The Demise of the Service Bureau?

Published in the September 1998 issue Prototyping Technology International, UK & International Press Copyright 1998
by Terry T. Wohlers

Service Bureaux (SBs) have played an important role in the development of the rapid prototyping (RP) industry. Many of the early adopters of RP systems were SBs. From the start, SBs have served as the ‘training wheels’ for many companies new to the technology. Even today, few companies buy an RP system without first having at least a couple of RP models built at an SB.

Over the past 18 months, many SBs have struggled to maintain profitability, due in large part to competition and a subsequent drop in prices. A few have gone out of business. Will further declines in the prices of RP systems and models signal the collapse of these important, pioneering companies?

About 10 years ago, an estimated 2,000 plotting and scanning service bureaux were in operation, according to Eugene Kleca of Image Peak Systems, a veteran in document imaging and reprographics. Today, this number has shrunk to a few hundred, he said. The primary reason is that many companies now have the capability in house. The cost of large-format plotters and scanners has dropped to a point where it makes sense to own them. Also, the volume of work has increased as companies have expanded their use of CAD systems on a larger number of new product designs and product variations.

Indeed, the average selling price of an RP system has dropped and the RP class of technology, as a whole, has become easier to use and maintain. Consequently, companies that once used SBs exclusively are buying RP systems. This trend is likely to continue, meaning that history may repeat itself. No one knows for sure to what extent or at what pace it might occur, but a reduction in the total number of SBs seems likely.

In 1997, the percentage of RP systems at SBs dropped to 19.7 per cent, down from 23 per cent in 1996. For nearly every year since the beginning of RP, the SB piece of the pie has shrunk due to end-user companies buying systems. These figures alone are a clear indication that companies are relying less and less on SBs. Further price reductions in RP systems could accelerate this decline.

Over the past two years, the average price of RP models has declined by as much as 50 per cent at most SBs in the USA. I recall that in the early 1990s, a small RP model would cost at least US$1,500. Today, similar models and prototype parts can be purchased for as little as US$150-200.

This dramatic price decline has created another dynamic in the marketplace. Many companies that were on the verge of buying a system are now reconsidering, because it is so inexpensive to buy models from SBs. Some companies prefer to let someone else own the problems associated with running and maintaining the equipment, and cleaning and finishing the models. This has slowed the sale of new RP systems. This situation will continue until system prices drop further, making it easier to justify the cost of ownership.

A mystery to some is the fact that the total number of SB locations has grown over the past couple of years. CAD/CAM Publishing’s Rapid Prototyping Directory, 1998 edition, showed that 331 SB locations are now in operation worldwide, compared to 284 locations the year before. Part of this increase came from including in the equation SBs that were inadvertently omitted the year before. For example, the 1998 directory showed the number SB locations in Japan jumping to 39, up from 10 locations the year before, yet 29 new SBs did not emerge in Japan over the past year. Revenue growth among SBs worldwide was an estimated 3.1 per cent in 1997 ? perhaps a more accurate reflection of RP growth. In 1996, growth in this market segment was 43.1 per cent.

Many SBs continue to rely heavily on RP models for revenues and profits. These are the companies that are experiencing the greatest problems. Other SBs have expanded their businesses to include a range of complementary services, such as design and engineering services, rubber tooling and vacuum casting, aluminum-filled epoxy tooling, CNC-machined molds, injection molding of plastic prototype parts, and reverse engineering. These companies have also felt the pressure from lower margins on RP services, but not nearly to the same extent. Many are doing fine, and some are doing exceptionally well. One of the keys to survival in this business will be to offer the right blend of engineering, RP, rapid tooling, and related services without stretching company resources too thinly. Some will do it successfully; others will not.

Many factors will separate the successful from the less than successful. SBs must provide on-time performance and deliver as promised. They must be responsive and dedicated. Because they must offer competitive prices, their operation must be lean and efficient. High quality is assumed and they must deliver highly finished models at a reasonable price. If possible, SBs should provide a service that distinguishes them from the competition, although this is becoming increasingly difficult.

Other factors are key, such as hiring and retaining talented, experienced people. Some customers want to see size and capacity. Visibility in publications and at industry events helps create identity and credibility. Taken together, these factors create character and a reputation that are critical in this relatively small, ‘word-of-mouth’ industry. So will RP service bureaux be a thing of the past in the future? SBs, as we know them today, will look and operate differently in 10 years, but it is very unlikely that all of them will disappear. Companies that offer a strong mix of services, run efficiently, and stay abreast of needs and industry trends will survive the pricing crunch. Many others, however, will fade away due to the competition, their inability to change with the times, and decreases in RP system prices. This painful process is underway.