Switch course ▾
Session 5. From Pilot to Programme: Scaling and Sustaining on a Low Budget
Objective. Run a pilot as a learning test with stop / continue / change criteria, decide what must stay the same when you scale, build a three-year growth plan and a low-cost budget in hours and money, and present a three-minute briefing to a board or education department.
Most programmes die not in the pilot but in the year after it: a decision made on feeling, a rapid roll-out, a leader who moves. This lesson goes beyond the key ideas. It shows how to read pilot results against rules you wrote in advance, how to look for the conditions behind the numbers, and how to put the case to a board so that the decision it takes is small, clear and possible to say yes to.
Key ideas
A pilot is a test designed to teach you something
- Keep it small and bounded: for example four coaches, eight to ten coachees, one term. A good pilot has three or four questions (Can we find the hours? Do pairs keep meeting? Do teachers say it helps? Is practice changing?), and answers to them are collected on purpose.
- Before the pilot starts, write the stop / continue / change rules, for example: continue if at least 8 of 10 planned meetings are held and most coachees report it is useful; change if time is the problem; stop if pairs break down in spite of repairs. Decisions made in advance are harder to bend later.
- Use short improvement cycles within the pilot: Plan a small change, Do it for two weeks, Study what happened, Act. The Institute for Healthcare Improvement PDSA worksheet is a free template from another field that works well for schools. Remember Kraft, Blazar and Hogan: programmes at larger scale had smaller effects, so learn what protects quality before you grow.
What must stay the same and what may bend
- Name the non-negotiables from the pilot: the features without which it stopped working. Typical ones: protected time, trained coaches, a coordinator, the no-appraisal promise, the clinic. Everything else (meeting day, form layout, the subject focus) may be adapted by each school.
- Grow capacity before reach: train the next group of coaches (often the best coachees after a year), add one school or department at a time, and set up a small coordinating team. Cynthia Coburn's article 'Rethinking Scale' argues that scale means deep and lasting change and not only more numbers; ask whether practice really changes in the new places.
- In a district, use a hub-and-cluster design: a hub school with trained coaches serves three to five nearby schools; coordinators meet monthly across the cluster. Keep a short handbook so new schools do not copy by rumour.
Low budget: the main cost is time, so spend it wisely
- List costs in hours first: coach time, coordinator time, training hours, clinic hours. Money is small: paper, printing, certificates, tea at clinics, perhaps an allowance. Counting hours makes the real cost visible to leaders.
- Cheap levers: swap duties for coaching hours; hold clinics inside existing methodical meetings; timetable common free periods; use parallel classes so a coach can observe; let coachees become coaches; use peer trios; ask the district to recognise coaching in professional development records where rules allow.
- Training alone is a weak buy. The Global Education Evidence Advisory Panel's 'Smart Buys' report rates structured pedagogy combined with teacher training and learning materials as a good buy, and warns that investing in inputs without complementary changes is a bad buy. The lesson for a programme: do not pay for training days that no one follows up in the classroom.
Sustain it and ask upward with a clear case
- Plan for the day the champion leaves: a one-page programme policy, a named deputy for the coordinator, a handbook, a budget line in the school plan and a yearly review date. Programmes tied to a person end with that person.
- Use a three-minute board briefing: (1) the ask in one sentence; (2) the need and the evidence; (3) what we already did with little money; (4) the cost in hours and money; (5) the main risk and how we will handle it; (6) the decision we ask for today.
- Know what to end. If a pair, a school or a form is not helping after repairs, stop it openly and say what you learned. A programme that can end parts of itself is one that people trust to continue the rest.
A short improvement cycle (PDSA)
Reading pilot results against rules set in advance
Write the rules before the pilot, for example: continue where at least eight in ten meetings are held and most coachees say it helps; adapt where the problem is time; pause where conditions are missing. Then place each school against the rules. Do not average across schools. A cluster average of 'good' can hide one school that failed and one that discovered a cheaper way to do the same thing.
Ask what was different about the best and the worst sites, and write the answer down. If the best school had a strong coordinator and the worst had none, the condition, not the school, is the lesson. This becomes the non-negotiable you write into the handbook, and it is the first thing each new school must show before joining.
