February 2006

Sun Mon Tue Wed Thu Fri Sat
      1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28        

New Blog

Subscribe w/Newsgator

  • Subscribe in NewsGator Online

« The Real Problem in Bay Area Real Estate | Main | 802.11b on edge of extinction »

Apr 07, 2005


Ramana Kovi

I belive Software-as-a-service is here to stay. I think all your points are valid, pricing, delivery are important components of the SaaS model. You are right on what cutomer is looking for - "flexibility to customize the applications to fit its needs, dispose it when done with it."

I think the big vendors (i.e oracle,SAP and siebel) most probably will play less of a role because it requires complete transformtion of changes in thier product line and thier sales processes.

Dennis Posadas

It will of course, benefit the software companies, in that they can forecast an NPV from their stream of cash flows in the future.

Dennis Posadas
author, Rice Bowl & Chips
How Asian countries are using the Silicon Valley
model to develop technology startups
(ISBN 0-595-34583-2)

Adam Messinger

SOA and SaaS are related by one of the efficiencies which drives SaaS, namely the reduced cost of engineering which comes from not having to support the software in varying environments nor support many versions of it. To maintain this efficiency one must provide an integration API (so that customers don't demand you run in their environment for integration purposes) which is easy to upgrade (so that you don't have to keep supporting old versions). SOA is the natural way of providing such an API because it is loosely coupled (and thus good for integration) and coarse grained (and thus good for versioning).


SaaS also supports agile software development much more effectively than what we have been used to, which in itself drives functional component development to enable the small parts loosely connected model.

Kip Watters

For the most part, I agree with your concerns and perception of software as a service changing the Enterprise software delivery/license model and being here to stay. Two points come to mind:

As for the possibility of a customer simply "canceling a subscription" and using another service, we need to remember that the end user of these services really doesnt know, nor do they care, that the software is being delivered as a service. All they know if that it took a hell of a long time to learn the software in the first place, and once a company overcomes the friction and pushback from users to a new application, it probably doesnt make sense to just "switch out".

Also, in the case of, for example, quite often companies use this service because it is a smart thing to do and a very effective way to organize sales leads, etc. However, the real value lies in the customer relationship which does not necessarily require CRM services to maintain, in which case does not become part of the backplane of their business. When we start talking about SaaS in the context of product catalogs, inventory systems, integration infrastructures, etc. -- there is a substantial non-software, non-subscription-based/license fee switching cost. Service based delivery of software does not change the fundamental economic principles of switching costs and barriers to entry that the software industry has so adeptly built it's market upon.

One other thing to mention -- basing a subscription on a by-the-slice/per function pricing model will not work, in my opinion. Subscriptions will work, usage-based subscriptions will not. Isn't this what AOL tried with internet access, what my former (now out of business) company tried with WS Management/SLA measurement, and what is at the root of why Vonage and other VOIP companies are growing like gangbuster? At the most basic level, it also discourages strong adoption, and ultimate reliance on the services that the software companies depend on.


I was going to include a few comments about but decided to not get into naming names, especially because I can't help but have a bias considering my position with SAP.

Nonetheless, I think the CRM/SFA space is unique in many respects in that all CRM/SFA applications and users tend to be pretty similar. It's about as close to a generic app as you can get to in the enterprise, and that's also why you are seeing strong uptick for SugarCRM as well... all at the expense of Siebel.

My point is simply that this space is somewhat of an anomoly, but over time it should be expected (and encouraged) to baseline many areas of enterprise applications in order to be multi-tenant hostable. My comments right now reflect today's perspective on market dynamics... so I'm not entirely convinced that represents a watershed event in the SaaS movement, but no doubt the success they are having fuels the trend.

Eyal Rosenberg

I believe SOA will play major roll in the SaaS arena, as enterprise applications are actually set of services implementing enterprise’s business processes. The shift, or I can say the upgrade, to SOA will enable SaaS vendors add value and improve the existing business processes in places traditionally preserved to the enterprise application vendors and to tailored customizations.
Moreover, SOA will simplify the integration with existing applications and business processes thus will accelerate time to market of SaaS vendors in the penetration process.
Therefore I think SOA will lead to faster adoption of the SaaS by enterprise customers

The comments to this entry are closed.