An improvement plan is not enough — you need a strategy 



When making plans for how to meet their goals, administrators sometimes neglect to think expansively about what it will take to get there. 


Not long ago, I visited a nearby school to learn about its approach to literacy coaching. However, when I met with the assistant principal and asked her about the specifics of the coaching model, she told me she didn’t know much about its details, so I asked to see the school improvement plan. The plan described the initiative, its instructional focus, and its objectives, and it listed various action steps to be accomplished, along with descriptions of who would be responsible for meeting each goal, a set of progress indicators, and a timeline. For example, it noted that instructional coaches would meet with teachers, run school data teams, and submit notes and agendas to the principal.  

But what I was looking for — and could not find — was concrete information about what those coaches were actually supposed to do to influence and improve instruction. Precisely what was supposed to happen when they met with teachers? What would they talk about? What sorts of practices would they demonstrate, and how? What practices did they aim to challenge, and why? What did they want teachers to learn, and how would they help them learn it?  

This particular school has seen no improvement in its test scores or other indicators in several years. Sooner or later, budget-conscious administrators will point out that the instructional coaching has had no obvious impact on teaching and learning, and they’ll ask whether the effort should be abandoned. But how can anybody defend the work, given how vaguely it has been described? If the improvement plan neglects to tell coaches to do anything more specific than “meet with teachers and report to the principal,” then who’s to say whether it has been implemented well, badly, or not at all? 

There is a great danger of investing in educational programs without offering a clear explanation of what, exactly, they entail and why they are worth pursuing. 

It has often been noted that while all organizations have a mission statement, not all of them have a mission. Just so, while most organizations have a plan, few of them have a strategy. A plan is just an account of who will do what and when. By contrast, a strategy is a tight, cogent set of ideas about how best to fulfill the mission of the organization. It provides context, goals, and priorities, giving people a sense of how they should implement the plan and what it will take to do so successfully.  

There is a great danger of investing in educational programs without offering a clear explanation of what, exactly, they entail and why they are worth pursuing. School and district leaders need to make sure they provide a clear through line to their vision of success, including the strategy for getting there and the specific steps to undertake to execute that strategy.  

The weakness of our current practice 

Most school and district leaders are fluent in the language of continuous improvement, data analysis, and theories of action. They know that they ought to have improvement plans — they just don’t know how to create excellent ones, what to do with them, or how to connect them to a larger cycle of inquiry. They are aware that these plans are, to borrow from Shakespeare, more honored in the breach than the observance. Old strategies are abandoned because scores have not improved, without sufficient questioning about why they failed and without adequate reason to believe that the next initiative will have a better chance of success. When things go wrong, they tend to assume that it is the fault of the people implementing the plan, not a flaw in the plan’s design.  

Given how often school improvement plans falter and fail, we should at least consider the possibility that our plans themselves — and not the millions of educators struggling gamely to implement them — are to blame. 

However, given how often school improvement plans falter and fail, we should at least consider the possibility that our plans themselves — and not the millions of educators struggling gamely to implement them — are to blame. 

Here are some of the mistakes that the typical improvement planning process encourages: 

Confusing the product with the process 

The way we currently engage in improvement planning, with the emphasis on generating an action plan, is antithetical to the idea of continuous improvement. The plan is a product that is supposed to be a proxy for a process, but an intense focus on creating the plan can itself distract from the details of how it will be implemented (or whether it will be implemented at all). Often, for example, when supervisors give feedback on improvement plans, they address the document’s format and whether or not it includes certain required elements, but they say little or nothing about the substance of the plan or its real-world likelihood of succeeding. For example, many districts require that school plans include SMART (specific, measurable, achievable, relevant, and time-bound) goals — which can easily send the message to principals that the priority is to make sure their plans include goals that fit the SMART definition, not to make sure the plan is, in fact, a smart one.  

Writing plans too quickly  

A typical template for a school improvement plan looks more or less like the one I described at the beginning of this article. It may include district and school mission and vision, SMART goals, priorities, focus areas, theories of action, and a root cause analysis, among other things. However, most principals are under pressure not just to produce a plan that includes these and other “must-haves” (the creation of data teams, for example) but also to meet tight deadlines, which only increases the likelihood that they’ll focus on checking off the necessary boxes, rather than thinking strategically about how to make the plan work. That’s why, when I consult with principals, I often urge them to ask their district for more time, so that they can figure out, for instance, what those data teams will actually do, what obstacles they’re likely to face, what kinds of support they’ll need, and so on. 

