The theory goes like this. Everything service the user touches can be separated from the underlying transport. The transport is packet-based, low latency and with as much bandwidth as you care to buy. Internet Protocol was deliberately designed to assume the minimum possible set of things about the application. Hence any reasonably foreseeable user application can be bought independently of access. A complete re-structuring of the industry will occur as a result, because the vertically integrated telco business model becomes obsolete. The end-to-end principle and it’s corollaries, in a nutshell.
It’s important to know the boundaries of the applicability of your theory. If you don’t know them then there are two possibilities. Either you’re deeply into string theory and cosmology, in search of universal truth without bounds. Or you have a dogma, not a theory or a principle.
There are many such boundaries to the end-to-end principle, but today I’m going to focus on only one of them: branding.
Suppose you’re interested in a service that performs a remote backup of your PC. You have a choice of suppliers, because that’s the joy of the dumb network — you’re not tied in to your access provider for the services you need. But you’re a small business, and you really rely on this service. Several employees have had their laptops go kaput in the last year. This thing needs to work when you have a problem. Who do you turn to for the service?
The problem is that smaller start-ups may not have the brand strength to compete against an incumbent telco offering the same service. If you really care about the supplier being around the day after a hard drive crash, then telcos have an impressive history of longevity. The nature of capital-intensive multi-billion dollar network industries is that things tend to move slowly. The turmoil of the last decade has been the exception, not the rule.
Once the dust settles from the decline of the PSTN, access providers will still be advantaged in offering critical services. This could be due to concerns about availability, privacy, reliability or other non-functional aspects of the service. Remember, in our laptop backup example you could be relying on that provider staying in business. The recent past would make many wary of “backups.com” becoming a key supplier.
At the same time, those access providers are unlikely to be performing much service innovation themselves. It’s a question of how to associate 3rd party services with their own brand. One common way today is through a portal, which acts as a trusted recommendation engine to premium content and services. You can go elsewhere for free, but this stuff is all guaranteed to work on your phone and be exactly as described. But the fenced garden isn’t the only way.
Another possibility is what you might call the “Marriott Model”. In this case, their hotel franchise has been extended into new sub-brands, such as Courtyard (low-end business travel) and Fairfield (family travel). The logo on the hotel reads “Courtyard by Marriott”. You’re getting the brand promise, but the ability to do service innovation goes beyond the management of the traditional Marriott hotel chain. Not very far, but it’s something.
It isn’t hard to imagine a telco hosting a service, and then imbuing it with its own branding and privacy/security/availability promises. For instance, you know your data is secure, because it is inside a hardened telco network operations center. The customer gets the best of both worlds: innovation as well as operational trust. Who knows, maybe one day we will be seeing “AOL Vonage Voice Service by AT&T”!
Telcos aren’t the only big players with strong brands. But given you have to buy access, they’re strongly positioned still to retail services (even if they rarely conceive or product manage them). The end-to-end principle allows the functional aspects of the service to be located anywhere. But it doesn’t automatically distribute to the network edge non-functional attributes such as financial stability of the service provider.
Radical industry change is in the offing, and many participants will become extinct. Yet it would be wrong to assume that end-to-end means decimation of all telco participation in service revenues. The brand, distribution network and existing customer bases are too strong for that to happen universally. The challenge is to understand all the things the customer values — not just the raw service functionality — and use your brand strength and operational capabilities to stay part of the value chain of a 3rd party product.
Posted by Martin Geddes at 12:43 AMTrackBack URL for this entry:
http://www.telepocalypse.net/cgi-sys/cgiwrap/mgeddes/MT/mt-tb.cgi/134.