyone gets a different result with vitamin A analysis."
Nutrition-testing volume contracted out has shrunk considerably in recent months, as more big food companies perform it in-house. Last October, NFL stopped offering nutritional testing services. Checking food for heavy-metals contamination is currently a busy area for NFL. The group's chemists use graphite furnace atomic absorption spectrometry to achieve the low detection limits required.
Besides pesticide testing, the NFL has developed considerable expertise in resolving off-flavors in food and packaging systems. "We have over 25 years experience with many different types of off-flavors and product types," Hill says. "Success with this type of work greatly depends on the experience of the chemists."
Losing control
Outsourcing holds definite drawbacks, including potential loss of control, confidentiality and quick turnaround times. When a firm submits samples to a contract lab, it loses a certain level of control over when the sample gets done, who performs the work, and how testing is performed. Contract labs schedule samples into their workloads. In effect, the contract lab assigns a priority to the work it receives. Even though the "when," "who" and "how" might be important to the firm submitting samples, those issues might not be the lab's No. 1 priority. Most contract labs offer a "rush" service so samples are returned more quickly than normal, but hefty surcharges are assessed for the privilege -- typically twice the normal cost of testing.
One of the biggest problems with contract labs is turnaround time. Delays are common. In recent months, however, several new analytical contract labs have opened their doors for business, and the increase in the number of labs has helped reduce sample backlogs and overloads.
Most contract labs won't discuss a firm's work with others. Nonetheless, the occasion might arise when a firm has certain sensitive samples it wants tested, and the fact is it simply doesn't have a 100% guarantee that the type of testing conducted and the analytical data generated won't be discussed with potential competitors. An organization might have developed a particularly tricky analytical test for a critical analyte. If this information is passed along to the contract lab, it might use the test methodology to analyze competitor products needing the same type of testing, but lacking appropriate test methodologies. In other words, any knowledge passed along to a contract lab can end up helping competitors.
Relying too heavily on technical expertise from outside the company is a mistake. If downsizing cuts are made too deeply and indiscriminately within analytical services personnel and other R&D areas, the talent, expertise and experience lost can be devastating and might prove more difficult to replenish than expected.
Unfortunately, many chemists, food technologists and other technical people aren't all that strong at self-promotion. As a result, management can easily overlook their technical and esoteric contributions. But once cuts are made, it isn't long before the loss of these vital contributions becomes apparent.
"If there's one thing I've learned in the past couple of years, it's that analytical lab managers need to do a better job articulating the benefits they provide to the company," says one analytical services manager who experienced significant staff cuts in a recent corporate downsizing.
Send it out
The best approach for handling analytical testing requirements is combining in-house testing and outsourcing. Organizations should outsource the complicated analytical problems they're not set up to handle -- either because they lack proper instrumentation or properly trained personnel, or both. The key is finding a reputable contract lab or consultant that routinely handles the type of testing needed and has the proper equipment and skilled people already in place. "Don't send contract labs extremely tough problems that they're not equipped to handle," Ellefson says. "If they can't do the work efficiently, or if they'll need to conduct significant analytical research work, or develop a new analytical procedure, it will be an expensive proposition. After all, contract labs have to pay their expenses, too."
Several excellent contract labs have developed specialized testing capabilities. For example, certain laboratories possess excellent pesticide residue-monitoring capabilities. Others can help solve a challenging off-flavor problem. If a quality control department has a customer complaint sample that is contaminated with unknown metallic particulates, it can send the sample to a contract lab. The lab can examine the sample with an electron microprobe equipped with an energy-dispersive X-ray spectroscopy system to determine the metal's composition and perhaps its origin -- all for a reasonable fee.
Get what you pay for
Another important point: Consider more than price when evaluating contract labs. A lab might not be noted for being the low-cost provider of analytical services, but it might excel in ensuring accurate answers. One lab that isn't noted for being low-cost, for example, will run a control sample with every set of routine samples submitted for analysis. If inaccurate results are obtained with control samples, test parameters and instrumentation calibration are carefully checked. If an inaccurate result is reported to a client, the lab will re-analyze samples at no charge. One way to check the accuracy of a contract lab is by sending them control samples (samples of known analyte concentration) along with unknown samples. Companies might even tell labs they're being sent a blind control sample in with the samples to be tested. Just knowing they're being monitored might be enough incentive to inspire a lab's most meticulous work effort.
