Agile Development Blog

This week both Jean and I delivered talks on the Agile organization at Agile 2010 in Orlando. Whether you were able to attend one, both or neither, this post shares the handouts and materials that we used in the talks.

If you attended, please provide comments on what you liked, were puzzled by and might change in the future.

Jean’s work was a three-hour tutorial on learning models for managing the Agile organization.   She ran three exercises and provided a bibliography of books/resources that we have used here at Rally:

Jean in action at Agile 2010

• True North Example

• True North Exercise

• Strategy A3 Template

• ORID Exercise

• Personal and Corporate Visioning Exercise

• Sample Quarterly Agenda

• Bibliography

In addition to Jean’s talk, I presented an experience report on our use of Plan-Do-Check-Act (PDCA) at Rally.  This report tells a story of our evolution of strategy execution from Gazelles/Scrum to Lean/Agile.

We hope these resources provide you with ideas for scaling your own Agile efforts beyond their current levels.  Again, please comment on the blog with what you got from the materials or the talks.  We want to hear from you on this topic.

Ryan Martens is a tomato grower, founding board member of the Entrepreneurs Foundation of Colorado, and CTO at Rally Software Development.

On April 21 in Atlanta, the Lean Software and Systems Consortium will come together for its second US conference.  Last year’s event in Miami was “amazing” according to Jean.  So this year, Rally is exhibiting, I am speaking while Jean and Aaron are running the open space on Friday.  The price per attendee goes up by $250 on March 31st, so if you do intend to go, REGISTER now.

At our booth, Rally will be demonstrating its product support for highly-visible Kanban, WIP/Cumulative Flow reports, and cycle-time  metrics. Join Alan Atlas, Jean Tabaka, Aaron Sanders and Craig Langenfeld in our booth.

I will be presenting an experience report titled: PDCA: Beyond Simple Inspect and Adapt. On spring break this week, I’ve been thinking more about the details of my talk. Here is my abstract and outline for those of you who might consider attending:

Lean and Kanban focus on practices of continuous flow of product delivery. Plan-Do-Check-Act (PDCA) is a Lean discipline that moves beyond inspect and adapt of Agile team-level processes.  At a corporate level, PDCA provides guidance for strategy as well as problem-solving work. In 2009, I led Rally’s move to PDCA for the company’s strategy process at both the annual and quarterly levels.  My primary guide was Pascal Dennis’ “Getting the Right Things Done”. In this experience report, I share Rally’s PDCA first year of adoption: where we started, how this impacted our corporate behaviors, and where we are now. I want to share Rally’s story to compel participants to embrace PDCA and get good at it. I ask each participant to come with its organization’s #1 goal and success criteria. I will close with a planning A3 exercise from Pascal’s book .

Outline

1. What brought me here—background on why I am passionate about sharing my organization’s overall Lean story including the addition of PDCA, A3’s and concurrent set-based development.  This talk focuses on PDCA as the critical step in increasing structure and discipline in strategy execution.
2. Point of View – Use PDCA to move your planning horizon out and as the principle governing mechanism for organizations in continuous flow.
3. Benefits — Mature your strategic planning and execution environment to handle the complexity of increasing speed, agility and scale and to gain alignment, pull and innovation.
4. Where we were and what was not working
   1. The context at Rally was based on a couple of key concepts:
      1. Core Values, Core Purpose, Sandbox and BHAG from Jim Collins
      2. 3 to 5 Quarterly Rocks, success criteria and Scoreboards from Gazelles
      3. Rock team structures  – cross departmental and story based
      4. Facilitated, highly collaborative cross-departmental meeting of 30+ managers and above
   2. What we noticed
      1. Rock work as a second or third job
      2. Wisdom of the Crowds to help fix over-reaction and group think
      3. Too much on the fly, not enough backlog grooming
      4. Highly critical, non-cross departmental initiatives were de-prioritized
      5. ORID process added to keep from jumping too solutions, but the data was not visible enough
5. What we decided to do about this:
   1. Explanation of PDCA — A brief overview of PDCA in general and then specifically what I used as guidance from the Pascal Dennis book, “Getting the Right Things Done: A Leader’s Guide to Planning and Execution”.
   2. Our initial experiments with A3 process the year prior — Working with our Ops team and product marketing teams on problem solving using real data
   3. First quarter — How we kick-started Rally’s company-wide adoption of PDCA . I describe our “Mountain Team” and their transitional role.
      1. Defining Rally’s True North
      2. Creating our second level tree with current and needed metrics
      3. Socializing these throughout the company seeking feedback in anticipation of our annual and quarterly planning
      4. Started new experiments based on quarterly planning decisions
   4. Next Quarter – Review new experiments, discussed learning and drive A3’s into the planning process
   5. Mid-course adjustment by Mountain team, in middle of the quarter – What we discovered working and not working
      1. The rocks were all dependent on each other.
      2. Had to run Rock of Rock team meetings to steer to a final solution
      3. Coordinated release planning would have
   6. Final quarter – We worked to expand the plan. We took the Mountain team’s True North and feedback to drive our PDCA for Rally’s Annual Corporate planning by:
      1. 1. Taking company-wide feedback into our Annual planning to collaboratively drive cross-department A3 creation around each branch of the tree
         2. Mountain Team retrospective over the course of year 1 that helped create a planning rock team.  The Mountain team’s role as a transition team ended.
   7. Year 2 – Doubling down our efforts to go from amateurs to intermediates —Changing our process to institutionalize A3’s and PDCA as our strategy execution approach:
      1. Quarterly rocks moved to a world of pre-defined from developed on the fly
      2. Quarterly planning moved from ad-hoc based on yesterday’s weather to more programmed based on True North and meeting the target metrics
      3. Strategic planning worked to validate annual True North in the context of long-term planning, shared vision development, cross-boundary collaboration and larger systems
6. What we learned and what you should do about it
   1. The cycle of adoption is a year, quarterly cycles work to improve the process, but it is hard to make leaps on a quarterly basis.
      1. Year 0 – Introduce Lean thinking (A3 in our case)
         
