Strategy As Learning

Continuous improvement methodologies abound (e.g., The Model for Improvement, Results Oriented Cycles of Inquiry, Carnegie Foundation for the Advancement of Teaching’s approach to improvement science, Kaizen, Six Sigma). While the particularities of these methodologies differ, they’re all designed to support stakeholders in answering three core questions:^{2Adapted from Langley, G. J., Moen, R. D., Nolan, K. M., Nolan, T. W., Norman, C. L., & Provost, L. P. (2009). The improvement guide: A practical approach to enhancing organizational performance (2nd ed.). John Wiley & Sons.}

• What are we trying to do?
  
  
• Is what we are doing working as expected?
  
  
• What changes can we make to improve what we’re doing?

When the answer to the second question is “no,” each of these methods use the following steps to answer the third question.

1. Select areas of focus: As a matter of course, all members of the learning community routinely examine data to determine whether current practice in the system is leading to desired outcomes. When they do not meet expected targets, stakeholders identify high-leverage areas for improvement.
   
   
2. Conduct root cause analysis: Stakeholders collaboratively explore the underlying causes of gaps between expected outcomes and results. A diverse cross-section of community members—especially those closest to or affected by the problem—conducts causal analysis so that the challenge can be examined from different perspectives. Tools like the 5 Whys protocol and Fishbone Diagram are often used.
   
   
3. Develop a theory of improvement: Once root cause analysis is complete, stakeholders generate a range of interventions (i.e., “change ideas”) predicted to improve outcomes, drawing from both their own experience and that of the field (e.g., empirical research). These ideas are captured in a shared theory that articulates the anticipated causal links between the proposed interventions and desired outcomes, typically using a tool like a driver diagram, a theory of action, or an operationalized theory of action.
   
   
4. Conduct short-cycle testing: Stakeholders then test proposed interventions, often using rapid-cycle testing processes like the Plan, Do, Study, Act (PDSA) cycle. The goal of testing is to rigorously vet ideas across various contexts to determine whether they are effective and suitable for scale.
   
   
   1. Plan: Stakeholders determine the scale of the test and develop an action plan, including: where, for how long, and at what scale testing will occur; who will be involved; and what data will be collected to measure outcomes. At this stage, stakeholders capture their predictions about the impact of the intervention idea so they can assess the gap between anticipated and actual results.
      
   2. Do: Testers implement the intervention, gathering data, tracking results, and making note of any deviations from the testing plan.
      
   3. Study: Stakeholders come together to compare actual results with predictions. By collaboratively analyzing data—often using visualizations like frequency tables, line graphs or run charts, bar graphs, scatterplots, and pie charts—the team gleans insight for the next problem-solving cycle and the overarching strategy. 
      
   4. Act: Stakeholders decide whether to abandon, adapt, or adopt the intervention for the next round of short-cycle testing. In instances where the intervention has achieved predicted outcomes across a diversity of contexts, leaders may opt to scale the intervention by integrating it into system-level strategy.
      
      
5. Scale up and spread improvements: Leaders integrate the intervention into system-level strategy and the daily practice of the organization, adapting application as needed across contexts.

Equity-advancing continuous improvement methodologies embed liberatory design principles throughout the process so that improvement efforts help stakeholders embrace complexity, build relational trust, heal, and transform power.^{3Anaissie, T., Cary, V., Clifford, D., Malarkey, T. & Wise, S. (2021). Liberatory design. http://www.liberatorydesign.com.}

Learning leaders design their systems to blur what Follet calls the “sharp-line between strategy and execution.”⁹Ansell, C. (2011). Pragmatist democracy: Evolutionary learning as public philosophy. Oxford University Press.  They create ongoing feedback loops between strategy and implementation, understanding that local challenges are the grist for system-level improvement. In practice, this dynamic disrupts traditional concepts of where expertise lies—in the many doers rather than the few directors—and enables the system to adapt system-level strategy more quickly. 

Learning Leaders:

• Design leadership roles that straddle strategy and implementation.
  
  Learning leaders lead from the ground. They embed themselves in ground-level work and team-learning processes as often as possible so they can understand implementation, experience challenges directly, and learn at pace with stakeholders. 
  