Making the case upward
A board decides quickly on small, clear requests. Lead with the ask in one sentence, link it to something already in the school development plan, and show what was achieved with little money. Give the cost in hours before money: boards understand that two periods a week for a coordinator is a real price. Have one teacher's short quotation ready, used with permission.
Name the main risk yourself, and the step that reduces it. Leaders trust people who say 'what might go wrong' more than those who promise nothing will. Finally, protect the decision: a one-page policy and a named deputy mean the next head can read what was agreed, and the programme does not rest on one person's memory.
Examples from the classroom
Applying the rules in Namangan
Scene. A Namangan cluster pilot has four schools. The rule set in advance says 'continue if at least 8 of 10 meetings are held'. School 1 held 9, School 2 held 7, School 3 held 3 and School 4 held 9. The cluster average, 7, looks acceptable, but the decision differs: School 1 and 4 continue, School 2 adapts with more coordinator time, School 3 pauses until it names a coordinator and protects a free period.
Insight. Decide school by school against pre-set rules; an average hides the conditions that matter.
Hub and cluster in Fergana region
Scene. A district cannot give every school a trained coach. It chooses a hub school with four trained coaches and links it to four nearby small schools. Each month, coordinators from the five schools meet for 90 minutes, run the clinic together and share one problem. The district writes a ten-page handbook from the pilot. New schools join in the second year only after naming a coordinator and showing their timetable.
Insight. A hub-and-cluster design extends skill without a coach in every school, and the handbook keeps quality.
The policy page in Bukhara
Scene. Deputy head Kamola knows she may be promoted, so in the third year she writes a one-page policy with the head: purpose, forms, the no-appraisal promise, the clinic, the coordinator's two periods, a review date each June. She trains a deputy coordinator and shows her the register. When Kamola leaves, the new coordinator takes over with the page in her hand, and the first clinic takes place in September as planned.
Insight. A policy page and a named deputy turn a personal project into a school system.
Common pitfalls
- Scaling because the pilot 'felt good', without rules set in advance.
- Growing the number of schools faster than the number of trained coaches.
- Asking the board for a large, vague approval instead of one clear decision.
- Letting the programme depend on one person with no written policy.
Decide on a Pilot, Plan Three Years, Brief the Board
- Trios (5 min): each person names one innovation in their school that started well and faded. At what point did it fade and why?
- Trios (20 min): decision exercise. Read the SAMPLE pilot results for a four-school cluster in the Kit. Decide scale, adapt or stop for each school and for the cluster, list the non-negotiables you would write into the handbook, and state the first condition for year two.
- Whole group (10 min): compare decisions. Which school did most groups want to stop or pause, and what did that tell you about the conditions?
- Part B, pairs (25 min): for your own cluster, fill the three-year plan (year 1 pilot, year 2 growth, year 3 embedding) and the budget in hours and money in the Kit. Name at least three low-cost levers and the risk that most threatens year two.
- Part B continued, groups of four (15 min): each person gives the three-minute board briefing using the six-part structure while the others play board members and ask one hard question (cost? risk? evidence? why now?). Feedback: one strength, one cut.
- Alone (10 min): write the leadership task from the Kit and name the person who would take over as coordinator if you left.
Your facilitator will guide you through this activity in the session.
Questions to discuss
- Which non-negotiable would be hardest to defend to your own head or board?
- Where in your three-year plan does the programme depend on one person?
- What did the board question teach you about how to present the cost?
Reflect
- What is the smallest version of this programme you could start next month?
- What have you seen end because it grew too fast?
- What will you say when a leader asks, 'Can we afford it?'
Try it tomorrow
- Write stop / continue / change rules for your pilot or your next term before you collect any results.
- Draw a three-year timeline on one page and mark the first condition each new school must meet.
- Practise a three-minute briefing with a colleague, with the ask in the first sentence.
Further reading and links
Pilot to learn, decide by rules set in advance, scale capacity before reach, keep the non-negotiables, count the hours and plan for the day the champion leaves.