Wednesday, November 19, 2008

CMMI and Agile: Why not embrace both!?

The SEI has released a Technical Note that I co-authored with Mike Konrad, Hillel Glazer, Jeff Dalton, and Sandy Shrum. This paper was written almost a year ago and went through a lot of revision and review. We published a public draft at the SEPG conference in March 2008 and held a panel session with all the authors to discuss the content. The paper is finally complete and published. It’s a 48 page document. This is no lightweight look at Agile and CMMI. It’s a very in-depth analysis that looks at why people perceived incompatibilities and why the two communities were not mixing or understanding each other’s point of view. I suspect it will be for some time to come, the definitive statement on Agile + CMMI.

Here are the official announcements…

Announcement 1<?xml:namespace prefix = o /?>

Think that Agile and CMMI are incompatible? You may be wrong according to the authors of a newly released report published by the Software Engineering Institute. The report CMMI and Agile: Why Not Embrace Both! explains why each of these two improvement approaches have been misunderstood by users of the other approach and describes how CMMI and Agile can be used together in ways that benefit from the advantages of both. Take a look at http://www.sei.cmu.edu/publications/documents/08.reports/08tn003.html for a copy of the report.

 

Announcement 2

CMMI® and Agile: Why Not Embrace Both!

Agile development methods and CMMI (Capability Maturity Model® Integration) best practices are often perceived to be at odds with each other. This report clarifies why the discord need not exist and proposes that CMMI and Agile champions work toward deriving benefit from using both and exploit synergies that have the potential to dramatically improve business performance. Get a copy of this groundbreaking report at http://www.sei.cmu.edu/publications/documents/08.reports/08tn003.html.

Technorati tag: SEI, Software+Engineering+Institute, Carnegie+Mellon, David+Anderson, Software+Engineering, CMMI, Agile

Sunday, November 02, 2008

Personal Hedgehog Revisited

[First pusblished at moduscooperandi.com] More than 4 years ago, I riffed off Jim Collins idea of a corporate Hedgehog Concept, with this blog post on Personal Hedgehog Concept. It’s proven to be one of the most popular blog pages on AgileManagement.Net since I started it in August 2003.

The original post used the career of Cameron Barrett as the example. At the time, Cam was pursuing his passion for politics supporting the campaigns of Democratic candidates Wesley Clarke and John Kerry. However, recently I was challenged by Liza Raiser to explain what the Personal Hedgehog Concept means to me.

Actually, I’ve been working on my own hedgehog concept for most of the past 8 years.

First, what am I passionate about? For a long time I’ve been passionate about the underperformance of the software engineering profession and the low rate of success on software development projects. In fact, I was so disgusted with the profession I intended to quit almost 10 years ago. It was thanks to Jeff De Luca, and the original FDD project in Singapore that I regained my enthusiasm for the profession.

So what can I be one of the best in the World at? It’s taken a while, but I started down the path to publishing and what we now call blogging in 1999 at the behest of Peter Coad and Jeff De Luca. 4 years later, Peter was instrumental in assisting me with the publication of Agile Management for Software Engineering. I’ve continued to work at improving my ideas on software engineering process and management of knowledge workers and I’ve continued to work as a practitioner in regular jobs managing software engineers - until recently, when I formed David J Anderson & Associates.

So what changed? Well finally, I was able to realize my Hedgehog Concept. Finally, my skills with software engineering process and management and leadership of knowledge workers were in sufficient demand that they could drive my economic engine. Let’s be under no illusion! There is little to no premium in the market for good management in the software and IT industries. While great individual contributors often become independent contractors and earn high hourly rates, the same does not generally apply to managers. And while employers might be willing to pay a 10%-20% premium for a decent person, often a great manager find him/herself earning far less than the top technical people on the team. This is despite the hard economic evidence that it is management talent that generally constrains the performance of software engineering organizations.

So, for a long time, I’ve known that I had to break out of working as a manager for other people and start my own firm. The question was when? When would the timing be right? Finally in 2008, with a track record that includes successful projects and teams at Sprint, Motorola and Corbis and with a catalog of intellectual property that includes my contributions to FDD, the MSF for CMMI Process Improvement and most recently my contributions to Lean in software development and the innovations with the Kanban method, I finally have sufficient recognition and respect in the industry for it to drive my economic engine.

Along the way, I’ve also resolved my own inner conflicts about whether I had taken the correct career path. I’ve finally come to realize that management and leadership is my real strength and that other things I enjoy are merely hobbies, like painting, art and design, and my synthesis of those talents in user interface and interaction design. It was in fact user interface design that got me started down this road, with my uidesign.net site. Recognizing in myself what I could be World class at, from the things that I can be merely good at, has been the foundation of a new happiness in my life.

