Pilot Planning & Scale Template

Purpose and how to use this template

This template helps teams design pilots that generate usable evidence and a clear path to scale. Use it during planning to surface assumptions and measurement needs, during execution to keep the experiment disciplined, and during closeout to make a go/no-go decision and prepare for scale.

Tips: keep each section concise; assign owners for each field; capture evidence in the data collection plan so the evaluation is auditable; keep the core hypothesis tight and measurable.

1. Objective and success criteria

Define what success looks like in operational terms.

• Pilot objective: (one sentence – what problem are we trying to solve?)
• Primary success metric: (the one metric that determines pilot success – e.g., % defect reduction, mean time to repair, on-time delivery)
• Target threshold: (numeric target and timeframe – e.g., reduce defects by 20% within 8 weeks)
• Secondary metrics: (quality, cost, safety, throughput, customer satisfaction, staff time)
• Timebound evaluation date: (when will we evaluate?)

2. Scope and exclusions

Be explicit about what is in and out of scope so results are interpretable.

• Process/area included:
• Shifts/lines/customers included:
• What we will not change: (exclusions that would confound results)

3. Stakeholders and roles

List people and their responsibilities.

• Pilot sponsor:
• Pilot lead / manager:
• Data owner / analyst:
• Operations lead (site/shift):
• Quality / safety / compliance:
• Communications / change lead:

4. Core hypothesis

State a falsifiable hypothesis so the pilot tests something specific.

Example: Implementing standard work for machine setup will reduce setup time by at least 15% and reduce variation in setup time by half.

5. Data collection plan

Be specific about what you will measure, how often, and who collects it.

• Primary metric (definition): how it is calculated and units
• Secondary metrics (definitions):
• Data source: manual log, MES, ticketing system, customer survey
• Collection frequency: shift/daily/weekly
• Data owner: name and contact
• Storage and dashboard: where data will be stored and how stakeholders will view it
• Quality checks: how data integrity will be verified

Sample size guidance (practical rules of thumb): pilots are not full-scale trials. Aim to collect enough observations to see meaningful change while keeping effort small:

• For rate or proportion metrics (e.g., % defects): aim for at least 30–50 independent observations per condition as a practical minimum; more if baseline variability is high.
• For averages (e.g., time, cost): 20–50 measurements per condition often shows directional effects; use statistical sample-size tools when precision matters.
• If events are rare, pilot duration may be more important than sample count—prepare to extend the pilot or use proxy leading indicators.
• When in doubt, run a short baseline period to estimate variability, then revise sample needs.

6. Timeline and milestones

High-level timeline with owners.

• Planning & approvals (dates, owner)
• Baseline data collection (dates)
• Pilot launch (date)
• Interim review(s) (dates & criteria)
• Final evaluation (date)
• Scale decision meeting (date & attendees)

7. Risk register

Capture key risks, likelihood, impact, and mitigation plans.

1. Risk: (short description)
2. Likelihood: Low / Medium / High
3. Impact: Low / Medium / High
4. Mitigation: (actions and owner)

8. Evaluation metrics & analysis plan

Explain how you will determine if the pilot met success criteria.

• Primary analysis method: simple before/after comparison, control group comparison, time-series, or statistical test
• Who will perform analysis and produce the report
• What constitutes a meaningful improvement beyond random variation (absolute/relative thresholds)
• Uncertainty and sensitivity: what would change interpretation (e.g., seasonal effects, concurrent interventions)

9. Decision gate criteria

Explicit go / adapt / stop rules make decisions objective.

• Go: Primary metric meets target AND no unacceptable safety/regulatory issues
• Adapt (iterate): Primary metric shows improvement but below target or secondary issues need resolution
• Stop: No improvement and/or negative impacts on safety, quality, or cost
• Document who signs the decision and what information they review

10. Scale-up checklist

If the decision is Go, use this checklist to prepare broader deployment.

• Define scale scope (sites/lines/shifts/customers)
• Confirm standard operating procedures and job aids are documented
• Training plan and materials ready; trainers assigned
• Required materials, tools, software, and spare parts procured
• Data collection and dashboards configured for scale
• Ownership and handover: site leads and support teams assigned
• Monitoring plan: KPIs, cadence, escalation paths
• Change management & communication plan completed
• Post-scale review date scheduled to validate sustained benefit

11. Example (short)

Objective: Reduce packaging line changeover time by 15% in 6 weeks. Primary metric: average changeover minutes per event. Scope: one line, day shift. Baseline: collect 3 weeks of baseline to measure variability. Decision gate: if average time reduced by >=15% and defect rate unchanged, proceed to expand to other shifts.

12. Common pilot pitfalls and how to avoid them

• Pitfall: Vague hypothesis — Fix: make success metric explicit and measurable.
• Pitfall: Poor baseline or no control — Fix: run a short baseline to measure variability or use a comparable control group.
• Pitfall: Data not collected reliably — Fix: assign a data owner and automate collection where possible.
• Pitfall: Focusing on a single bright result — Fix: check secondary metrics (safety, quality, workload).
• Pitfall: No plan for scale — Fix: include the scale checklist and assign resources during planning.

13. How to capture and share evidence

Attach or link to: baseline data, daily/weekly dashboards, photos/videos of new work, training materials, the final evaluation report, and the approval record for the scale decision.

Final note: Treat the pilot as a learning experiment. A transparent failure with clear lessons and a documented adaptation is better than an ambiguous result that is later scaled and fails. Use this template as a starting structure—adapt fields to your industry, regulatory needs, and operating rhythms.