I work for a custom software development and consulting company that employs the Rational Unified Process. In support of this process we often quote a statistic published in around 1995 that states that 70% of software project fail, either by going over budget, delivering late, failing to meet the customer's expectations, or failing to be completed altogether.
Is anyone aware of more recent and up to date statistics regarding software project failure rates?
Karen, delivering late is a major problem that RUP or OOD has not solved. In addition, you have 40% of IT projects being terminated during the course of a fiscal year at most major corporations as business needs change. RUP/OOD does provide higher quality software but increases your overhead and does not, in my experience, shorten the software development life cycle as promised. Many companies that use RUP have not done the training to properly leverage Rational Rose. OOD is something that works poorly on the cheap. Saving Changes...
Anonymous
Tom, it's interesting that you say that, because I'm actually developing marketing literature for a new technology and process we've developed that allows us to cut development time approximately in half, guarantee delivery, and cap the cost. We're actually able to do this without sacrificing the quality of OO and n-tier design.
We're currently developing web-based applications at lightening fast speed, and we're very eager to get the message out that we can do this.
Karen, OOD does produce a higher quality product and certainly much easier to maintain once the 1st release goes live in the long term. Still, shortening the development cycle is another matter. Now, IBM, at least according to its TV ads, promises to get companies on the net for $498 full price, but that's using labor from India [at 1/16 labor rate of American developer] and various set formats, so I'm told. Your product will be a success if you address SIZE and COMPLEXITY issues that Rational Rose and other methodologies do not address, except for Productivity Plus. Saving Changes...
It'll cost you; but the updated information can be found at the Standish Groups website. The Chaos studies are the ones that Rational refers to. The stats are getting better; but the message is the same - most projects fail to meet requirements on time and on schedule. Check out www.standishgroup.com or
Karen - Are you really interested in promoting this new process? I do not see a way to contact you, access a demo or ask more questions? Saving Changes...
Karen - I'm a PM working on software development and have seen and/or used just about every approach to development (RUP, OOD, off-shore/onshore development etc etc). All have inherent virtues but it is often the way they are used and the lack of proper training that causes problems. No news there. If you have a process that can reduce the cycle time without compromising quality and scope, I'd like to hear more. Can you tell me where I can see a demo? Saving Changes...
IT Project failure is a myth! On time/on budget is relative. Is success based on delivering everything promised in the initial estimate? The software process is a complex system. We refuse to accept the fact our estimates are based on "partial knowledge". You never know true time and cost until you're finished. IT projects are a joint venture between IT and Business. Business is a part of the process. As feedback occurs during the project the Business helps make decisions on modifying cost and schedule. There also has to be a culture of accepted failure and the ability to fail early. This, of course, assumes the staff is competent. Can you predict from the initial estimates that ALL functionality will be delivered with quality? Can you tell me if it?s going to rain on Monday six months from now?
Iterative processes like RUP try to accommodate this ability to modify or fail by having working and use accepted pieces of the system built in increments. These milestones provide the proper feedback to help EVERYONE make informed decisions on cost and schedule. Easier said than done but far better than traditional waterfall. Saving Changes...
Frank WintersPhotographer and ConservationistSandwich, Ma, United States
The Stadish Groups Chaos report is an excellent source and it just so happens that there is a copy of the 1999 version of the report available on the gantthead site under "White Papers". Its dated but still useful -- things have not changed too much, and anyway this report will give you a look at the kind of information available from Standish. Saving Changes...