Making too many assumptions about what will work 

The world of education is filled with too many unexamined assumptions. As the authors of Learning to Improve (Bryk et al., 2015) put it, school and district leaders tend to suffer from solutionitis: “a form of groupthink in which a set of shared beliefs results in an incomplete analysis of the problem to be addressed and fuller consideration of potential problem-solving alternatives.” For example, many plans include some form of instructional coaching. However, few of those plans address the challenges involved in making instructional coaching worthwhile: How, exactly, will the coach assess the 1st-grade teachers’ current approach to reading instruction? How will they share their findings with the teachers? Rather, most improvement plans appear to rely on the assumption that veteran teachers already know how to coach novice teachers effectively. Thus, plans typically address technical and procedural matters (e.g., “Each teacher will have the opportunity to work with a coach for two coaching cycles”) but say nothing about how to identify coaches, support them, and ensure that they do the job well.  

We all suffer from an optimism bias — thinking that things are going to be easier and quicker to accomplish than they actually are — and this plays out in the creation of improvement plans. Leaders would do well to seek out multiple perspectives on where the bumps are likely to appear along the way. 

Trying to do too many things at once 

This problem is a corollary of the optimism bias. Because many of us are unrealistic about what it will take to get things done, we list more priorities than we can possibly accomplish in the given time frame. One of the exercises I use with leadership teams is to ask them to write every initiative going on in their districts on a different sticky note. Often, they cover an entire piece of chart paper with sticky notes, which is a pretty good indication that they’re trying to do too many things at once. Similarly, many districts make the wise decision to focus on just one big priority, only to assume they can address all of its many parts simultaneously. For example, they might give themselves only one year to create and implement a new approach to classroom formative assessment, even though that will require surveying teachers’ current practices, reviewing the research literature, considering various approaches, creating professional development opportunities, and so on. 

Neglecting to clarify who will do what  

While an improvement plan might include a column for the “person responsible,” it may nonetheless exclude other educators with key roles in implementation. For example, under an action step such as, “Meet monthly to talk about math curriculum,” the math coordinator might be specified as the responsible party. Or, the task of “providing time for departments to meet regularly” might be assigned to the principal. However, neither of these indicates what teachers’ roles might be, or whether and how they are supposed to teach differently. It is possible for teachers — who, let’s face it, are the key players in any plan to improve student achievement — to look at a school improvement plan and have no idea what, if anything, they are being asked to do. This is a problem. 

Underestimating the urgent need for professional learning 

In any given organization, as W. Edwards Deming (2000) famously pointed out, most people are already doing the best they know how to do. Thus, the most important goal of any organizational improvement strategy is to give people the sorts of professional learning they will need in order to do their work in new and better ways. In many school improvement plans, however, the writing of a professional learning plan is listed as one of the action steps (i.e., the plan is to come up with a plan to learn how to make the plan work), which seems like a surefire recipe for failure. If we want educators to do something different, we can’t afford to put off the details of professional development until some later date.  

What should we be doing instead? 

I work with schools and districts on designing and leading school improvement strategies, and before that I wrote a few plans of my own. The longest plan I ever had to write was more than 70 pages long, before educators like Doug Reeves (2006) and Mike Schmoker (2011) cried uncle and suggested that there is an inverse relationship between the length of a plan and its potential to make a positive difference. That was certainly a useful step toward halting the madness, but it still didn’t make the plans much better, just shorter. 

My big aha moment came when I was involved in developing a logic model for evaluating a grant for a coaching program. A logic model is a tool researchers use to measure the effectiveness of a program. It lays out the actions in a given project in a causal chain, describing short-term and long-term outcomes of those actions. It is an important tool for researchers because for evaluators to understand what’s really going on, they have to know not only the intended ends, but also the means.  

