Friday, January 30, 2009

Era of On-demand

Soon we will see an era that will change the way people buy, use and experience software. Customers will be able to choose between buying as a whole or just based on usage or any other similar factor. This will help us to see software as a commodity like electricity. We do not have a power plant at our home to use electricity. We buy as much as we need and pay for what we use. On Demand Information Management concerning content management and business intelligence will gain a lot from the on demand model. Today it take a lot of time and money for a company to implement a full fledged CMS or BI system. Based on certain parameters it will be prudent for some companies to adopt the on demand model because this will allow them to save on the upfront cost.
Posted by at

2 comments:

  1. Anand MohanFebruary 7, 2009 at 5:48 AM

    One of the challenges of the on-demand model is the pricing strategy. Mostly all on-demand vendors operate at low margins ranging from 10% to 15%. Companies come up with different types of pricing strategies to maximize profit. Some try to give the basic offering for free and then charge for the value adds. Some charge based on the value derived by the customer out of the service. Another model might purely go on customer usage or productivity. The pricing strategies might also differ by the type of customers (big, medium and small). The vendor should try and identify the right strategy for the service which is offered on demand.

    ReplyDelete
  2. Anand MohanFebruary 8, 2009 at 7:09 AM

    From Gianpaolo's blog an interesting concept on the SaaS maturity models can be understood. Though there will be lots of questions this is a good starting point:

    Saas Maturity Models
    Level 0
    1. Each tenant run on separate instances
    2. Each instance has customization done for the customer
    3. Economy of scale is not achieved because each customer runs on a different instance of the software
    Level 1:
    1. All tenants are running the same code and customization can be done through configuration
    2. Each customer still owns their instance - the computing power is not shared among the instances and therefore the provider cannot achieve economy of scale, putting it at a competitive disadvantage vs. a multi-tenant solution
    Level 2:
    1. Includes multi tenancy concepts
    2. UI, business rules and data model can be configured for each customer and all customers share the same application.
    3. Customization through configuration is done through a self service tool so provider intervention is not needed
    4. Economy of scale is still not there because there is only one instance of the software
    Level 3:
    1. All capabilities of level 2 + scale out of the application
    2. New instances of the application can be transparently added to the instance pool to dynamically support the increasing load.
    3. Appropriate data partitioning, stateless component design, shared metadata access are part of the design
    4. Load balancing to maximize the utilization of the hosting resources like CPU, storage etc.
    5. Data periodically moved to average data load per instance
    6. Architecture is multi tenant, scalable and customizable through configuration
    7. Single production environment to support all customers

    ReplyDelete

Subscribe to: Post Comments (Atom)