May 7, 2003: In this week's commentary: To make money, service providers should concentrate on delivering impeccable service quality, rather than on performance-enhancing service management tools.
Developing commercially viable online services is hard enough. Imagine how much harder it must be to make money from a network dedicated to accelerating online services. However good its technology is, the venture is always going to be dependent on the emergence of a large enough pool of online providers that it can sell its services to.
No Lack of Willing Providers
Yet despite the formidable odds stacked against success, there has never been a shortage of startups coming forward with the aim of claiming this market as their own. Last month, the latest emerged from stealth mode when Netli unveiled its NetLightning service (see Netli Speeds Web Apps).
Read and React
"Service has more dimensions than the "speeds and feeds" by which products are measured. A good service is not necessarily the fastest. Being reliable, consistent, responsive when things go wrong and proactive about continuous enhancement — these things count far more than raw performance."
It takes nerve to enter this market, as the stakes are high even to get to the table. Netli, backed by $21 million in venture finance, has taken three years to develop its technology and build out its infrastructure. An earlier entrant, Bang Networks, which launched in April 2001, had more than $32 million behind it. Volera, a joint venture by Novell, Accenture and Nortel, started out in February 2001 with more than $80 million in cash. Accellion, which began life as Spacedisk in early 2000, has raised a total of almost $30 million. These initial stakes always seem to be justified, however, because the potential returns are reckoned to be huge.
That's Why They Call It Gambling
The trouble is, the gambling analogy is apposite — success is very much a lottery. It's simply a matter of being at the right place and the right time with the right level of infrastructure. Take content delivery network (CDN) giant Akamai as a case in point. If you're successful, you gain a virtual monopoly and your winnings start pouring in.
Or at least, that's the theory. As Akamai will tell you, content distribution hasn't in practice turned out to be a goldmine, any more than other, similarly hyped communications opportunities of recent years, such as broadband connection or Internet backbone — despite all the money staked on each of those territories. Thus Akamai and its rivals such as Mirror Image Internet and Cable & Wireless are themselves moving in to add Web services and application delivery capabilities to their offerings. Akamai's linkup with IBM last week was another variation on this theme (see IBM, Akamai Boost 'Virtual Capacity').
Why Fund a Service Management Startup?
With Akamai and other equally well-established companies as competitors, how are hopeful startups still able to command funding? Perhaps the real reason so much is staked on these ventures is a combination of factors that make them easy to sell to venture capital investors:
The amount of investment required acts as a deterrent to potential competitors
The technical proposition relies on patentable, proprietary technology, and thus it's easy to demonstrate in the business plan that market share, once gained, can easily be defended
Various market projections can be assembled to demonstrate that customers are certain to emerge, but since the customers don't already exist, no one can go and ask them tricky questions like, do they actually need the type of service proposed in the business plan?
Fortunately for Netli, it seems some customers have already signed on for its services, but the current fashion for XML-based standards to underpin services architectures may yet remove any proprietary advantage it might be hoping to establish.
Meanwhile, despite all the money that has been poured into providing the physical infrastructure for distributing Web services and online applications, very little development work has been done to provide assistance in monitoring service quality and delivery performance.
The irony is that this is probably where the real goldmine lies. What technologists often forget about service is that it has more dimensions than the "speeds and feeds" by which products are measured. A good service is not necessarily the fastest. Being reliable, consistent, responsive when things go wrong and proactive about continuous enhancement — these things count far more than raw performance.
Quality Still No. 1
Service providers that aim to make money will concentrate first and foremost on making sure they deliver impeccable service quality to their customers. Only if they succeed in doing so will they then explore whether they need to deliver their services faster, but those who choose to buy into performance-enhancing delivery networks will always be a subset of the wider market for service management tools.
Fortunately, that need has not been totally neglected by startups, with Web services management software vendor Amberpoint this week unveiling the first such product specifically designed for Web services. Its Service Level Manager product may not be as sexy as Netli's NetLightning service, but chances are it fulfills a more pressing need for the average online service provider.
Do you have a comment or question about this article or the ASP industry in general? Speak out in the ASP Discussion Forum.
Phil Wainewright founded ASPnews.com in 1998 and is the publisher of Loosely Coupled. He can be contacted at
IE 7
Firefox 2.0