  For example, in the High Tech High’s CARPE College Access Network, hub leaders coach school-based improvement teams in addition to strategic planning responsibilities. This structure differs from that of many networks, where coaches sit on teams that report to, but do not include, hub leadership. The CARPE network’s hybrid structure allows hub leaders to anticipate and understand emergent challenges and quickly adapt their leadership approach and network strategy in response. 
  

• Include representative ground-level stakeholders in strategy and decision-making spaces.
  
  Learning leaders ensure that ground-level actors participate in developing system-level strategy. Formal strategy teams include a diverse cross-section of the organization and its stakeholders. Equally as important, the daily application of continuous improvement principles in interconnected teams means that daily discoveries inform overarching strategy and decision-making.
  
  For example, Partners in School Innovation embeds ground-level voice into formal strategic decision-making structures by establishing rotating roles filled by elected ground-level staff on several national planning teams. Further, ground-level insights gleaned from the application of the organization’s continuous improvement methodology, ROCI, are continually funneled to strategy leads through the organization’s robust shared learning spaces.
  

• Build strong andon structures.
  
  To ensure daily practice informs organizational strategy and decision-making, learning leaders build strong alarm systems that let anyone in the system flag problems and opportunities for improvement—large and small—as they occur. Toyota epitomizes this practice. When line workers bump into a serious issue, they pull the andon cord—an alarm that halts the production line and prompts managers to work in partnership with line workers to solve the problem before the vehicle is completed.
  
  Effective education learning leaders establish andon structures that similarly allow all stakeholders, from ground-level actors to leaders, to quickly flag misalignment between strategy, expectations, and experience during implementation. The goal of this work is not to determine good versus bad practice but to reveal where there are misunderstandings in strategy that should be investigated.¹⁰Spear, S. J. (2009). The high-velocity edge: How market leaders leverage operational excellence to beat the competition (2nd ed). McGraw-Hill.
  
  Andon mechanisms are intuitive, embedded into practice, and consistent across the system. The technicalities of andon processes look different in each system, but at the highest level, they begin with frequent, cross-functional learning spaces (e.g., huddles) that enable stakeholders to raise concerns and instances where the implementation of strategy is not producing expected results. In many cases, actors present in the learning space are equipped to adapt ground-level strategy and address the challenge immediately. Intermediaries—actors tasked with keeping knowledge, support, and resources flowing from one part of the system to another—efficiently funnel those “alarms” and suggest adaptations to those ultimately responsible for updating system strategy (e.g., the strategic planning team, school leaders, a network hub). In cases where additional support is needed, intermediaries connect with leaders to mobilize higher-level action.
  
  In small systems (e.g., a team, a small organization), these structures may be relatively informal. In larger organizations, leaders may need to develop more technical processes to ensure knowledge is being effectively funneled from local spaces. Whatever the process looks like, most critical is that all stakeholders are empowered to participate—and that leaders will listen when they do.
  

• Create cross-functional, goal-focused teams.
  
  Learning leaders build cross-functional, goal-oriented teams and task them with end-to-end responsibility for those objectives. Learning leaders select a diverse set of representatives who bring varying expertise and access to local sites of implementation. Collaboratively, the team addresses boundary-crossing challenges, moving beyond siloed tinkering and toward systems-level solutions.
  
  Once a team accomplishes its immediate task, learning leaders collaborate with team members to reflect on learning, set new goals, adjust team membership as needed, and launch new learning and improvement work. In some cases, the team may no longer be necessary. Over time, this type of agility means that, on an org chart, the system’s structure looks something like an evolving organism.¹¹Aghina, W., Ahlback, K., De Smet, A., Lackey, G., Lurie, M., Murarka, M., & Handscomb, C. (2018, January 22). The five trademarks of agile organizations. McKinsey & Company. https://www.mckinsey.com/business-functions/people-and-organizational-performance/our-insights/the-five-trademarks-of-agile-organizations
  
  This model relies on competent, high-capacity teams. Learning leaders develop and document clear roles, expectations, and guidelines for teams (see an example from High Tech High’s CARPE College Access Network here) and appoint or help select a team lead. Learning leaders often find it valuable to codevelop standards for team excellence —as New Visions for Public Schools has done for its College Ready NSI—and share those standards directly with teams and those supporting capacity building (e.g., coaches, team leads). Learning leaders leverage these standards as a way to monitor their own leadership practice: If teams are falling short, leaders interrogate where system strategy, design, or their own coaching needs to be adapted.

