On November 5, Matthew Heusser posted on an article on CIO.com, "'No Estimates' in Action: 5 Ways to Rethink Software Projects." The article purports that estimates are not essential in development, and this idea has been pushed by Woody Zuill, a development manager at Hunter Industries, through the use of the hashtag #NoEstimates.
Obviously, we at DCG believe in estimation and its value. Our clients do too. It's a key part of our business.
However, I quite understand the distraction that an over-focus on estimates can bring to the development process. But, whether we like it or not, software development mainly takes place in a business context.
This brings two constraints to bear:
- At the micro level, selection of the best option for any development problem has to consider effort (cost) and duration;
- At the macro level, while I agree that tracking against customer satisfaction is a powerful strength of Agile, at least part of the customer's satisfaction will always relate to on-time, on budget. Hence, some form of estimation is not just management overhead but a development best practice.
So, it's fair to say we at DCG disagree, and we say #YesEstimation! Still, we'd love to hear your thoughts - so please share in the comments below.