POINT OF VIEW
Sharing software platforms and hardware infrastructure makes great sense. We embrace the cloud for convergent infrastructure to make the best from fewer resources; why not share the software platforms as well?
Every business will be changed by demands for mobile connectivity and customer empowerment. Digital architectures and Digital-SOA (Service Oriented Architecture) rely heavily on shared infrastructure to achieve time to value and cost targets.
Software platforms that deliver significant value when shared across projects and departments include your Enterprise Service Bus (IBM Application Connect Enterprise, WebSphere, Responsiv Unity), used to deliver connectivity and data replication controls, Process Automation (IBM Business Automation Workflow, RPA, and Case Management), used to develop efficient business operations.
Sharing software platforms can help accelerate project delivery, improve skills, and bring forward the time when you see value from the project.
Sharing these capabilities spreads the cost of software licenses and maintenance, reduces procurement delays for individual project, and allows your business to develop deep skills and experience with the technology.
Shared infrastructures that are less obvious but more critical to brand and reputation include Web APIs and B2B connections.
Building a shared computing capability is expensive and can incur risks which are often outside the experience of many project managers. A shared platform must do a lot of different things to keep everyone happy and the pitfalls are plentiful.
How is shared infrastructure funded?
Possibly the biggest challenge is to create a funding model that supports the immediate needs of the business as well as its long-term strategy.
Project based funding
Many organisations fund all of their IT by providing capital budgets to each project and expecting the investment to sustain the technology over several years. This approach requires significant guess work and leads to equipment and licences laying idle for a predicted “peak year” in the future which may never happen.
A better way
The ideal would be to balance cost recovery and sustainable funding against the business benefit realised by each set of business users.
A critical success factor is that all participants feel they are getting value for money and that the infrastructure is governed in a manner that does not risk their business. A funding model is needed that properly allocates cost and that allows proactive maintenance and capacity management. Such a model must be transparent and flexible to accommodate changes to usage patterns.
It is essential to begin by determining a solid and predictable basis for the funding of your shared solution. Basic cost will vary according to size and quality of solution but should include everything, including design, implementation, maintenance, and training.
The way a solution is priced can affect the way it is used. For example, models that charge for each request will result in fewer requests, while a cost-avoidance technique like batching requests may increase them.
The way a solution is priced can affect the way it is used. Financing models that link cost to benefit tend to have the greatest success.
Some systems can be designed to make the most of human behaviour, thus spreading demand and cost – perhaps across time zones if you are a global enterprise. The answer to this could be anything from a pay-as-you-go system, which charges per operation, to a generalised “tax” on the firm.
Funding models that draw a budget tax each year to cover the known cost of running technology tend to hide the true costs of some applications and do little to inform business decisions.
But finance models that link costs to benefit tend to have the greatest success. If you can demonstrate a clear cost to benefit relationship you can identify the economic value that is added by your IT infrastructure more easily. For example, ascribing a cost to a business activity like the process of performing a meter reading can expose the true costs and benefits to the business. Cost information can be used to calculate the economic value added by your IT.
Taking an open market approach that requires services to attract customers and fund themselves, allows each service to have its own price. The result is to link costs and benefits.
Services effectively declare their cost of operation in a catalogue and users can decide whether the cost is justified. In this model the Enterprise Service Bus and other middleware deliver a service and as such set the price.
The price of a transaction is often asked, seldom known, and almost never used as the basis for spending decisions. Responsiv believe that a pricing model based on easily measured metrics aligned to consumer values and technical strategy will always win.
Richard Whyte has been building enterprise IT solutions for over 20 years. He is known for creating innovative practical solutions that provide a strong foundation for future development, whilst solving immediate problems. Previously the European CTO and Principal Architect for IBM Systems Middleware at IBM, he has an MBA, a degree in Statistics and Computing, is a Chartered Engineer, a Chartered IT Professional, and Fellow of both the Institute of Technology and the British Computer Society.