Learning leaders build, prioritize, and protect time and space for collaborative learning, reflection, and improvement routines, adjusting other applied work accordingly. In describing her experience working at Partners in School Innovation, ground-level implementer and regional leader Uchenna Lewis said, “There’s a spaciousness created for learning that I had never experienced as an adult.”

To create a similar “spaciousness for learning” in their systems, learning leaders set shared learning goals and roles, establish a regular cadence for learning routines, and build strong connections between individuals, teams, clusters, functional responsibilities, and content areas so the system functions as a holistic learning engine rather than as a series of discrete, disconnected learning workstreams. 

Learning Leaders:

• Co-develop an organizational learning agenda.
  
  In many education systems, ground level workers (e.g., teachers, coaches) are accustomed to “learning” spaces designed as one-directional “sit-and-get” professional development. But to leverage the power of the collective and accelerate change, learning leaders create learning spaces that are the exact opposite: collaborative, cross-functional, action-oriented, and aligned to shared learning goals.
  
  To set the conditions for that type of work, learning leaders set a strong learning arc and vision for the organization, using the system’s shared strategy, learning methodology and tools, and feedback from ground-level stakeholders to prioritize and sync improvement efforts around shared short- and long-term goals.
  
  
• Create a clear, consistent structure and purpose for each learning space.
  
  To create a sense of stability in a dynamic, constantly learning system, learning leaders establish a consistent cadence, style, and set of objectives for each learning space. Common structures include:
  
  
  • Huddles.¹²Institute for Healthcare Improvement. (2012). Huddles. http://www.ihi.org/resources/Pages/Tools/Huddles.aspx These brief, frequent, goal-oriented meetings support reflection on short-term goals and challenges, especially during the implementation of short-cycle testing.¹³Provost, S. M., Lanham, H. J., Leykum, L. K., McDaniel Jr., R. R., & Pugh, J. (2015). Health care huddles: managing complexity to achieve high reliability. Health Care Management Review, 40(1), 2–12. doi: 10.1097/HMR.0000000000000009 In huddles, a group of stakeholders checks in, raising concerns, sharing learnings about recent practice, and noting where expectations and results differ during strategy implementation. The huddle structure may be used within a single team (e.g., within a seventh-grade teaching team, within a regional service team) or across several teams (e.g., across seventh- and eighth-grade teaching teams, across regional service teams).
    
    Effective huddles are structurally consistent—they’re held at the same time and place and use a standard agenda and set of roles. They are organized around several simple questions, such as: What did I learn yesterday? What challenges are top of mind? What can we do differently tomorrow to address those challenges? Critically, the outcomes and immediate action items of the meeting are recorded on a running notes document.
    
    In general, the huddle team is able to address most challenges raised through adjustments to daily practice, but when larger issues emerge, the huddle leader escalates the challenge to appropriate system leaders. In practice, huddles serve as an andon routine—a way for ground-level stakeholders to quickly raise implementation challenges and contribute to solving them. 
    
    
  • Data reflection spaces. These meetings afford stakeholders the opportunity to come together to reflect on and interrogate data to surface and interrogate problems of practice, identify bright spots, and understand the success of interventions. These meetings supplement, rather than replace, the use of data in other spaces (e.g., data reflection in huddles). They are particularly useful in systems in which capacity for data analysis is low and where dedicated time and space for capacity building and growth around analysis may be helpful.
    
    
  • Stakeholder-specific or role-alike affinity spaces. These opportunities support system actors (e.g., principals, coaches, teachers, families) in comparing notes and consolidating learning. They are particularly useful in systems where power differentials or accountability structures make it difficult for ground-level stakeholders to express uncertainty or share failures in front of supervisors or system leaders. These learning spaces need not be limited to stakeholders within your system—system leaders (e.g., superintendents), for example, often find it beneficial to establish ongoing learning relationships with leaders in similar systems.
    
    
  • Individual learning spaces. Recognizing that collaborative learning requires individual reflection, learning leaders protect and help organize individual learning space for those they lead. In practice, this may look like demarcating meeting-free times each week, setting learning goals with supervisees, or simply relaxing top-down time management, trusting those you lead to make choices about where and when they need to take time for individual reflection, learning consolidation, or reading.
    
