The more detailed we made our plans, the longer our cycle times became.
—Don Reinertsen
In this chapter, we discuss the next ‘critical move’ in the SAFe Implementation Roadmap: Create the implementation plan.
It’s a big deal to implement organizational change of this magnitude, and some strategizing and planning help. But Reinertsen’s quote reminds us not to overthink the problem. It’s far better to plan a bit, execute a bit, and learn a bit. Then repeat. In other words, we need to take an Agile, incremental approach to implementation—just as we do with solution development.
We’ll do that by picking one Value Stream and one Agile Release Train (ART) to serve as the vehicles for our journey.
In the last chapter, Identifying Value Streams and ARTs, we described a process typically executed in one or more workshops, where the key enterprise stakeholders gathered to identify the flow of value through the organization. By ‘stakeholders,’ we mean SAFe Program Consultants (SPCs), members of the Lean-Agile Center of Excellence (LACE), newly trained Lean-Agile leaders, and other essential team members.
Using their new knowledge from SAFe training, the Lean-Agile mindset and principles, and the value stream workshop, the enterprise stakeholders start to identify strategies for implementing this new way of working. That brings us to the next step, creating the implementation plan. This is where the rubber meets the road in a SAFe implementation.
Until now, it’s been all talk. The next move, however, requires real and tangible changes to individual and organizational behavior. Specifically, creating the plan involves three activities:
Pick the first value stream
Select the first ART
Create a preliminary plan for additional ARTs and value streams
Each of the development value streams identified in the prior step is a candidate for the new way of working. A large enterprise offers a lot of opportunities for improvement, and while there’s no one right way to begin, for many companies the next smart move is to just pick one target. After all, it’s likely this significant change is untested in this business’s environment. Picking one target allows the newly trained SPCs and leaders to focus their full attention and resources on a specific opportunity.
Once a value stream is selected, some additional analysis is required to further define the development value stream boundaries, people, deliverables, potential ARTs, and other parameters. To assist, we offer Figure 1, a Value Stream Canvas that stakeholders can use to capture their emerging understanding [1].
Filling in the fields typically requires some homework. It calls for an understanding of how things work now, as well as how they’re intended to work in the future. As highlighted in the bottom section of Figure 1, some additional analysis is needed to define prospective ARTs and governance in a Solution Train (multi-ART value stream).
As the value streams are identified, it becomes obvious that development value streams often cross many boundaries, as illustrated in Figure 2.
In turn, many of the development value streams—and, as a result, the ARTs—will be geographically distributed rather than collocated. While that might complicate things, it’s a reality, and it doesn’t change the basic operating model. ARTs use a variety of techniques to mitigate this challenge, including multi-location face-to-face Program Increment (PI) Planning. Although we’ve observed that a SAFe implementation provides opportunities to advance more geographically collocated development practices, companies simply have to start with their current situation.
Once the first value stream has been identified, it’s time to create the initial ‘short-term win’ by focusing on the first ART. That will yield institutional knowledge that can be applied to other ARTs. In some cases, the first value stream is also the first ART, and no other decisions are needed. In larger value streams, however, the next step will require the active support of many more leaders and other stakeholders in that value stream. Many organizations decide to look for a first, ‘opportunistic’ ART, one that can be found at the intersection of the factors illustrated in Figure 3.
The target for the first ART is often one that best meets the following criteria:
Leadership support – Some senior leaders may have already been trained in SAFe and will be anxious to put their training to work. Moreover, it’s likely that many of these leaders have previous experience with Agile development.
Clear products or solutions – SAFe is most easily applied to a clear and tangible solution, something the company sells directly or values highly.
Collaborating teams – Somewhere in the enterprise, there are already teams collaborating on building a larger solution. Some may be Agile, some not. But given the business’s current challenges, the teams may be ready to embrace this change.
Significant challenge or opportunity – Change is hard. The smart enterprise selects a subject that is truly worthy of its effort, ideally a large existing challenge or a new opportunity. Creating a short-term win there will produce immediate benefits and facilitate faster and broader adoption.
Once this ART is selected, the enterprise is nearly ready to move forward.
Before we move on to launching that first ART, we note that it’s likely that a broader implementation plan may already be forming. Although it’s still early in the process, strategies for rolling out additional ARTs and for launching additional value streams may be starting to take shape. In short, change is starting to happen, and the signs are everywhere:
The new vision is being communicated around the company
Principal stakeholders are aligning
Something big is in the air, and people are catching on
As we described in Creating a Lean-Agile Center of Excellence, the LACE and various SPCs and leaders typically guide the transformation using Agile and SAFe as their operating framework. In accordance with SAFe practices, the LACE holds internal PI Planning and invites other stakeholders, such as Business Owners, to help further define the implementation strategy. One natural output would be a PI Roadmap for the implementation, as illustrated in Figure 4. The roadmap further details the plan and a PI cadence for implementation.
Before committing to the roadmap, it’s probably a good idea for stakeholders to reflect on the existing culture and the ‘how’ of the larger implementation strategy. Yes, it’s a committed change initiative, and that means it’s a largely centralized decision (see the chapter Principle #9: Decentralize decision-making). The change is not optional, but how it is received depends on many factors. Oftentimes, mandated change can be uninspiring to those who are on the receiving end of the decision. In that case, you may want to try the approach described in Yuval Yeret’s chapter, Invitation-Based SAFe Implementation, which describes how to create a more collaborative organizational change effort.
Don’t be too concerned about getting your strategy perfect right at the start. Any such preliminary plan is only the current hypothesis and will be incrementally improved as your implementation evolves. We’ll revisit the larger planning cycle later in Launching More ARTs and Value Streams.
In any case, with the first value stream selected and one or more initial ARTs defined, it’s time to move on to the practical matter of implementing the first ART. That is the subject of the next chapter, Prepare for ART Launch.
LEARN MORE
[1] Thanks to SPCT Mark Richards for contributing the Value Stream Canvas concept.
[2] Martin, Karen, and Mike Osterling. Value Stream Mapping. McGraw-Hill, 2014.
[3] Knaster, Richard, and Dean Leffingwell. SAFe Distilled: Applying the Scaled Agile Framework for Lean Software and Systems Engineering. Addison-Wesley, 2017.
