It works, but does it taste good?

Like parents wishing only the best for their children, you had high hopes for your latest prototype. You nursed that baby for weeks – maybe months. From the day the new product idea was conceived, until its birth on the bench top, you made sure it had all the right ingredients, in the proper quantities, at the right temperatures, for the required amounts of time. That prototype – your pride and joy – had a brilliant future. And then it came time for scale-up.

Sure, you understood that your little tyke might undergo some changes at this point. But according to your calculations, those alterations should have been minimal – a little darkening, crisping, sweetening – no big deal. Marketing would get their product when they wanted it, accounting would be happy with the numbers, the engineers would marvel at your ability to predict process interactions and you'd be free to move on to your next innovation. But alas, such was not to be. Like a teenager plagued by peer pressure and pimples, your angelic bench sample came back from the pilot plant bad.

So now what? Beat your breast and curse the Fates? Unload on Dear Abby? Fine options, both are. But there's also a third: Retrace your steps, determine the error in your ways and try again -- all the while consoling yourself with the fact that you're not alone.

"It's a huge problem. We see it all the time," says Gail Vance Civille, president of Sensory Spectrum in Chatham, NJ. "Marketing puts aggressive time lines around scale-up, and R&D is left to react to those demands. Ironically, instead of cutting scale-up time, that pressure instead makes the process longer because more mistakes happen."

Poor research and development. Once again it's the victim of circumstance. But it doesn't have to be that way.

"No one ever sits down and makes a conscious decision to fail. But if they fail to communicate, that's when problems occur," says Nancy Rodriguez, president of Food Marketing Support in Oak Park, IL. "A team approach from square one could eliminate a lot of wasted time as a product moves through the various stages of development -- including scale-up. As the design is developed by R&D, designers should give input as needed, but also receive input from everyone."

Rodriguez explains that taking a team approach lets members of all departments voice their concerns up front; that way, there should be no surprises as development progresses. If purchasing foresees a problem with seasonal ingredient supplies, those concerns can be put out on the table and last-minute ingredient substitutions can be avoided. If manufacturing and packaging lack required capabilities, they have some time to work up appropriate suggestions and/or solutions.

"Yes, it takes time to gather a team," Rodriguez says, "but the payback is a shortened launch time because the scale-up went smoothly."

Developing a product profile

Amassing a team is one essential step; another is developing a comprehensive profile of the product prototype that will serve as a reference throughout scale-up.

Don Goodman, president of M&D Technologies in Columbus, MS, recommends using as many descriptive techniques as possible to ensure an accurate profile. In addition to writing down its formula and processes, document the food's physical attributes – moisture levels, nutrient content, etc. – everything that can be measured analytically. Have sensory use descriptive tests to work up statistics on appearance, taste and texture. Quantitative descriptive analysis and flavor profile analysis panels provide sensory data in graph form; these diagrams can be referred to as needed.

If a picture truly is worth a thousand words, one might be tempted to believe a few snapshots of a prototype would facilitate this documentation process. But Goodman cautions against heavy reliance on photographs; though color pictures can indeed be helpful, they also can be misleading if colors alter during the film's processing.

With baseline parameters thus established, scale-up can proceed. But, Goodman adds, don't abandon your prototype. "Keep making it on a small scale, mixing and forming by hand and cooking on your Jenn-Aire," he says, "That way, you'll have fresh samples to compare against."

The paper trail continues in the pilot plant, where Goodman urges more copious record keeping. "Do process audits with every run," he says. "Document everything – ingredients added in this order, at this speed, cooked for this long at this temperature – because without this data, you're totally lost. With it, you can draw some correlations between how some slight change in process caused some rather dramatic changes in the product.

"You have to expect some changes when you go from mixing product by hand to processing 290 pounds per minute. But by having reference data and product to compare with, you can pinpoint what those changes are, as well as what caused them."

Once changes have been enumerated, it's time again to enlist the services of sensory; assorted tests can be administered to determine whether those differences are detectable, and/or significant.

"In an ideal world, pilot-plant runs reproduce the prototype exactly, and you needn't worry that there might be differences," says Ellen Daw, sensory analyst for J.M. Smucker Co., Orrville, OH. "But that just doesn't happen very often in the real world. That's why it's so important to establish the sensory attributes before scale-up – so you have information to refer to when changes start cropping up."

Daw describes two means of tracking changes in pilot-plant products. "One way is to test the pilot product by itself," she says. "If it doesn't score at least a seven on a nine-point hedonic scale, don't go with it – there are problems that need to be worked out. But if the hedonic scores are OK, go ahead and do a two-sample test, bench top vs. pilot plant product."

Daw adds that using freshly prepared bench product is crucial to obtaining accurate results. "By this point, you're probably at least six weeks into the project," she says. "Even frozen prototypes will begin to undergo flavor, texture and color changes. By working with fresh product, you're limiting the variables to the scale-up issue."

Although these discriminative tests will show if a difference does exist, the results won't tell why the change occurred, or whether the altered product is still considered acceptable. Check your process audits for clues as to why the changes occurred; as for if the product is still satisfactory, consumer testing is needed.

Sensory Spectrum's Civille says that any time any sort of change is made during scale-up, it is important to once again recheck all product attributes – physical, chemical and sensory.

"You'll probably encounter resistance to so much testing, but it really needs to be done to ensure that the end product replicates the prototype as closely as possible," she says. "If you find there's any significant variance, use the 'Don't pass go' method: Don't proceed to the next stage until you've straightened out the existing problems."

Despite all the testing and retesting, products do eventually reach the manufacturing plant. So what happens then? Should product designers abruptly wash their hands of the product and let QA/QC take over?

"Technically," says Civille, "full production runs do fall under the auspice of QA/QC. But it's my personal feeling that, for the first year, R&D should continue to keep an eye on things."

She notes that the first few plant runs are usually under the best possible circumstances. It's a new product, and everyone's babying it. But later runs, say a month or three months later, also should be spot-checked by R&D to make sure standards are being upheld.

Civille adds that, often times, manufacturing's standards for quality are altered, depending on prevailing conditions. "There's so much going on in a plant on any given day; it can be tough to use the same rigid standards all the time," she says. "One of my favorite QA/QC sayings is, 'Well, it's not bad, considering the fire we had in the plant yesterday.' "

Pam Erickson is a St.Charles, IL-based contributing writer specializing in topics relating to the food industry.