Most mobile operators have started by rolling out as much infrastructure as possible and throwing in the software to run the entire set up. They grow so quickly that their only focus is on adding capacity. Then they find themselves hitting a plateau and realize their systems are unmanageable and have to reverse engineer the management infrastructure.
TM Forum president and CTO Martin Creaner says operators are beginning to understand that there will always be new infrastructure technologies, which means their differentiation will have to come from their execution capability and management software.
"The profitability of 21st century telecom is all about management software. It reduces your cost and increases your flexibility if it's well-executed to enable you to roll out new services," Creaner explained.
However, he adds that it is not the new services themselves that make a company profitable in the long term but the flexibility to adopt new services when they come along. "That may sound heretical, but it is really your ability to turn on a dime, not the new services or infrastructure, that makes you profitable," he said.
He argued that most operators don't think up new services. "There may be a hundred or so interesting new services a year. We spend years looking for the killer application, which doesn't exist. The killer application is just the next clever thing somebody comes up with."
Creaner said flexibility becomes a competitive advantage when confronted with the hottest new service. Those who do come up with it might get a six-month start, after which everybody else rolls in behind. "If you've got a really flexible software infrastructure, you can roll in behind maybe in two or three months as opposed to six to 12 months," he said. "That's where the profitability comes from."
There are many challenges in developing this operational flexibility. A major issue is how to push a huge back-office transformation through management and get a buy-in for what could be a six-year project when the executive team is thinking of a three-year cycle. Considering that the lifecycle of the average transition project is far longer than the lifecycle of the average executive, he suggests that the best way to get the project through the top brass is to build in short-term milestones.
"It has to be built on six-month deliverables," he said. "You have to keep giving people a win even if it's the less efficient way of doing it - by doing it in six-month chunks, it is going take seven years rather than five."
The decision-makers can feed a win out to the marketplace every six months or so, by picking off the easy wins where you get the biggest bang for the buck. "The management team is going to change and you're going to go through recessions and booms. Anybody who started a transformation project 18 months ago didn't see where we are today. You can't build a seven-year plan; you have to build lots of six-month plans."
Avoiding the one-stop shop
Another stumbling block in the transformation process is that a company may be closely affiliated with certain vendors, with ties that can impede the telco's progress. However, Creaner said that while these links are a matter of course, operators want to have their cake and eat it too.
"They don't want to do a one-stop shop [with a vendor] because then they would be locked-in," he said. "But very often they end up doing that because it is the quickest solution. This has been a tug-of-war for years."
Creaner said the operator then finds itself in "a horrible locked-in situation and they grow to hate the vendor" because the vendors charge them dearly for every single change request. Aside from that, there are proprietary interfaces in one-stop shops, "so the operators cannot pull out, say, their performance management system and drop in a new one - and they wouldn't anyway because everything is tightly interlinked."
The other option is the best-of-breed approach where there are open interfaces against standards like those of TM Forum and other interface standards. Of course, an operator needs a systems integrator that is going to knit everything together.
"Our business is far too complex; it's never going to be plug and play," said Creaner. "You are not going to be able to buy six different software systems and build them all together within a week - it will be a major integration challenge."
He believes that nearly all the major tier 1 and tier 2 firms are moving away from the one-stop shop approach. On the other hand, the "vast majority" of the tier 3 and tier 4 firms like a one-stop shop because they have no technical expertise, they have to get up and running quickly, and they have a very limited budget.
"What these smaller operators want is to spend their budget on hardware infrastructure and get the quickest, dirtiest integration solution. They will worry about the software, making it more sophisticated, at some stage in the future," he said. "That's probably the right business decision, but they are building a huge headache for themselves in the long run."
The TM Forum president said that many tier 1 providers, which ramped down their strong in-house technical expertise in favor of outsourcing, are beginning to rebuild because they feel that some projects are better done themselves. Many operators continue to handle these projects in-house because they feel it gives them more control and because they consider their market different. "There are some companies like Verizon, Chunghwa Telecom and NTT DoCoMo whose OSS/BSS developments are nearly all in-house," Creaner pointed out.
"I've heard that so many times from companies like NTT; 'No, we can't buy a commercial billing system because the Japanese market is different,'" he added.