Rickerism

Since 1997, I have been building software in what would now be called an agile development methodology. I am an avid believer in fix time box delivery cycles of 90 days. Half the people argue that it should be longer and the other half argue that it should be shorter. To me that means 90 days is just right. Besides, companies have a biorhythm of quarters. Delivering a product release every 90 days enables the software team to meet the biorhythm of the rest of the company as well as the biorhythm of the customers.

The deadline is the deadline. Once the deadline is set, it is inviolable. Usually the deadline is set a year ahead. You scope the product functionality to meet the timespan. You do not increase the timespan to meet the scope of functionality.

There is always one more thing we could put in the product. There is always one more bug. There is always one more critical demand from a customer. There is always some salesman pleading that if he only had this one feature he could close a million dollar deal.

If you let the deadline slip once, then it will slip again and again and again. You will never ship product unless we keep deadlines inviolable.

The issue is not the deadline. The issue is prioritizing and staffing. It is far easier to move the deadline than it is to make the hard decisions of prioritizing what gets in the release and what gets pushed out 90 days. When you prioritize, someone will inevitably get told “no.” If no one is hearing “no”, then you are not prioritizing.

If everything absolutely must be in the release (which I doubt), then the issue is not the deadline but staffing. If management believes that the scope cannot change, then management should have made the investment to hire the number of developers necessary to meet the scope.

As they say in Texas, crap rolls down hill. Slipping deadlines is letting the crap roll down hill onto the staff, making the staff take the brunt of a lack of decisiveness. If you want a successful software company, your management team better keep the crap off the development team. Your management team better make the hard decisions and own up to the responsibility for those decisions.