    

Perhaps most critical to ensuring that these structures produce strong learning is respect—nothing devalues a learning space faster than insufficient planning and lackluster facilitation from learning leaders. Effective leaders collect consistent feedback; shift the structure, facilitation style, and content of the spaces in response; and are explicit with participants about how they have integrated feedback.

Learning leaders apply the same improvement standards to themselves that they expect of others. They have an approach for designing, guiding, and participating in system learning and improvement and constantly test their practice to achieve desired results. Once their approach is captured in a Theory of Leadership, leaders transparently and methodically improve their own leadership approach, the same way they improve their organization, team, or network strategy. Being explicit helps guard against a natural inclination to adjust recollection of anticipated outcomes once results are in (e.g., “I always thought this would play out like this.”).

Living out this approach requires a number of “soft” behaviors: humility, vulnerability, a willingness to constantly learn, and an openness about when and about how you’ve fallen short or where you do not have answers. Underlying this practice is a simple question: “How can I do better today than I did yesterday?” 

Learning leaders create and highlight a sense of collective identity by working with others to recognize the qualities of their system that differentiates it from others. They capture those norms and values, ideally in a short and quippy list, and keep them front and center, posting them publicly in shared workspaces and grounding any collaborative work by returning to them as norms. Being explicit about these values is critical—keeping them tacit makes it more difficult to communicate, measure, refine, and apply them consistently across the system. 

Once these norms and values are defined, learning leaders regularly assess their own leadership practice and system design against them and help those they lead do the same. At High Tech High, for example, “sparking joy” is a value that leaders and community members say is critical to effective learning. Leaders and staff across the organization constantly assess their design decisions and behaviors against that value : “Is this activity we’re designing for the network convening going to spark joy? Is how I’m coaching teams sparking joy?” When behaviors or system design is not living up to or furthering those values, leaders respond and help others do the same.  

Over time, these values become integrated into the fabric of the community through institutional myths and stories.¹⁷Bolman, L. G., & Deal, T. E. (2017). Reframing organizations: Artistry, choice, and leadership (6th ed.). John Wiley & Sons.

Learning leaders leverage markers of community membership to bring social cohesion and a sense of belonging to diverse learning cohorts. Learning leaders may elect to pick a unique, thematically compelling name for the system, develop a logo and a website, reinforce a community-specific lexicon, and invest in branded clothing and other physical signals of membership. They enact rituals and traditions that reflect shared values—for example, closing each quarterly full-group learning session with storytelling, applauding outside a retiring faculty member’s final class, or inviting new staff to dine one-on-one with organizational leaders. And they identify and model shared “ways of being” that distinguish members of the system.

Intentionality around these efforts is especially important in new and geographically dispersed systems (e.g., multistate networks, national organizations) where stakeholders have had fewer opportunities to develop organic community ties and symbols. The goal is not to exclude outsiders—in fact, systems that inspire transformative change are always ready to welcome new members¹⁸McCannon, J. & Margiotta, B.K. (2015, January 21). Inside the command center. Stanford Social Innovation Review. https://ssir.org/articles/entry/inside_the_command_center – but to establish a social glue to support collaborative learning work. 

“A capacity to translate private feelings into knowledge and then public action, when warranted, has been an engine of every movement for social change.”

Parker J. Palmer¹⁹Palmer, P. J. (2007). A new professional: The aims of education revisited. Change: The Magazine of Higher Learning, 39(6), 6–13.

Learning leaders recognize feelings, experience, and stories as critical data for transformative systems change.²⁰Safir, S., & Dugan, J. (2021). Street data: A next-generation model for equity, pedagogy, and school transformation. Corwin. When staff feel overwhelmed, students feel unsupported, and families feel excluded, these are not just personal challenges. They are indicators of system breakdowns that can reveal specific, actionable areas for intervention. 

Because this perspective is antithetical to how many professionals have been taught to behave, learning leaders train stakeholders to “stay close”²¹Palmer, P. J. (2007). A new professional: The aims of education revisited. Change: The Magazine of Higher Learning, 39(6), 6–13. to feelings and translate them into actionable knowledge, asking questions like: How did you feel today? What, if anything, do these emotions tell you about your practice? System design? What would need to change to shift how you’re feeling? 

Learning leaders create communities where stakeholders feel comfortable being vulnerable and sharing the knowledge gleaned from feelings to incite action.