EA‘s Selling Point

How to draw the interest of the managerial level of an organization when the enterprise architect is trying to sell their concept and methodology? Or how to discover the real need for EA within an organization?

I’ve got some idea when relating these questions to what I noticed during my internship in HSBC Insurance. I was in the IT project team working on the upgrading for their policy management system. In order to better understand how this system interacts and exchanges data with other systems in the company, I referred to their IS documents. I found that the main business and services is supported by five systems, including policy management system in the back office, system for sale force, financial management system, system for communication among the sale force, back office and the intranet, etc. Each system though has clear functions and efficient in its own functional sphere, the collaboration among the system is a tricky problem.

I think that is the case with many other organizations as well. They have DB2, Oracle Database and Access to store their data in various departments. The policy management system which is the core system for their daily transaction has many interfaces for those database and management systems. An upgrade of that system may cause changes of the format of output and the manner of data exchanging. So the project team had to communicate frequently with staff from other department to make sure their change to the current system is applicable. This seems to be a trivial work but in the cases when you have an outside developing team and testing team, this communication greatly impacts the schedule of the project.

One unique selling point of EA which provides solution to the problem mention above is its emphasis on collaboration. The deputy CIO of the US government agency Small Business Administration sees the enterprise architecture to be SBA's path to modernization.

“Enterprise architecture plays into the larger governance processes. When you’re doing some good governance, you’re able to see, earlier rather than later, that things may be running into challenges or running into opportunities that we can take advantage of and correct early,” he said. “It’s a continuation of the evolution of breaking down the old stovepipes and being able to share data and utilize data for new and innovative purposes.” Garwood started at SBA on July 30.
FCW (http://s.tt/1mdtC)

However, in my opinion, the architect should always stress the specific business or organization environment when trying to implement the concept. The overhaul of the IS in the Homeland Security Department has achieved some success. In reviewing of their achievement in the information-sharing initiatives, the CIO Richard Spires said,

“We brought together DHS from 22 separate organizations, all crammed together – you’d expect to have a lot of duplication and overlap, right?” he said in June at an ACT-IAC event in Washington. “We all know in government you can’t [achieve efficiencies] in six months or a year; these are multi-year kinds of things we’re trying to achieve. We’re doing this from a portfolio standpoint by function rather than looking at each one of the components individually. That’s my strategy.”
FCW (http://s.tt/1nWex)

And similarly, if the target is the insurance company I mention above, there will be elements about risk management and compliance, all of which should be take into consideration when trying to sell EA or adopt any methodology.

IT Failures

The IT failure in the system upgrade of Royal Bank of Scotland caused massive impact on transactions of their customers and even of customers of other banks. In a report titled “IT Spending in Banking: A Global Perspective” the problem that may lead to this IT failure is pointed out.

The report says:

Of the total investment in IT in 2012, a whopping 77.6% goes to maintenance. European banks are having the most difficulty in reducing their maintenance spending.
Banks will need to spend on new investments, at least partly as a result of needed system upgrades. Frequently, financial institutions are running systems that are too obsolete, too slow, and inflexible. Systems like these are impediments to achieving optimum operational efficiency as well as to developing products.

I find the situation ironic: on one hand many companies rush for some fancy applications and ideas, such as the cloud which is believed to provide more flexibility and productivity, and the SCM which take advantage of social media to get closer to their customers. On the other hand, however, inefficiency in some core business is neglected because those transactions are supported by their traditional systems and they are the maintaining part not where the main investment goes. What they want is transformation of the organizations but some perhaps fail to fully considerate the priorities in both business and IT perspective.

An earlier report from the Standish Group issued in 1995, says the critical factors contribute to the failure or success of an IT project are involving user participation, requirement specification, planning, executive management support, ect.  My understanding is that during these processes, various needs are weighted and balanced on the basis of analysis on how the organization functions to meet its long-term goal. IT is adopted to keep the continuity and robust development of the business as well as its transformation.

In that sense, how could RBS implement the upgrade system before it is fully tested? How could many IT failures are merely regarded as some accidents?

EA First Week Journal

I'd love to start with our first case Ms. Rettig's The Trouble With Enterprise Software. I was kind of surprised that the current implementation of enterprise software in organizations is regarded hopeless in solving the problems exists in the legacy systems. But it is provoking indeed. At least one thing is shown to us that the software or IT ,though fancied by many executives of organizations,isn't a solution to all the requirements. Yet the trouble of it appears in certain stage of the overall development of IT of organizations. Such as the integration with and among current systems, the collection and presentation of the data and the demand for changes in business. I think all of these entail EA. 

In my point of view, EA is a tool or technique used to interpret the vision of organization into business processes, into technologies infrastructure. Due to various reasons and context, no one can simply throw all the old systems away and unite all the functional parts within the organization using a highly compressive system. But they can manage to partly implement the technology in core department in early stage and then form regulations for operations of their business processes and finally can reach a level in which standard data can be generated for use in relative compressive enterprise systems. The whole procedure can only be realized when there is proper EA.