The grant evaluator in this instance asked us to describe leadership behaviors that new principals were supposed to demonstrate as a result of being coached, which would indicate what they learned from their coaches. Then she asked us to describe the actions of the coaches, and what they had to learn in order to coach effectively. Then she asked us to describe our actions and how they would lead to learning on the part of the coaches. 

I was taken by surprise at how hard this was to do, because I had to make explicit all my thinking about the relationship between coaching someone and their behavior changing. I realized that, for all my knowledge about coaching, I was not completely clear on the chain of causation (what Michael Barber, 2015, calls a delivery chain) from the activities described in the grant proposal to improved school leadership. It took several months and several drafts to get to a tightly focused program plan.  

That’s when it hit me that this process — thinking through the proposed actions and how they were supposed to connect to intended results — tends to be missing from school improvement planning. As I’ve come to understand, if organizational consultants hope to guide schools and districts in developing strategies that actually have a good chance of succeeding, they will need to make sure their plans meet at least six essential conditions: 

  1. The strategy employed is powerful enough to reach the goal.
  2. The through line from central office or school leadership to what students will experience in classrooms is clearly articulated.
  3. The steps involved are sufficiently detailed to make the strategy high leverage.
  4. The educators in roles germane to the strategy have expectations clearly laid out for them — the “do” columns are, in effect, job descriptions.
  5. The learning that is required in order for educators to do what they are being asked to do is described — essentially, there are professional learning plans for everyone involved in the plan.
  6. Evidence of progress for each column is specified, which often entails creating instruments and/or procedures that do not yet exist.

The information needed to demonstrate that these six conditions are met can be represented on one (usually large) sheet of paper, which I call a strategy map. The fact that it is only one sheet of paper is important for reasons having to do with simplicity and portability. (And paper is preferable to an online document because it is too easy to change an online document and lose track of which is the authoritative version.)  

There are plenty of resources available to schools and districts wanting to think more deeply about strategy. For example, there is a guide available online from the W.K. Kellogg Foundation (2004) on how to build good logic models, and there are well-regarded books, such as Strategy in Action (Curtis & City, 2009) and Learning to Improve (Bryk et al., 2015), that offer principles and procedures for developing and executing strategy. However, the problem with these books and guides is that they are, in a sense, too good. They lay out an in-depth approach to improvement, embracing the best tools in the action science canon. But in practice, I’ve found, the steps specified and the time required put too great a demand on educators who are already pressed for time. The strategy map is altogether more manageable and more easily translated into a source document that can drive the agenda of every meeting. 

My experience with using this template to map strategy is that people don’t like it very much — at least not at first. They often find it overwhelming and difficult until they’ve had several chances to interact with it, such as by taking an existing school improvement plan and trying to translate it to this format and being surprised at what is missing. But they appreciate being given a tool that is relatively straightforward and that supports them in thinking through everything that needs to be done to move closer to a specific goal. 

So what does this look like? 

The superintendent of a district in Connecticut and her team created a strategy map focused on two aspects of students’ experience of school: climate and feedback. The district has also embarked on an initiative to make teacher leaders central to their school improvement efforts, by providing them the opportunity to take a summer course on teacher leadership. The work of creating the strategy map, then, was to draw the through line from what the central office team had to do to support this work, and what they had to learn in order to do that, through to what classrooms are supposed to look like in Region 14. A fragment of their strategy map is shown in Figure 1, describing a part of the causal chain that they believe will result in aspects of instruction they want to promote. 


Along the way, a number of groups were involved in providing input to the map. Not only is it important that teachers are afforded the time and space to talk about how they do their work and what they need to learn, but it is also important that they be asked to describe the support they need from others. This is an important step not just in terms of buy-in, but because we are frequently more insightful about what others should be doing than we are about ourselves. 

This process — thinking through the proposed actions and how they were supposed to connect to intended results — tends to be missing from school improvement planning.  

Once the input had been collected, the next step was for the central office team to determine how they would know whether the learning and doing described in the map were happening as intended. This can be tricky in schools and districts where teachers and administrators are not used to being observed. While this was not the case in Region 14, figuring out how to assess whether the strategy was being implemented as designed still took considerable time and thought, and included in some cases the creation of surveys and observation tools.  