      2. Year 1 – Introduce PDCA (Novice)
      3. Year 2 – Invest or abandon (Your choice)
         1. A3 is now the language for problem solving
         2. Making sure we are solving the right problem (aka slowing down to speed up)
   2. Do not have a overall guidance team steering the continued PDCA process – it is owned by the “team”
   3. Putting pressure on the organization to get more clear about our economic models to mature from “Theory-based decision making toward the right solution,” Now “Data-driven decision making toward the right problem”
   4. Where to start – The Strategy A3 an exercise
   5. What is next? – We call it the Innovative or Lean Organization. Seeing large systems, collaborating across boundaries and creating your reality.
      
7. Point of View – Use PDCA to move your planning horizon out and as the principle governing mechanism for organizations in continuous flow.
8. Call to Action – Introduce the language of A3’s through problem solving or Strategy A3’s
9. Benefits — Help build a company of problem solvers to focus your efforts on the critical few things.
10. Where I hope you go with this: Great companies build great software, great experiences and work on creating win/win scenarios.
    

Last week, I ran our quarterly planning session and was very positively surprised with the results of testing Toyota’s A3 method in the meeting. Before the meeting, I was convinced that I was trying to “push” too much, but instead the team sucked them up.

Breakout groups at Rally quarterly planning

If you don’t know what an A3 or problem sheet is, I recommend starting with John Shook’s article in the MIT Sloan Management Review.

What is an A3?

John says in his article, “At Toyota, there exists a way to solve problems that generates knowledge and helps people doing the work learn how to learn. Company managers use a tool called the A3 (named after the international paper size on which it fits) as a key tactic in sharing a deeper method of thinking that lies at the heart of Toyota’s sustained success.”

An A3 is a problem sheet in which the owner attempts to:

1. “Establish the business context and importance of a specific problem or issue

2. Describe the current conditions of the problem

3. Identify the desired outcome

4. Analyze the situation to establish causality

5. Propose countermeasures

6. Prescribe an action plan for getting it done

7. Map out the follow-up process”

There is much more information on this topic on the Lean.org blog.

Introducing A3’s to the Rally team

At Rally, we started using A3’s to diagnose issues due to growth and market maturity last summer.  I hired a trainer and friend from my US WEST days Jay Arthur and focused our first effort on product marketing and product management.  In that three-month effort, we had very positive results from working 4 simultaneous A3’s and built a well prioritized backlog of countermeasures.  We knocked off the highly feasible and highly effective ones that next quarter, but we continue to work off that list today.  After product marketing and product management, we brought the technique to operations team, too.

Given this foundation and 25% of our team trained on A3’s, we brought them into quarterly planning as part of our rock prioritization process. Jean and others warned me that it was a ton to grasp in fully packed day of checking the last quarters results and adapting for the next quarter. For that reason, I was prepared to fall back to our old way of working and prioritizing.

Homework is good?

What happened was even more than I could expect. Instead of throwing out this process, the team embraced it and even worked through breaks. Using A3’s to structure your thinking requires discipline and critical thinking.  I guess there was more pull waiting in the room than I expected; maybe it was the required reading of John’s article that helped?  My conclusion – homework is good?

As Jean and I continue to explore what makes an Agile organization, I am curious as to what others are using in their business planning processes to prioritize their quarterly or annual efforts.

About the Author: Ryan Martens is an avid outdoorsman, founding board member of the EFCO, and Founder and CTO at Rally Software Development. Subscribe today to get free updates by email or RSS.