Professional Learning  ›  Leadership & Coaching  ›  Workshops  ›  Piloting Small Tests of Change
a ready-to-run workshopPiloting Small Tests of Change
Leadership & Coaching · Workshop

Piloting Small Tests of Change (Workshop)

Everything you need to lead this with your staff or PLC: a minute-by-minute run-of-show, what to say, what to hand out, and how to make it stick. Pick the length that fits your slot.

Key takeaway

Launching a change school-wide all at once bets everything on getting it right the first time and fails large when it does not, while testing the change on a small scale first, a pilot or a small test of change, lets a leader learn what works, fix what does not, and build evidence and buy-in before scaling, which de-risks the change and makes the eventual rollout far more likely to succeed.

Format
PLC / team meeting
Length
45 minutes
Group size
Any (works 4-40)
Who can run it
Any leader or administrator
You will need
Slides (5)Printed handoutA change you are consideringA timer
Aligned to
Learning Forward: Learning CommunitiesInTASC 10: Leadership & CollaborationImprovement science and research on piloting change
Share / assign

Why it works

When a school decides to make a change, a new approach, program, structure, or practice, the default is often to launch it everywhere at once: announce it, train everyone, and roll it out school-wide from a start date. This all-at-once approach has a serious flaw: it bets the entire change on getting it right the first time, with no chance to learn before committing everyone, so when the change has problems, as nearly every change does, those problems hit the whole school at once, the change fails at scale, staff conclude it does not work, and the failure sours them on the next change too. Piloting, or running small tests of change, offers a fundamentally better approach drawn from improvement science: test the change on a small scale first, with a few willing teachers or one grade level or a limited version, learn from that test what works and what does not, refine the change based on what you learn, and only then scale it up, informed by real evidence rather than hope. The logic is that you cannot know in advance exactly how a change will play out in your specific context, so rather than betting everything on an untested design, you find out on a small, low-stakes scale where problems are cheap to discover and fix, and you scale only what has been shown to work and been refined through the testing. The benefits are substantial. You learn and adjust: the pilot surfaces the practical problems, the things that do not work as imagined, that you can then fix before they affect everyone. You de-risk the change: a pilot that fails costs little and teaches much, whereas a school-wide launch that fails costs enormously. You build evidence: a successful pilot provides proof that the change works in your context, which is far more persuasive than a leader's assurance. And you build buy-in: pilot teachers who had a good experience become credible advocates, and evidence from a local pilot overcomes skepticism in a way that mandates do not. Running pilots well involves several principles. Start small and willing: begin with a manageable scale and, ideally, willing volunteers rather than conscripts, since the goal is to learn, not yet to mandate. Define what you are testing and how you will know: be clear about what the pilot is meant to find out and what evidence will tell you whether it is working, so the pilot actually produces learning rather than just happening. Genuinely learn, including from failure: the pilot's value is in what it teaches, so you must be genuinely open to finding that the change needs adjustment or does not work, not run a pilot that is really just a pre-decided rollout in disguise. Adjust based on what you learn: refine the change in response to the pilot rather than scaling it unchanged. And scale thoughtfully: use the pilot's evidence, refined design, and advocates to inform a wider rollout, which may itself proceed in stages. A caution: a pilot is only useful if you are genuinely willing to learn from it, including to change course or stop; a pilot run to rubber-stamp a decision already made, ignoring what it reveals, wastes the opportunity and breeds cynicism. The workshop helps leaders test changes on a small scale first, learn and adjust from the pilot, and scale what works with the evidence and buy-in a pilot builds, rather than launching untested changes school-wide.

The research: Improvement science and research on piloting, small tests of change, and scaling.

The run-of-show

Choose your slot. The agenda, timings, and length update to match.

    The core activity: test small, learn, then scale what works

    The heart of the session is testing a change on a small scale first, learning and adjusting from the pilot, and scaling what works with the evidence and buy-in a pilot builds.

    Test small where failure is cheap, then scale what worksAn all-at-once launch bets everything on a first try and fails large. A pilot finds and fixes the problems on a small, low-stakes scale, builds evidence and advocates, and lets you scale a refined change that is far more likely to succeed.

    Facilitator notes

    A pilot you will not learn from is theaterIf the decision is already made and the pilot exists only to rubber-stamp it, you gain none of the learning and breed cynicism when you ignore what it reveals. A real pilot means being genuinely willing to adjust, or even stop, based on what it shows.

