Wednesday, December 24, 2008

Success secrets of Portfolio Management

Yesterday, I had an opportunity to talk with divisional CIOs of a large consulting firm about my experiences of implementing portfolio management and CIO dashboards. One question from them kind of intrigued me – What it the success secret of portfolio management? Is it a process, tool or people?
I reflected on my previous experiences and stories that I have heard from my colleagues. Many large organizations put in millions of dollars in setting up Portfolio management tool, only to abandon it later for want of value. Almost in every case, the root cause is – lack of quality data and inability to keep up with changes in the data. Garbage in, Garbage out !
So what could be secret for the success? I think, it is simplicity of the solution. Strive for it without compromising the analysis value. More often, the portfolio management exercise often starts with selection of a tool and implementation of that tool. The tool is often comprehensive and offers to capture a whole lot of data and provides numerous reports. The question one should ask – is it worth the value? The first time, you can capture the data, but can you keep with the changes in the data? IMHO, the first step should be to decide what you must see every day, every quarter and during your annual budget exercise. Keep it as simple as possible. Develop a data capture process, simpler tools and governance process around it. Try to integrate tools to your operational systems so as to keep the data entry minimum. For example, get the operational FTE cost by integrating your tools with timesheet systems. Enhance your process and tools over the period in agile way as you start using it.
After a year or so, once you are confident of the captured data and the process, you may want to look at professional portfolio management tools that can act as your repository. That is solely to keep your future maintenance costs minimum and get in-built integration capabilities. It may appear reversed approach as you will need to migrate data to the tools, but trust me, if you know what you want from tools, often it will be simpler exercise to select, setup and maintain tools. Otherwise, tools are sure to overwhelm you and set you in undesired direction.

Wednesday, November 26, 2008

In the eye of the storm

Great interview with Citigroup CEO Vikram Pandit at Knowledge@Wharton -

Vikram explains the root cause of financial crisis and his solution - going back to basics of banking. He has a great advise for leader wannabes .. A must read.

Friday, November 21, 2008

Open source within your organization

I recently made a trip over the seas in order to rescue a very critical project out of missing deadline syndrome. The project timelines were very aggressive with many parallel work streams. When I was involved remotely, I could not figure out why a bunch of sharp developers can not deliver as agreed. More painful was not able to estimate correctly for the remaining work and commit to next deadline. When I sat down face to face and talked, I discovered a shocking truth - people are wasting most of the time in waiting on each other. There were defined custodians of different parts (or services if you may) and almost everybody was dependent on each other. The problem was everybody was too busy with their own priorities and not able to get the time to help out others. To fill the waiting time, they did stubs and mocks, but it all contributed to the wastage.

I read recent blog entry by Martin Flower - and I think it is very interesting solution to such problems. He proposes to adopt the 'Open source' strategy within the organization. Rather than waiting on custodians to enhance or fix their services, go ahead and change the code as per your requirements. Send the patch to custodian who will review and merge with the main code. It is always easier to review and merge, rather than code and test. Now, I am aware that it is not as easy as it sounds, otherwise, there would have not been any need of custodians. However, if it works for open source world, it should sure work within your own organization. I guess, it depends on maturity of the existing code and developers as well. It may be possible after the initial versions of services are released. It is MUST that you have a mechanism for knowledge sharing and collaboration, otherwise, no other option than waiting on others.

Wednesday, September 3, 2008

Google Chrome - Loved it !

Just downloaded and played with google chrome.

The speed is impressive, so are cool features of showing history with timestamps and ability to search through the history. How many times we need to go back to the link that we had crossed on our way, but not able to find again !

I liked 'most visited' on the homepage. Great feature to quickly launch your everyday sites. It is like browser determining your favourites rather than you having to specifically mention it.

Incognito mode will be handy on shared machines.

There are some glitches as well - most irritating being 'favorites'. It is so complicated to add any link to favorites and manage it.

I am not sure if it can be real threat to IE. Most of the audience that get benefited or excited by such features is probably already on FireFox and rest (like me) probably hate change !

Thursday, April 24, 2008

Role of an Architect

I recently published an article in Microsoft Architect Journal about the 'Role of an Architect' from system integrator perspective.

For HTML version, click -

Any comments ?