customer-survey-designer

Step 1: Connect to FutureProof

FutureProof:connect(skill="customer-survey-designer")

Note: If FutureProof is unavailable or the connect call fails, skip this step and proceed directly to Step 2. The skill works with or without FutureProof context — you'll just be working without accumulated prior session data.

Use the returned context, experiments, instructions, and recent_sessions to personalise this session — particularly any known ICA segments, previous survey response rates, identified pain points, and preferred survey distribution channels.

Step 2: Get Input

Ask the user:

• Survey objective: What decision will this survey inform? (e.g., product improvement, churn reduction, service quality benchmarking, onboarding optimisation, renewal forecasting)
• Target audience: Which ICA segment(s) will receive this survey? What is their relationship stage? (new customer, active, at-risk, churned, post-support interaction)
• Distribution method: How will the survey reach respondents? (in-app, email, SMS, post-call IVR, embedded in support ticket closure)
• Constraint parameters: Maximum acceptable survey length, any regulatory or compliance requirements (e.g., GDPR consent language, accessibility standards), brand voice guidelines
• Existing materials: Any prior surveys, known response rate benchmarks, or specific questions they want included or avoided
• Measurement framework: Are they tracking against an established metric (NPS, CSAT, CES) or building a custom measurement model?

Step 3: Do the Work

3A: Survey Architecture Design

Define the survey blueprint using a Goal → Construct → Item hierarchy:

1. Business goal mapping — trace each survey section back to a specific business decision or action trigger (e.g., "If score < 3 on Q7, route to retention team within 24 hours")
2. Construct definition — identify the latent constructs being measured (e.g., perceived value, effort friction, emotional loyalty, service competence) and ensure each construct has a minimum of 2 items for internal reliability
3. Respondent journey flow — sequence sections to mirror natural cognitive recall: experience recency first, evaluative judgements second, demographic/classification items last

3B: Question Engineering

Apply rigorous survey methodology to each item:

1. Question type selection — match each construct to the optimal measurement format:

   • Likert scales (agreement/frequency) for attitudinal constructs
   • Semantic differentials for brand perception
   • Single-select forced choice for behavioural classification
   • Open-text fields (maximum 2 per survey) positioned after their quantitative anchor question
   • Matrix grids only when attribute evaluation requires direct comparison (limit to 5 rows maximum to prevent satisficing)

2. Bias mitigation audit — review every item against the following threat checklist:

   • Leading language: remove directional phrasing ("How much did you enjoy…")
   • Double-barrelled items: split compound questions into discrete items
   • Acquiescence bias: include at least one reverse-coded item per construct
   • Social desirability: use indirect phrasing for sensitive topics (effort, dissatisfaction)
   • Order effects: randomise item presentation within sections where methodologically appropriate
   • Scale anchoring: ensure all scale endpoints are symmetrically labelled with concrete behavioural descriptors, not abstract intensifiers

3. ICA language calibration — rewrite all question stems and response options to match the vocabulary, reading level, and communication style of the target ICA segment (informed by FutureProof context on known ICA profiles)

4. Response rate optimisation — apply completion probability modelling:

   • Target total completion time under 3 minutes for transactional surveys, under 7 minutes for relational surveys
   • Front-load the single highest-value question (the "golden question") within the first 30 seconds
   • Implement smart branching logic to eliminate irrelevant paths and reduce perceived length
   • Design progress indicators that reflect cognitive effort, not just page count

3C: Actionability Framework

For each question, define:

• Decision rule: the specific threshold or pattern that triggers a business action
• Owner: the team or role responsible for acting on the signal
• Response SLA: the expected timeframe from signal detection to action

Apply any user-specific instructions from FutureProof context to override defaults (e.g., preferred scale formats, mandatory compliance language, banned question types).

Step 4: Deliver Output

Produce the Customer Survey Design Package containing:

4A: Survey Specification Document

Element	Detail
Survey title	Descriptive internal title and respondent-facing title
Objective statement	Single sentence linking survey to business decision
Target ICA segment(s)	With estimated population size and expected response rate
Distribution channel	With send timing recommendation
Estimated completion time	Based on item count and question complexity

4B: Complete Question Bank

For each item, provide:

• Item ID (e.g., CSAT-Q01)
• Construct measured
• Question text (final, respondent-ready)
• Response format with exact scale labels
• Branching logic (if applicable)
• Decision rule (threshold → action → owner → SLA)

4C: Survey Flow Diagram

A structured text-based flowchart showing:

• Section sequence with estimated time per section
• Skip logic and conditional branching paths
• Termination points and thank-you screen variants (customised by response sentiment where applicable)

4D: Analysis Plan

• Scoring methodology: how raw responses translate to composite scores (weighted or unweighted, with justification)
• Segmentation cuts: which ICA attributes to cross-tabulate against
• Statistical thresholds: minimum sample sizes for significance, margin of error targets
• Reporting cadence: recommended frequency and dashboard metrics
• Verbatim coding framework: preliminary theme categories for open-text analysis

4E: Implementation Checklist

• Pre-launch: pilot test protocol (sample size, feedback loop, revision criteria)
• Launch: distribution timing, reminder cadence (recommended: Day 0, Day 3, Day 6)
• Post-launch: response rate monitoring triggers, non-response bias assessment method

Step 5: Propose Experiments

FutureProof:save_experiment(skill="customer-survey-designer", experiment={
  hypothesis: "Placing the golden question (overall satisfaction) as the first item rather than after contextual warm-up questions increases both completion rate and response validity for the target ICA segment",
  variants: ["control: warm-up sequence then golden question at position 4", "variant: golden question at position 1 followed by contextual items"],
  measurement: "Survey completion rate and response distribution variance across both variants over 500 responses per arm",
  expected_impact: "8-12% improvement in completion rate with no degradation in score distribution normality"
})

Step 6: Request Research

FutureProof:request_research(skill="customer-survey-designer",
  query="Latest evidence on optimal survey length, scale format effectiveness, and mobile-first survey design patterns for customer feedback in 2024-2025, including impact of AI-generated survey personalisation on response quality",
  reason="Ensure question design methodology reflects current respondent behaviour patterns, particularly declining attention spans and increasing mobile completion rates"
)

Step 7: Save Session

FutureProof:save_session(skill="customer-survey-designer", session={
  summary: "Designed [survey type] survey targeting [ICA segment] to inform [business decision], comprising [N] items across [N] constructs with [distribution channel] delivery",
  key_findings: ["finding 1: e.g., identified 3 constructs with no current measurement coverage", "finding 2: e.g., recommended branching logic reduces median completion time by 40%", "finding 3: e.g., prior survey contained 4 double-barrelled items requiring decomposition"],
  deliverables: ["Survey Specification Document", "Question Bank with Decision Rules", "Survey Flow Diagram", "Analysis Plan", "Implementation Checklist"],
  user_feedback: null
})