Driver C

Measurement

Use observation and information to turn doing into learning

The Playbook helps you develop a customized Theory of Leadership that outlines your role as a leader, in designing, guiding, and acting to advance equity at scale.

This section describes how to use practical, integrated measurement processes that close the gap between doing and learning. Once you’ve reviewed this section, review the other three drivers. Together, they constitute a Leading Through Learning approach that you can make your own through your Theory of Leadership. 

To achieve equity at scale, learning leaders use measurement to uncover opportunities—large and small—to improve services so that they meet the needs of every young person, family, and community. Measurement makes visible how, for whom, and under what conditions your strategies and approaches work, or if they work at all. Integrated fully and seamlessly into everyone’s standard work, measurement continually gauges whether you deliver equitable services and processes throughout the system. 

Measurement becomes your leadership competency and accelerator of service quality and delivery. It is no longer a technical, isolated activity, tethered only to the latest discrete problem-solving effort. Rather, it is an outgrowth and a manifestation of strategy. You serve as an active processor, gatherer, and generator of measures, and you give others the capacity to do the same. You use measurement as the mirror to observe your system in action, detect ways in which your understanding, reach, and quality of work needs to or can improve or change, and act on those opportunities so that individuals and the system as a whole can become ever better.

Because measurement creates expectations, and expectations create obligations, and obligations structure actions, and actions reflect values, measurement influences how you work and to what end. Leverage measurement to monitor and improve what matters most. To transform your system, you must transform the way you think about measurement.

To get started:


I believe that we reveal values in where we look and how we tell.

Imani Perry1Perry, I. (2022, May 13). Welcome to Unsettled Territory. The Atlantic. https://newsletters.theatlantic.com/unsettled-territory/627d6a1133833200211dca36/free-speech-student-protestors-freedom-fighters

Learning leaders use measurement to continually and comprehensively understand, test, and refine decisions, actions, and strategy. Faced with complex systems they alone can never master, learning leaders use measurement to make doing and learning inseparable for all system actors. Measurement is key to improving system-wide practices, including their own practice as leaders. And by using measurement to improve practice, they practice improvement, becoming ever better at becoming ever better and incrementally learning new and more effective ways to achieve their ultimate strategic aim: equity at scale. 

This approach represents a significant departure from measurement for compliance or measurement for accountability. Measurement for improvement prioritizes integrated, rapid, system-wide learning. Like a learning methodology, measurement is a component of system design, culture, and practice. Measurement for improvement helps answer the question, “How will we know a change is an improvement?” and requires: 

  • aligning measures to strategic inputs, drivers, and outcomes; 

  • measuring quality and quantity of services and effect;

  • setting before-the-fact performance expectations; and 

  • collecting information in responsive and affirming ways that support the participation of stakeholders throughout the system.

Done right, measurement communicates core values, focuses stakeholders on the strategy, and supports localized and system-wide learning. Learning leaders home in on select measures that allow them to gauge the most important parts of strategy. They focus on elements that are essential to advancing equity, most tightly aligned to the ultimate aim, and able to serve as early alerts when things fall off track.

To avoid hindsight bias2Roese, N. J., & Vohs, K. D. (2012). Hindsight bias. Perspectives on Psychological Science, 7(5), 411-426., learning leaders set performance expectations in advance. These expectations capture what should be observed if the hypothesis is valid, and they set success at a level that increases equity. Learning leaders build systems so that they can immediately detect deviations and respond to the constant influx of new insights generated every time measurement reveals performance gaps. They revise structures, teams, processes, and strategy in light of new learning.

REFLECT & ACT:

How does your system use measurement? 

  • How distributed are measurement practices and responsibilities? How might you restructure measurement so that it is integrated into everyone’s daily work?

  • What are the most core elements of your strategy? How do you measure them? What measures are most in focus? Do they also align with your core strategy?

  • Have you set performance expectations for each measure?

  • What alert systems (andon structures) exist in your system? Which work most reliably? Are there pockets where problems languish unaddressed? As a leader, how might you help set up more sensitive detection apparatuses?   
 