    Adapt it

    Participant handout

    One page for every teacher. Print it, or save it as a PDF.

    K12 Academics · Professional Learning

    Piloting Small Tests of Change: learn before you scale

    De-risk change by testing it small first: pilot with a few willing teachers, learn what works and what does not, refine it, and scale with the evidence and advocates the pilot builds.

    1. A change I am considering:
    2. How I would pilot it small (who, how limited):
    3. What the pilot is testing and how I will know if it works:
    4. How I stay genuinely open to learning, including that it needs changing:
    5. How I scale with the pilot’s evidence and advocates:

    Make it stick

    Piloting de-risks change and builds buy-in. Close by setting the follow-up:

    Common questions

    Isn't piloting just a slow way to do what we could do all at once?
    It is slower at the start and faster overall, because the time a pilot takes is far less than the time, cost, and damage of a school-wide change that fails because it was launched untested. The all-at-once approach looks faster, you announce, train, and launch everywhere, but it bets the entire change on getting it right the first time with no chance to learn, and since nearly every change has problems that only surface in practice, an untested school-wide launch routinely hits those problems at full scale, fails, wastes the enormous investment of training and effort everyone made, and sours staff on the change and on the next one, all of which costs far more time and goodwill than a pilot would have. A pilot front-loads a modest amount of time to find and fix the problems cheaply on a small scale, so that when you do scale, you are scaling a refined change that works, with evidence and advocates behind it, which makes the rollout far more likely to succeed the first time at scale. So the comparison is not pilot-slow versus all-at-once-fast; it is a small up-front investment that greatly raises the odds of eventual success versus a fast launch that often fails expensively and has to be redone or abandoned. For trivial or low-risk changes, you may skip piloting and just proceed, but for significant changes, the pilot is the faster route to a change that actually sticks, because it avoids the far larger time sink of a failed school-wide launch.
    What makes a pilot different from just rolling something out to one group first?
    The difference is genuine learning: a real pilot is designed to find out whether and how the change works and to adjust it based on what is found, whereas rolling something out to one group first without that learning intent is just a staggered launch that gains little. A staggered launch starts with one group but treats the change as settled, collecting no real learning and making no adjustments, so it misses the whole point and often just means the first group hits the same unaddressed problems a bit before everyone else. A true pilot, by contrast, has several features that make it a learning activity: it is clear about what it is testing and what evidence will show whether it is working, so it actually produces knowledge; it is approached with genuine openness to discovering that the change needs adjustment or does not work, rather than as a foregone conclusion; it is used to refine the change based on what is learned before scaling; and it is run on a scale and with participants (ideally willing volunteers) chosen to maximize learning at low stakes. The mindset is the key difference: a pilot asks what can we learn here that will make this change better and more likely to succeed, while a mere staggered rollout asks only how do we get this going group by group. So doing it to one group first is necessary but not sufficient; what makes it a pilot is the deliberate design to learn and the genuine willingness to adjust, which is exactly what turns the small-scale start into the de-risking, refining, evidence-building activity that piloting is meant to be.
    What if the pilot shows the change doesn't work, haven't we wasted the effort?
    A pilot that shows the change does not work has succeeded, not wasted effort, because it has saved you from the far greater waste of launching a failing change school-wide, which is exactly what piloting is for. This reframe is central: the purpose of a pilot is to learn, and learning that a change does not work in your context, or needs significant adjustment, is immensely valuable learning, because it lets you fix or abandon the change on a small, cheap scale before it has consumed the whole school's training, effort, and goodwill on something that would have failed anyway. A pilot that reveals a fatal flaw has done its job perfectly, discovering cheaply what would otherwise have been discovered expensively at scale. The only way a pilot wastes effort is if you ignore what it tells you, running it as a rubber-stamp for a decision already made and scaling the change despite the pilot showing it does not work, which is both a waste and a breeder of cynicism. So the honest stance going into a pilot is genuine openness to all its possible outcomes: it works and you scale it, it works with adjustments and you refine then scale, or it does not work and you change course or stop, and all three are successful uses of a pilot because all three produce the learning that improves your decisions. Treating a negative pilot result as a failure rather than as valuable information gets the logic backwards; the pilot that stops a bad change early is doing you a great service, and recognizing that is what makes you genuinely willing to learn from pilots, which is the condition for their value.

    Go deeper

    Build it into a bigger day

    Workshop
    Leading Continuous Improvement
    Workshop
    Scaling What Works
    Workshop
    Fostering Innovation & Risk-Taking