The final strategy was considerably simpler than many school improvement plans, but it provides just as much helpful information. For example, the high school devised a deceptively simple guide for teachers, and observers of teachers, that summarizes what all teachers should focus on. The guide includes a section on student ownership of the learning, which specifies that students should be working on challenging tasks, with each other, for the majority of instructional time. This, in itself, is not revolutionary, but it is part of a larger plan of providing professional learning to all teachers so that the tasks they assign are indeed challenging, that students have voice and choice, that failure is a normal and valued part of the learning process, and that students receive actionable feedback on their work. And, perhaps most important, everyone is clear that this is the focus for the district for the foreseeable future. 

Strategy over planning 

I don’t wish to suggest that everyone who writes a plan makes the mistakes I’ve described — indeed, I know they don’t, because I know many thoughtful, strategic, and successful educators who swear by their plans. But too many schools and districts think of improvement planning as an exercise in compliance because the format used encourages completion at the cost of contemplation, and accountability at the expense of capability.  

Just to be very clear, I am not suggesting that schools and districts should not plan. I am suggesting that planning has been corrupted by the need to create a plan, rather than a good one. We should move away from a format that is all about deadlines and responsibility and instead devote our time to creating a strategy and studying how that strategy could be improved.   


Barber, M. (2015). How to run a government: So that citizens benefit and taxpayers don’t go crazy. London, UK: Penguin. 

Bryk, A.S., Gomez, L.M., Grunow, A., & LeMahieu, P.G. (2015). Learning to improve: How America’s schools can get better at getting better. Cambridge, MA: Harvard Education Press. 

Curtis, R. & City, E.A. (2009). Strategy in action. Cambridge, MA: Harvard Education Press. 

Deming, W.E. (2000). Out of the crisis. Cambridge, MA: MIT Press. 

Reeves, D.B. (2006). The learning leader: How to focus school improvement for better results. Alexandria, VA: ASCD. 

Schmoker, M.J. (2011). Focus: Elevating the essentials to radically improve student learning. Alexandria, VA: ASCD. 

W.K. Kellogg Foundation. (2004). Logic model development guide: Using logic models to bring together planning, evaluation, and action. Battle Creek, MI: Author.


Citation: Stevenson, I. (2019). An improvement plan is not enough — you need a strategy. Phi Delta Kappan, 100 (6), 60-64. 

ISOBEL STEVENSON (; @isobeltx) is director of organizational learning at the Connecticut Center for School Change, Hartford, Conn. She is the author of the weekly newsletter, The Coaching Letter ( 


  • Thank you for a thoughtful and practical article that I consider a ‘must read’ for every school and district leader. Click the share button to get this in the hands of educators!

  • Jenny M

    Thank you for posting this. I am wondering if I am missing the image for Figure 1, referenced in the text. Is this available? I think having an example of the strategy map would be very helpful in understanding the format suggested.

    • Michele Jacques

      Thank you so much for noticing! The figure was inadvertently left out. It has now been included in the article.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

WP_User Object ( [data] => stdClass Object ( [ID] => 651 [user_login] => IStevenson [user_pass] => $P$BHCSDF9OD8GPNau80pSBchVFFYPT0y/ [user_nicename] => istevenson [user_email] => istevenson@fake.fake [user_url] => [user_registered] => 2019-02-15 19:18:49 [user_activation_key] => 1550258330:$P$BKv5GUxtruHciHD6i5lHJER.pkymjt/ [user_status] => 0 [display_name] => Isobel Stevenson [type] => wpuser ) [ID] => 651 [caps] => Array ( [author] => 1 ) [cap_key] => wp_capabilities [roles] => Array ( [0] => author ) [allcaps] => Array ( [upload_files] => 1 [edit_posts] => 1 [edit_published_posts] => 1 [publish_posts] => 1 [read] => 1 [level_2] => 1 [level_1] => 1 [level_0] => 1 [delete_posts] => 1 [delete_published_posts] => 1 [author] => 1 ) [filter] => [site_id:WP_User:private] => 1 ) 651 | 651


What leaders can (and cannot) learn from scientists 

Shifting from data to evidence for decision making 

Retool state-to-district intervention for better outcomes

Improving 100,000 schools, one at a time