Measurement in Practice Achieve Atlanta

Explore how Achieve Atlanta has used measurement to drive equity and improvement.

Read the case.

Once committed to using measurement as a driver of equity and improvement, learning leaders develop the infrastructure to make this vision operational. Measurement serves as a unifying mechanism, not a structural barrier. Learning leaders architect their systems to derive learning from the many distinct yet interrelated dynamics that sustain them. They design integrated, self-reinforcing measurement structures that sit within the organizational learning structures discussed in Driver A and rely on the same networked, cross-functional, team-based configurations to generate system-wide learning. They leverage measurement as a part of this overarching systems design to make the interrelatedness among system actors and work more visible and more usable.

Learning leaders design systems to amplify common performance expectations so that everyone has a clear picture of the best-informed definition of success and is therefore able to detect when it is and is not met, no matter where in the system they operate. They also reject the inevitability of persistent failure, designing systems to detect deviations so that they may be addressed instead of designing systems to perpetually tolerate deviations or generate workarounds.

By demonstrating time and again that what happens in one area of the system affects others, measurement unifies; by demonstrating that what is uncovered, discovered, and improved in one area expands the capacity of the whole, measurement transforms, allowing the system to deliberately and continually advance.   

Read on to see how learning leaders in a well-architected system:  


Learning leaders build integrated measurement systems that gauge both the parts and the whole, the process and outcomes. To achieve this, learning leaders cease operating as mere auditors, who are spatially and temporally distanced from the site of the activity they are measuring, and enforcers, who use measurement solely to pass judgment. Instead, learning leaders concern themselves with keeping tabs on measures that apply across teams and functions and involve others in doing the same. They use measurement to make clear at the outset the definition of success for the system and the process expected to achieve that success, and to provide rapid alerts when there is a departure from either so that the entire system can improve its imperfect knowledge.

They also measure the quality and effect of their own leadership approach. Having previously captured their leadership approach as an explicit hypothesis, leaders set expectations for, observe, and improve their approach to designing, structuring, and supporting processes. As chief designers and managers of integrated processes, leaders rely on timely detection of challenges or flaws at every level of the system so that they might improve their understanding of and design of structures they put in place. Leaders acknowledge that when there is a departure from expectations anywhere in the system, that likely indicates a flaw in their leadership (or their own understanding of their leadership); therefore, they see it as an opportunity to become an ever better leader.

For example, the ultimate focus of the Tulare County Office of Education (TCOE) Network for School Improvement is ensuring students are on track for college and career, but student outcomes are not the only measure monitored. Adult behaviors and mindsets are also closely observed as proxies for the efficacy of leaders and the structures and processes they design. TCOE coaches log the behaviors of the school teams using a heat map and communicate results to hub leaders during regular meetings. Hub leaders use these data to interrogate their own leadership and shift network practices, for example by adjusting the learning arc at convenings. They also support coaches in doing the same, adapting their approach to coaching sessions based on feedback from school teams. In this way, TCOE avoids two common missteps: (1) tracking only lagging indicators and (2) viewing deviations from expectations as a problem with people, not a flaw in strategy or structure. Instead, TCOE leads through learning, positioning itself to know when work is off track well before high school students graduate, at a moment when there is still something to be done to address it. And they actively seek out and act to resolve structural flaws. 


Learning leaders design systems in which they and others actively identify, collect, and monitor leading and lagging indicators for each key measure. They prioritize proximity to process and ensure that as soon as there is a departure from expectations, alarms sound and teams coalesce to investigate and address the deviation. 