Many contract labs are, or were, directly associated with a major food company. Examples include: R-Tech (Land O'Lakes), Medallion Laboratories (General Mills), and TPC Labs (Pillsbury). Once part of Sara Lee Corporation, Northland Laboratories (formerly Sci-Tek Laboratories) has recently become an independent laboratory, and Covance was once a Ralston Purina Company division.
Many food companies have attempted to pay for some of the costs associated with in-house analytical labs by taking on analytical work-for-hire from other food companies. This affords the dual advantage of having an on-site lab available for high-priority analytical problems demanding immediate attention, while also generating revenues to help defray the costs of maintaining and staffing the lab. Instead of looking at whether analytical work should be outsourced, performed in-house, or via combining the two, these companies have decided to expand their labs so they can become profit centers, or at least, revenue generators.
Pillsbury's analytical-testing approach might be the wave of the future for big food companies. The firm's well-equipped, sophisticated analytical lab operates in a highly efficient manner. It outsources analyses that can't be performed efficiently in-house (or that can be performed at less expense at another contract lab); it fulfills most of Pillsbury's analytical in-house testing demands; and it accepts analytical work from outside clients (as a business-operating unit known as TPC Labs). "This allows us to focus our efforts on what we do best," says Paul Thompson, director of TPC Labs.
TPC has experience with doughs, vegetable products, meats, packaging materials, flavors, and identification of foreign materials in foods using various microscopy techniques. "We are an analytical problem-solving lab and don't take on much high-volume, low-cost work like nutritional analyses," Thompson says. Approximately 10 percent of Pillsbury's analytical workload is contract testing for other food companies.
One disadvantage for clients using a contract lab associated with another food company is that the contract lab's in-house requests tend to receive higher priority than outside clients' requests.
Should food companies outsource all of their analytical testing needs? Small food companies might have to send out most of their analytical work. But Thompson makes this point: "Smart food companies will always have an in-house analytical lab to handle important problems that require results in minutes or hours."
Do it on-site
Concerned about costs associated with assembling on-site labs, some smaller operations have traditionally shied away from in-house microbial testing. Expenses linked with testing equipment and additional time and labor required seemed to be insurmountable barriers. As a result, many small processors have relied on outside laboratories. With the advent of simpler, faster and more precise methods, on-site microbial testing of food products and contact surfaces has become a cost-effective and time-saving option for many processors. One small food processor in Alabama outsourced its microbial testing until 1996, when changes in customer needs prompted the quality control department to evaluate other options.
"We picked up a number of major customers who wanted us to do more microbial testing," said the quality control supervisor for this 250-employee operation. "But performing more tests would have significantly increased our testing costs at the outside lab."
To minimize additional costs and still address customers' needs, the company began investigating the possibility of establishing an on-site laboratory. As part of that assessment, they decided to try 3M Petrifilm plates. These plates are coated with rehydratable nutrients and a gelling agent that activate when the sample is added. The sample-ready media eliminates the time-consuming media preparation steps required with traditional agar pour plates, which can prove intimidating for small operations that have little microbial testing experience.
The company performed a month-long trial run to ensure a smooth transition from outsourcing to on-site testing. They tested 10 to 15 samples daily with the new method and sent similar samples to the outside lab. "The results we saw with Petrifilm plates were like those we received from the outside lab," the quality control supervisor says. "After 30 days, we knew that our new on-site testing program would be well-accepted within the company and with our customers."
The new microbial testing laboratory didn't require additional space; the entire lab fits right into a corner of the quality control supervisor's office. On average, food processors pay anywhere from $6 to $15 for each test they send to an outside testing facility. By comparison, testing in-house with Petrifilm plates can cost as little as $1 per test, plus labor. While labor costs are unavoidable regardless of which microbial testing method is used, this simple three-step process -- inoculate, incubate and read -- requires little training and reduces the chance of human error.
Obtaining quicker testing results has been of the biggest advantages of this processor's on-site testing program. Faster results mean more time to identify and isolate any potential contamination problems -- a benefit that can save processors time and money.
"Within 48 hours, we have confirmed total aerobic counts," the quality control supervisor says. "We feel safe releasing product within three days vs. waiting five to seven days for the outside lab's results to come in. Petrifilm plates give us a good means of monitoring our products and sanitation, with much speedier results than we were getting from the outside lab."
SupplySide