So here we are! I’m having the best fun at work since I quit the games industry in the late 1980s and I’m happier than perhaps I’ve been since leaving Singapore in 1999. Finding my Personal Hedgehog Concept has been at the root of that happiness. It’s been a long slog - more than 8 years. A journey of personal discovery. But ultimately it’s been worth it. And now I am excited about the future where I intend to continue innovating in leadership and management of knowledge workers and helping teams deliver superior economic performance.

Are you working on your Personal Hedgehog Concept?

The Tactical Transition

[First published at moduscooperandi.com in a slightly different form]

At David J Anderson & Associates our strategy is to help clients achieve long lasting, institutionalized, enterprise scale agile change. We help them to become what the SEI calls a “high maturity” organization while continuing to use Agile and Lean methods throughout their technology functions. To achieve this we go about changing the organization’s culture. Lasting change takes time. To do it properly can take 9 months to several years. It requires a serious commitment to achieving high maturity - quantitative management, predictability and continuous improvement - from the senior leadership. That’s why many of our clients have C-level titles.

However, not every client needs long term institutional change. So should we turn those other clients away? Perhaps! But not if they truly need us to meet their immediate, tactical goals. I’ve been amazed by the clients we meet who open up the discussion with “I’ve been reading your ... <insert book, article, etc.> and I’ve decided that the solution to our current problems is… <insert methodology FDD, MSF CMMI, Kanban>.”

I’ve been amazed at the demand for FDD and MSF for CMMI Process Improvement. By adding Daniel Vacanti and Eric Willeke who can help us deliver FDD and MSF CMMI projects, we have the skills and experience to respond to demand and provide staff augmentation when necessary.

With these types of clients they have an immediate tactical need. Perhaps they have a mission critical project that is late and over-budget. They need us to dig them out of the hole. So we do that for them. Their need is tactical. They are not concerned about institutionalized change. They are not concerned about resistance to change. They will use positional power and require staff to acquiesce or drop out. Delivery of the project is success for them. And if the process doesn’t survive past the delivery of the project then so be it. Technorati tag: David+Anderson, agile+management, CMMI, FDD, Kanban, MSF

The Relevance of Level 4

[Over the summer, I wrote a number of blog posts for Modus Cooperandi. I’d like those posts to get a wider audience. Starting with this one about the relevance of level 4 organizational maturity…]

CMMI Model Level 4 is often thought of like Nebraska or Kansas - it’s the flyover territory of CMMI. The big offshore outsource companies often think of Level 4 as something that they can skip - jumping from level 3 to level 5. After all, there are only 4 process areas. Two in Level 4 and two in Level 5.

When I was at Microsoft, working on MSF for CMMI Process Improvement, we talked about the future prospect of an enhanced edition that provided full coverage of Level 4 and 5. [The current release has about 80% coverage of Levels 2 and 3, and 20% coverage of Levels 4 and 5.] There was no market demand for a Level 4 solution. Our market research was telling us that there was a market for a Level 3 solution - the one we produced - aimed at the government contracting market in North America and the ISO 9000 compliance market in South America. We also knew that there was a market for a Level 5 template for TFS - mostly aimed at the offshore outsourcing companies. Level 4 just didn’t come in to our plans. It was flyover territory. It seemed no one does Level 4. If you look at the list of CMMI appraised firms, there are very few at Level 4. So why am I suddenly a big advocate of Level 4?

Well, it seems from discussions with clients and potential clients in America and Europe, our clients need to have the equivalent of Level 4 organizational maturity in order to meet their business goals and strategic objectives. They don’t need to be an optimizing organization at Level 5 - that would be icing on the cake. But they do need to be predictable. They want to have strong delivery with low variability. The want to be proactive and drive down cycle times using objective quantitative management. They need all of this to deliver on business goals within the tight financial controls and corporate governance that they now find themselves under. They need to be the equivalent of Level 4.

The real problem is that typical Agile methods can only take them to Level 3. So Agile isn’t enough. That’s where we come in. Our experience in creating cultures that drive towards high maturity (Levels 4 and 5) while implementing Lean and Agile techniques is still fairly unique. We help clients reconfigure their organizational culture to enable a high maturity organization to emerge while still gaining all the benefits of Agile and Lean methods.

The aim is to generate a clutch of Level 4 equivalent organizations. Clients who can estimate projects and iterations and deliver results with a low degree of variation from the original estimate. Firms who use predictive methods and leading indicators to learn and adapt quicker than those simply using retrospective methods and lagging indicators. And businesses who are led by objectivity and have left superstition and subjectivity behind in their organizational past.

CMMI Model Level 4 has real business relevance. Business that achieve it will achieve their business goals, hit their numbers and delight customers, shareholders and employees. Getting to Level 5 will allow a firm to become ever more competitive and to dominate their market. But for many firms the need to achieve the equivalent of Level 4 maturity is a business imperative, now! Anything less will leave all stakeholders dissatisfied. Technorati tag: David+Anderson, agile+management, CMMI