article page | 1 | 2 | 3 |

done an estimate and decided they wouldn't go ahead with the project because there simply wasn't enough time or enough people to dedicate to the effort of doing this product. The approval process went all the way to the top, and then they realized they should have been doing this using re-use and componentization from the catalog in the first place, and they hadn't realized what the difference in estimate was to do it that way. It was a great case study because it came back that it was about 80% cheaper to do exactly what they wanted to do, and it was the difference between launching a brand new set of 3G services in the market or not. And they could do it cheaper, and faster, and in a much more efficient way.”

Lancaster stepped in, at that point, saying “You raise an interesting point there, Ernest. The good news is that Product Catalog works. The not-so-great news is that until it gets to be integrated into the CSPs' process, they're not going to reap the benefits.”

Rainger built upon that sentiment, saying “A lot of that's around domain expertise. When you're talking about getting a process right, you've got to have more than just a good software package. You're got to have good domain expertise, and be able to advise on lexicon, and the rationalization effort.”

“You must also consider the organization needed to run it,” added Margitta. “The new people. Where are you going to get those people. That's one of the things we work with customers on is making sure to build an organization around [product catalog] and adapt their processes to use these tools. Everybody we spoke to today has a PLM process. In the telco world, everyone's got one. However, it's a matter of what the process is, how they follow it, how they measure it, how automated, how manual, how structured, gated, etc. What we're doing in every case is plugging into those

optimization, process definition, and fit the process in.”

Willetts, seeing the root of the disagreement as being a part of a larger issue, said, “This is right at the heart of the problem of re-systemization and transformation of telcos. Do you do that, do you fit the process to the system? Or do you do what telcos have done for 30 or 40 years and continue to do, and try to fit the system into the process by bastardizing it? If it's that, you don't get any economy of scale. I get a market of one and pay high prices for everything. I'm hoping that this time around, the recession will cause companies to say 'you know what? We aren't going to do all that custom engineering. We are finally going to bite the bullet and the process will be driven off of the system.' The luxury has always been there, especially among Tier Ones to say 'No, no. We do it the AT&T way or the Verizon way or the BT way'. This is especially true with unions and all sorts of other constraining factors on you, and you've got the sheer inertia of retraining staff. The whole idea of standardized process doesn't mean anything if you bastardize it.” Willetts added, “I use the word 'bastardize' these days as a contrast to 'customize'.”