Aware that the passage of time erodes situational specificity, learning leaders design systems that steadfastly collect and make available “green” data,3Caillier, S. (Host). (2021, June 17). Don Berwick on building courageous networks” [Audio podcast episode]. In Unboxed. High Tech High Graduate School of Education. https://hthunboxed.org/podcasts/s2e21-don-berwick-on-building-courageous-networks/ which are as close as possible to the inciting action. These leaders sit within the day-to-day minutiae, not above it, establishing mechanisms for rapidly detecting deviations and just as rapidly communicating that they have taken place.4Spear, S. J. (2009). The high-velocity edge: how market leaders leverage operational excellence to beat the competition (2nd ed). McGraw-Hill. These mechanisms may come in the form of designated reporting and response teams, communication tools (see Driver A Andon Cord), or regular incident review meetings. Because of this rapid detection, communication, investigation, and resolution, system actors are able to craft solutions with the most accurate information at hand and are able to act fast, springing into action before any deviation becomes calcified. 

Rapid detection requires leaders to design systems that track leading indicators as well as lagging ones. Leading indicators allow for the prediction of future outcomes by gauging performance right now. In other words, leading indicators help measure a process in progress, while the process is live (not stale), and therefore more readily adjusted, inspected, and improved. 

Strategic teaming is one approach the Institute for Learning takes to ensure a steady supply of green data from the ground level. In particular, they leverage their coaches as antennae for the system. Coaches are regularly in touch with system actors, their finger always on the pulse of day-to-day activity, feelings, and challenges so that they can funnel information to those responsible for designing structures and processes. High Tech High also structures work so that green data can flow through intermediaries. Their data lead and lightning rod roles ensure open flows between various levels of the system.


To ensure the uptake and feasibility of tracking key measures, learning leaders design measurement to be simple, accessible, and built into everyone’s everyday work. One of the more complicating aspects of measurement is data collection and management. To assess performance on key measures, learning leaders use qualitative and quantitative data to generate an accurate read of system health and equitable service provision, not just a confirmation of biases. As chief aggregators, processors, and collectors, leaders are well-positioned to support these kinds of strong data practices that make it possible to derive learning and track progress. To do so, learning leaders ensure that learning goals drive data collection, not the other way around. They then define measures—alongside practitioners and other stakeholders with expertise—and ensure that the volume of data is manageable and the information provided is closely aligned with measures and learning goals.

Learning leaders also design smooth pathways for accessing data, so that any system actor can leverage data for improvement. They manage data-sharing agreements with partners that might otherwise complicate or prevent collaborative learning. They navigate and maintain relationships with various parties who supply data as well as the complex, technical and logistical aspects (i.e., completing paperwork and compliance documents) of communicating data across traditional boundaries.

They also ensure system preparedness for collecting and mining data. They select and design data tools for their accessibility, cost-effectiveness, and likelihood to be used, even by those without an extensive data background. Learning leaders design systems that make it as easy as possible for system actors to share interesting insights. One way they do this is by integrating data tools and communication tools. For example, they use shared platforms (e.g., Google Data Studio) to store data, design roles that take on the often time-consuming process of consolidating data so that they are accessible to others, and normalize the practice of embedding graphs or data set links into emails. The more readily available these data are, the more usable and useful they are to collective learning. This design feature allows learning leaders to maintain their ability to set up and support strong measurement practices, while also making it possible for others to participate in the facilitation of data monitoring, sharing, and interpretation. 


REFLECT & ACT:
  • How have you set up measurement to monitor core parts of the system, including your own leadership approach?

  • When and why do you measure both process and outcome? When and why do you not do so? Are there neglected leading indicators that would give you an early read on performance?

  • How are you smoothing the path for others to access and make use of data? What obstacles to accessing data could you mitigate?

  • Which of your data tools do not work for you or the system? In what ways might you simplify these tools or make them more accessible?

Even leaders with the most well-designed improvement-driven measurement processes recognize that structures alone are not sufficient for transforming their systems to sustain equity at scale. They acknowledge that people drive systems, not tools. And culture drives people: providing an essential grounding, motivating, and unifying force that turns individuals into movements—“I” into “us.” Leaders who foster a strong culture of measurement for improvement design systems in which people share data freely, approach measurement with an excitement and openness to discovery, and are agile and facile data analysts.

