Saturday, November 12, 2011

Before the First Scrum: Management

You're a manager, a process person, developer, or product manager and your business success has been OK but you think it could be [much] better if your development strategy switched from waterfall to Scrum. How will you begin the transition? This entry will provide some advice on what needs to happen to convince a reluctant management team of the advantages of adopting the Scrum framework.

First Steps For Management

You've heard it over and over again that management needs to support the philosophy and practical applications of Scrum in order for it to work and be successful.  In my situation, the R&D Manager and Product Management Manager were both experienced with Agile and were eager to experiment  with Scrum to determine if the results would be better than the waterfall methodology then currently followed. The key point was that it was an experiment and not a full commitment; Agile would have to prove itself and that proof would be measured in the quality of the product and the speed at which the delivery was made. The business itself was not fussed on the means but were very keen on the results. Having the two department managers who are directly responsible for new product development supporting Scrum was, I believe, crucial for our ultimate success of adopting Scrum.

But what if there's no management support or worst, opposition to change i.e., adopting Scrum.

I worked at a defence company in the early 80's and the company's policy was no one would be hired as an engineer unless they had a 4 year college degree. My manager wanted to bring in someone as an engineer but without the necessary 4 year degree. The manager ended up writing a memo (remember those?) and eventually got the person in as an engineer. He later told me that the length of the memo was directly related to the hurdles before him and to get the person hired as an engineer required an eight page memo. The point of course is that the bigger the obstacle blocking your goal, the bigger the effort in making your case to reach that goal. To convince management that Scrum is a better alternative to current development methodologies, usually waterfall, you need to make the business case that Scrum can improve both quality and productivity; get better products out to the paying customers more quickly.

You must sell the idea of Scrum to management starting the first day and every day following without letup. Management needs to learn that through Scrum, there's higher productivity, greater innovation, quick reaction to customer or competitive changes, high visibility to progress (or lack of progress), higher product quality, and more potential evaluation deliveries to customers after each sprint. At a training session given by Jeff Sutherland, he stated that even if you do nothing else but have a daily scrum meeting, you should see a 20-30% increase of productivity. Jeff Sutherland has lots of practical information and statistics that can be used to help convince your management that Scrum can help your business. There are also many, many resources on the web that have anecdotes and other stories that can help bolster your arguments for Scrum adoption. You will need to understand what exactly the manager's opposition to Scrum is and address those issues directly.

I also think that trialling Scrum is a better approach than trying to completely change the way things are done all at once. By trialling, management is more likely to allow a single team to adopt Scrum without, in their minds, risking everything. When my company first trialled Scrum, only one team was formed. After the team was able to deliver a new product in only three months, a feat that had never happened before with waterfall, the CEO asked why all of development wasn't doing Scrum. 

If you have any stories on how you succeeded in winning over management to use Scrum, please let me hear from you.