Editorial : IP Content Delivery Network Readiness

About the only thing service providers have in common today—other than investor loathing—is they all want to jump into IP content delivery networking. But are they ready, from a billing and OSS perspective?

First, let’s define IP content delivery networks (CDNs). Here the term means the transport of information from a source (content provider) to destination (consumers) using the managed IP network(s) of service provider(s), and using their billing systems. The plurals are important, because CDNs will deal with multiple IP networks and billing systems. The key networks we are talking about for consumers are 3G mobile wireless, cable TV and DSL; the content at issue is the content consumers are willing to pay for.

So, what are the billing system considerations for value-based content delivered by IP networks? Below are six models you see in service provider’s networks today.

1) Home Built: Billing systems built by service providers are typically referred to as legacy systems. For better or usually worse, they are found in RBOC networks and go by the names of CRIS, CABS, TIRKs and others. The advantage of these and the reason the RBOCs continue to maintain them centers on the upward volume scalability and reliability of these beasts. The downside is they aren’t flexible, require massive IT departments to run and are not easily adaptable to IP content billing.

2) Smokestacks: These are billing systems with their own associated OSSs and customer databases, and they are rampant in today’s RBOC and large IXC networks. The RBOCs deployed them when they started to create separate business units for new services, thereby allowing product managers to get the job done no matter what it takes. Large IXCs, like bankrupt WorldCom, created smokestack billing and OSSs via all of their acquisitions.

The problem with adding another smokestack for IP content is that operators are trying to consolidated their billing systems, not add more. They can’t justify the additional operational expenses and inefficiency, and this model offers no opportunity for service bundling, cross discounting and a single customer service representative (CSR) view for customers.

3) Outsourcing: Major wireless players today such as AT&T Wireless, Verizon and Sprint PCS use this billing method, as do many cable companies. Its advantage is greatly reduced IT departments, scalability and—in the old days of simple service (2G voice or cable)—speed to market.

4) Managed Product: The alternative to outsourcing is to own the billing system but have the billing system vendor manage the product. Again, the advantage is reduced IT department staff, scalability and system integration with other OSSs. Wireless players such as Cingular, VoiceStream and Nextel have gravitated to this model.

5) Off-the-Shelf: This model is popular with Europeans, particularly GSM wireless players. You buy an off-the-shelf billing system, the vendor installs it and supports it, but not nearly to the degree of the managed product model regarding vendor staff on site.

6) Best-of-Breed Components: This model calls for breaking out the middle layers of billing such as mediation, rating and provisioning. The alternative to decoupling these functions is to have them embedded in the billing systems under models 1 through 5.

Billing for IP value-based content stands out as the first challenge because as we have said flippantly many times, "If you can’t bill for it, it’s just a hobby." But other business support and middle layer requirements can make or break CDN business models.

3G Wireless Readiness

The wireless industry in North America is chomping at the bit to get into IP content delivery. But how do they stack up in the readiness chart?

First, on IP billing, with the exception of Sprint PCS and VoiceStream, combinations of all the above six billing models exist due to the many mergers and acquisitions of Cingular, Nextel, AT&T Wireless and Verizon over the years. And it’s only going to get worse. Come January 1, the FCC will lift the spectrum cap for PCS/cellular license holders.

However, because of likely Department of Justice opposition, I don’t think you will see en masse mergers, but instead an unbundling of licenses so carriers can up their spectrum in certain markets a year from now. I see three or four stronger players with a hodgepodge not only of wireless technologies (TDMA, CDMA, iDEN and GSM) but also of billing models. Adding an IP-CDN billing requirement to one model is tough enough, but having to merge it into multiple billing models at the same time will be a challenge that wireless service providers may not be in a position to handle right now.

What about the other items in the above table? Equally challenging. Take for example the product catalog. When you come down to it, the single most important factor in Japan’s DoCoMo i-mode is that it lets users choose from more than 1,500 content products. U.S. wireless carriers don’t have anything near ready to match DoCoMo. If you can’t catalog what content you have, you can’t order it, if you can’t order it, forget about billing for it!

Decision support systems (DSS) are another function required for success of IP wireless content. Today wireless carriers in general haven’t a clue about the content preference of their users, let alone other basic demographics. Yes, you can bill for WAP usage by the minute, but people use WAP-enabled phones for content—yet what content? Wireless DSS is not ready for prime time, either.

Finally, features such as provisioning for QoS sessions, rating of content sessions for advice-of-charge, cross-product discounting, and revenue sharing and settlements with content providers are not very high on many wireless providers’ budgets today.

Cable Readiness

Billing for IP content delivery over cable is still on the drawing boards for many reasons. Here are three.

• Before you can bill for IP content, you need IP content delivery technology. Yes, cable companies can provide Internet access via cable modems, but they use DOCSIS 1.0 cable modems. The next generation is DOCSIS 1.1 with packet cable add-ons (multimedia termination adapters for the true QoS, multimedia, VoIP and more). The first problem is that this stuff is not ready for massive deployment (see my June 2002 editorial ).

• Problem two is that in general the IP multicast of content goes to a PC, not a TV set. Does the cable industry want to lose eyes in front of the TV which holds lots of non-IP profit potential for movies on demand and so on, or see "the eyes" migrate to PCs, for potential free IP content?

• Finally, with regard to billing and OSS standards, yes, the cable industry has CableLabs, but its focus on standards tends to be on cable network elements (modems, set-top boxes and the like), not on billing and OSS interfacing. Bottom line: Billing and OSS for cable are going to be a tough nut to crack for some time to come.

QoS provisioning requires the next-generation DOCSIS 1.1 modems. Rating flexibility doesn’t exist in cable systems at the present time, either. All products are hard-wired to subscription rate plans, and content settlements are there but they are generally tied to simple subscriber number metrics.

But on the bright side for cable, it does have a product catalog infrastructure in place. Go to a certain channel on cable or satellite, and you can order up what you want. Also, cable companies have DSSs in place and therefore know a lot about user content preferences.

DSL Readiness

Next-generation DSL content delivery networks will carry special content created for their high-speed (384 Kbps and higher) incoming delivery capability. What you get today over DSL (and cable modems, for that matter) is the same stuff you get for free over dial-up, except faster and always on. The other major consideration for DSL IP-CDNs is that RBOCs and portal companies are forming partnerships, such as Verizon with MSN, and SBC with Yahoo. The billing and OSS infrastructure responsibilities are not only between the RBOCs for DSL line provisioning and the portal for content packaging, but also the managed IP carrier for delivering content from a server to the RBOC’s DSLAM.

The business model that includes all the players needed to make DSL IP-CDNs a success has yet to surface. But here’s how I see it rolling out. The portal (Yahoo) or content aggregator (AOL, for example) will do the billing. They will settle or share revenue with data center operators (Akamai, for example) and RBOCs for DSL access.

Another reason the portals will likely be the billers for premier content is that they have a product catalog system in place and DSS (they know a lot about their customers and their content preferences). What about a billing system for portals? That’s a work in progress. And what about backbone IP networks, QoS provisioning, and IP rating and settlements? Still on the drawing boards.

The good news for all service providers, billing and OSS professionals, and their vendors is that they can expect to see lots of fascinating challenges ahead. The bad news is that it’s going to take a lot of resources to meet them head-on. Meantime, if you need more information about IP-CDNs, TeleStrategies has a full slate of seminars and conferences scheduled for this fall. Check out www.telestrategies.com or call 703.734.7050 for program information.