Learning leaders foster a culture in which the benefits of being open and transparent far outweigh risks conventionally associated with data, measurement, or accountability. They counter any stigma or fear around data and measurement by normalizing the practice of being transparent and forthright about results and the lessons they provoke. To start, they apply this level of transparency to their own and system-level data, openly exploring improvement opportunities for their own leadership and system design. 

In such a culture everyone sees and experiences measurement as valuable to their individual and collective goals, feels compelled to share data—favorable or unfavorable— and uses measurement to further progress. As discussed in Driver A, at every turn learning leaders reinforce a revised definition of success—equity through the pursuit of ever better—and build into their systems a sense of collective responsibility to that chief goal, motivating system actors to admit to and interrogate their missteps and to seek improvement even in areas in which they are less directly involved. 

Measurement supports learning when leaders:


As a part of their measurement structures, learning leaders collaboratively define key learning goals and measures (what matters most) that are applicable system wide. As a cultural reinforcement, learning leaders make these goals and measures visible, reinforcing the reality that all system actors use data to fuel individual and collective learning, from ground-level actors to leadership. Learning leaders make explicit what the group gains from ongoing inquiry, building awareness and connections across various areas of the system. They build an appetite for new insights and reliably distribute that learning across the organization (see Knowledge Management). They continually connect learning goals to broader aims, showing how measurement helps propel everyone forward and making it clear that each person’s learning is connected to and depends on others’. 


Learning leaders develop a keen understanding of how familiar and comfortable system actors are with data monitoring, including by having explicit conversations about their capacity to engage in effective data practices. Once they have a sense of the existing capacity of actors within their system, learning leaders make available professional development and one-on-one data support to build capacity and meet immediate data needs. They democratize the measurement process so that those closest to the problem participate in meaning making, while also structuring systems to remain efficient. They spot and take advantage of opportunities to centralize measurement and make it feasible for system actors with a range of experience levels to engage in measurement. For example, learning leaders may opt to introduce measurement to the newly initiated by using readily available data, rather than relying on data that cannot be accessed without formal agreements. This helps less experienced system actors experience more quickly the benefits of measurement for learning and avoid getting mired in the complexity and tedium of securing hard-to-access data.


Learning leaders design systems that seek out and value measures beyond the quantitative. Recognizing that measures are mere proxies for what can often be challenging to define or gauge, learning leaders build flexible, humane systems capable of bending and deferring to the voices of stakeholders and system actors, particularly when they capture something measures cannot or have failed to. 

To do this, learning leaders construct direct communication channels between system actors closest to the ground level and those further removed so that all levels of the system have access to “street data.”5Safir, S., & Dugan, J. (2021). Street data: A next-generation model for equity, pedagogy, and school transformation. Corwin. They value insights from the ground level as critical, as much or more so than more traditional quantitative indicators. Without these insights, collective understanding remains incomplete, hypotheses are flawed, and systemic transformation is continually thwarted. Because learning leaders know that understanding their system is key to transforming their system, they think expansively, not narrowly, about the sources of understanding. Learning leaders supplement quantitative data with the riches in the softer spaces, the conversations at student council meetings, the check-ins with family members, and the chats before community board meetings. They create space for the messy, the uncomfortable, the honest; conversations and testimonies that round out understandings about how the system is truly functioning. They structure teams to include those closest to implementation and impact, particularly when generating strategies or approaches. 


REFLECT & ACT:
  • What do you do to ensure you measure what matters most? How do you remind yourself and others of these key measures so they build connections and structure actions?

  • Have you consistently involved in measurement key stakeholders? 

  • What are system actors’ capacity and appetite for engaging in measurement? What capacity building structures have you put in place to support them? 

  • When and how do you look beyond qualitative data to understand your system? When and how do you collect qualitative data directly from people closest to the problems? In what ways are you intentionally including the voices of those who have historically been locked out of your system, in discussions on measurement and strategy? What opportunities for deepening your understanding of your system and its effects have you missed?

What’s Next?

Driver A Treat every strategy as learning

Driver B Foster democratic participation

Driver D Build a democratic knowledge-management cycle

Companion GuideDevelop Your Theory of Leadership

X