Frontline Experiment Tracker & Hypothesis Log

{ "Title": "Frontline Experiment Tracker & Hypothesis Log", "IntroductionHtml": "

Use this form to make small, fast, safe experiments repeatable and learnable. Capture the problem, a measurable hypothesis, how you will collect data, who owns execution, and how you will verify results. Good experiments reduce bias, protect safety, and create reliable evidence that teams can act on.

Quick practical guidance

• Keep your outcome measurable: pick one primary metric (throughput, cycle time, scrap %, downtime minutes, etc.).
• Define success up front: state a clear numerical or categorical success criterion.
• Sample-size rules of thumb: For simple shopfloor A/B checks with proportions (e.g., defect rate), aim for at least 30–50 events per group as a minimum pilot. For continuous measures (time, output), aim for 20–30 independent observations per condition for a rough check. When expected effects are small or decision risk is high, consult a statistician or escalate before acting.
• Control bias: document how you will keep conditions comparable, who collects data, and how measurements are taken.
• Safety first: stop immediately if a safety or quality risk appears. Escalate high-risk changes to safety or regulatory owners.

This tracker saves experiment records so teams can review outcomes, avoid repeating mistakes, and scale successful changes.

", "SubmitLabel": "Save Experiment", "SuccessMessage": "Experiment saved. You can reopen to update results, attach verification evidence, or run another experiment.", "DataType": "experiment-tracker", "SchemaVersion": "1.0", "Fields": [ { "Key": "title", "Label": "Experiment title (short)", "FieldType": "text", "HelpText": "A concise name that helps others find this experiment.", "Required": true }, { "Key": "problem_statement", "Label": "Problem statement", "FieldType": "textarea", "HelpText": "What problem are you trying to solve? Who or what is affected, and what is the current impact?", "Required": true }, { "Key": "baseline_metric_name", "Label": "Baseline metric (name)", "FieldType": "text", "HelpText": "Name the primary metric you will measure (e.g., 'first-pass yield', 'cycle time (s)', 'downtime minutes').", "Required": true }, { "Key": "baseline_value", "Label": "Baseline value", "FieldType": "number", "HelpText": "Current measured value for the baseline metric. Use the same units you will measure during the experiment.", "Required": true }, { "Key": "baseline_period", "Label": "Baseline measurement period", "FieldType": "text", "HelpText": "Describe how the baseline was measured (dates, shifts, sample size).", "Required": false }, { "Key": "hypothesis", "Label": "Hypothesis (if then because)", "FieldType": "textarea", "HelpText": "State the hypothesis in the format: 'If we [change], then [measurable effect] because [reason].'", "Required": true }, { "Key": "primary_outcome_metric", "Label": "Primary outcome metric (restate)", "FieldType": "text", "HelpText": "Repeat the exact metric you will use to judge the experiment's success.", "Required": true }, { "Key": "success_criteria", "Label": "Success criteria", "FieldType": "textarea", "HelpText": "Specify the numeric or categorical threshold that will count as success (e.g., 'reduce scrap from 4% to <2.5% over two weeks').", "Required": true }, { "Key": "experiment_design", "Label": "Experiment design (what you will change and how)", "FieldType": "textarea", "HelpText": "Describe the experimental action(s), control vs variant, timing, who performs the steps, and any process parameters to hold constant.", "Required": true }, { "Key": "design_type", "Label": "Design type", "FieldType": "radio", "HelpText": "Choose a simple design type for clarity.", "Required": true, "Options": [ {"Value": "side_by_side", "Label": "Side-by-side A/B"}, {"Value": "before_after", "Label": "Before vs after"}, {"Value": "step_change", "Label": "Step change (single condition)"}, {"Value": "other", "Label": "Other (describe below)"} ] }, { "Key": "sample_size_estimate", "Label": "Planned sample size (per group if A/B)", "FieldType": "number", "HelpText": "Enter the number of observations or events you plan to collect. Follow the guidance above and escalate for small effects or high risk.", "Required": true }, { "Key": "sample_size_basis", "Label": "Sample size basis", "FieldType": "select", "HelpText": "How did you decide the sample size?", "Required": true, "Options": [ {"Value": "rule_of_thumb", "Label": "Rule-of-thumb heuristic"}, {"Value": "historic_variation", "Label": "Based on historic variation"}, {"Value": "calculator_or_statistician", "Label": "Power calculation / statistician"}, {"Value": "operational_limit", "Label": "Operational time limit"}, {"Value": "other", "Label": "Other (explain)"} ] }, { "Key": "period_start", "Label": "Planned start (approx.)", "FieldType": "text", "HelpText": "Date or shift when you intend to start.", "Required": false }, { "Key": "period_end", "Label": "Planned end (approx.)", "FieldType": "text", "HelpText": "Date or shift when you intend to stop.", "Required": false }, { "Key": "data_collection_plan", "Label": "Data collection plan", "FieldType": "textarea", "HelpText": "Who collects the data, how often, which tools or forms, and how will you ensure measurement consistency?", "Required": true }, { "Key": "owner", "Label": "Experiment owner (name & role)", "FieldType": "text", "HelpText": "Person responsible for execution and record accuracy.", "Required": true }, { "Key": "expected_benefit", "Label": "Expected benefit (qualitative and quantitative)", "FieldType": "textarea", "HelpText": "Estimate the expected impact (savings, quality improvement, time saved) and any qualitative benefits.", "Required": false }, { "Key": "risk_assessment", "Label": "Risk assessment", "FieldType": "textarea", "HelpText": "List potential safety, quality, regulatory, or production risks and how you will mitigate them.", "Required": true }, { "Key": "safety_review_required", "Label": "Safety/regulatory review required?", "FieldType": "yesno", "HelpText": "If yes, secure approval before starting and attach evidence in your local system.", "Required": true }, { "Key": "contingency_plan", "Label": "Contingency plan (if things go wrong)", "FieldType": "textarea", "HelpText": "How you will revert or pause the experiment if safety or unacceptable quality issues occur.", "Required": true }, { "Key": "verification_checklist", "Label": "Verification checklist (select items you will complete before closing)", "FieldType": "checkbox", "HelpText": "Mark items you will verify and sign off.", "Required": true, "Options": [ {"Value": "data_collected_as_planned", "Label": "Data collected as planned"}, {"Value": "sample_size_reached", "Label": "Planned sample size reached"}, {"Value": "measurements_consistent", "Label": "Measurements taken consistently"}, {"Value": "no_safety_incidents", "Label": "No safety incidents or stopped and escalated"}, {"Value": "results_reviewed_by_owner", "Label": "Results reviewed by owner"}, {"Value": "decision_documented", "Label": "Decision and next steps documented"} ] }, { "Key": "verification_signoff", "Label": "Verification signoff (name & date)", "FieldType": "text", "HelpText": "Who verified the checklist and when?", "Required": true }, { "Key": "outcome_summary", "Label": "Outcome summary / results", "FieldType": "textarea", "HelpText": "Summarize the measured results versus the baseline and whether success criteria were met.", "Required": false }, { "Key": "lessons_learned", "Label": "Lessons learned and recommendations", "FieldType": "textarea", "HelpText": "What did you learn? Should this change be adopted, adapted, or abandoned? Any follow-up experiments?", "Required": false }, { "Key": "next_steps", "Label": "Next steps / owner for follow-up", "FieldType": "textarea", "HelpText": "Assign follow-up actions with owners and deadlines.", "Required": false } ] }