Critical Chain Project Management

While in the UK recently, I co-presented a paper entitled "A Critical Look at Critical Chain". Critical Chain Project Management (CCPM) has emerged in the last few years as a novel approach for managing projects. In this paper we analysed CCPM in the light of its contribution to project management practice and to project success. We started with a brief review of the key elements of CCPM - the revision downwards of duration estimates; the use of explicit buffers, progress measurement through measurement of buffer consumption; and priority setting.

CCPM is claimed by its supporters as "The most significant advance in Project Management since Critical Path", but its detractors dismiss it as old ideas recycled with a lot of hype. The paper examines some of the evidence. It also considers the place that CCPM occupies in the broader project management context, and the costs associated with its adoption.

Our conclusion is that, although CCPM has valuable concepts, it does not provide a complete solution to the needs of project management. Consequently, organizations should be very careful when considering the adoption of CCPM to the exclusion of conventional project management techniques and methods.

Our paper is available at http://www.robertb.co.nz/CLCC.html, and our PowerPoint presentation at http://www.robertb.co.nz/Presentations/RazBarnes4.ppt

Revolution or Evolution?

In the dot-com era's heyday, "Revolutionary Change" was the watchword. "The Internet Changes Everything" dominated the debate (and the NASDAQ), and many companies which rushed to adopt an Internet strategy forgot that business is all about business, and not about gee-whiz technology. Harvard Management Update has a annual "Burning Questions" conference - 'Revolution" was last year's theme. A year later, the theme has changed. Forget revolution, now it's evolution - an emphasis on small, frequent managerial adjustments. The analogy is that of Mississippi river boat captains, who only navigate as far as they can see - to the next bend.

The same theme has long been a staple of well managed IT projects. Management may want the detailed project plan covering the full five year implementation and the multi-million budget, but Grosh's law - "If a system doesn't produce something useful in six months, it never will" - remains as true today as ever. Inexperienced analysts often produce a single phase project to avoid the obvious and considerable amount of extra cost in the scaffolding of an incremental approach - building temporary systems to allow phase 1 of the new system to operate smoothly alongside the old, and so on. Less obvious are the costs of not building the scaffolding, but not discovering that the solution was wrong until implementation time arrives. Just ask the developers of INCIS!

The Semantic Web

An article in May's "Scientific American" http://www.sciam.com/2001/0501issue/0501berners-lee.html envisages the future of the Internet. Computers navigating tomorrow's Web will understand more of what's going ongoing on - making it more likely that you'll get what you really want. Most of the technology components to achieve this, such as XML, are already in existence.

This is an intriguing (spelling?) article. The "Semantic Web" feels like a combination of Knowledge Engineering (expert system) concepts combined with the Internet? Will this be a major, or (like expert systems) end up as a niche (although valuable) technology. Time will tell, but the concept of the Semantic Web is worth watching.

From HBR, Leading Through Rough Times: An Interview with Novell's Eric Schmidt

(HBR June 2001)

"Few large companies have soared as high, sunk as low, and struggled as long as the 18-year-old networking software maker Novell. For years, the company dominated the market for local area networks, but by 1997, it had faltered due to misguided acquisitions, product missteps, and large unsold inventories. That's when Eric Schmidt arrived from Sun Microsystems to take over as Novell's third CEO. He turned the company around with a deft combination of cost reductions, divestitures, and new product rollouts, and by 1998, it was back in the black. Unfortunately, the good times didn't last, and like most technology companies, Novell is once again struggling with a slowdown in demand. But Schmidt is optimistic about returning Novell to good health, and his strategies suggest ways for other organizations to handle themselves during downturns. He counsels against being overly cautious during such times. His advice: keep new products coming out to sustain the interest of customers and the press, pay attention to your cash position, stay focused on your desired outcomes, and take heart from other industry leaders."

In my "Strategic Use of IT" course "Demand-Side Economics" is a term used to describe markets, such as software products, in which the marginal cost of production is relatively insignificant. Such markets do not tend towards stable equilibria of several suppliers, but tend towards total dominance by the leader. It used to be IBM, now it's Microsoft. This article is interesting from the view - how do you fight off the gorilla?