Scenario 1 — Technical Architecture Decision (Monolith vs. Microservices)

Prompt
A software team must choose between microservices and a monolith for a new project.

What a generic AI would say (avoid):
“Both have pros/cons. Microservices scale but add complexity; monoliths are simpler but can get unwieldy.”

Strategic Decision Playbook

Inputs to collect (Context)

• Team size & seniority; DevOps maturity; on-call model
• Time-to-market deadline; release cadence targets
• Scaling profile (users, throughput, hotspots)
• Domain boundaries (DDD), ownership lines, compliance needs

Decision criteria (weight, example)

• Time-to-market (25) • Operability/MTTR (20) • Independent deployability (15)
• Cross-team coupling (15) • Performance at load (15) • Cost (10)

Evaluate options (score 1–5 each, weight × score)

• Monolith: _____
• Microservices (minimal viable set of services): _____

Risk register

• Monolith: eventual module entanglement, slower parallel delivery
• Microservices: distributed tracing/ops overhead, data consistency, infra cost

Recommendation synthesis

• Choose X because {top 3 criteria wins}, with mitigations for {top 2 risks}.

Implementation plan (30/60/90)

• 30d: Architecture spike, golden path, CI/CD baseline, observability SLOs
• 60d: First vertical slice prod-ready; error budgets; runbooks
• 90d: Performance test at target load; de-risk data strategy; hardening

Success metrics

• DORA (deploy frequency, change fail %, MTTR)
• Lead time to first feature; p95 latency & error rate under N RPS
• Cost per request; on-call toil hours/month

Scenario 2 — Marketing Campaign Optimization (Budget to Max ROI)

Prompt
Allocate budget across channels for maximum ROI.

Strategic Optimization Playbook

Data aggregation

• Last 6–12 months by channel: spend, impressions, clicks, conv., CPA, ROAS, LTV, attribution model; seasonality flags; caps/constraints.

Pattern analysis

• Trend/seasonality; diminishing returns curves (spend → marginal CPA/ROAS); lagged effects.

Audience & fit

• Segment (geo, demo, intent); map channels to segments & funnel stages; note creative requirements.

Scenario modeling

• Objective: maximize expected profit (or conversions) subject to total budget + per-channel min/max + brand/saturation guardrails.
• Test: Baseline, +10% Brand, +15% Performance, Creative refresh, New channel pilot.
• Use simple response curves or historical elasticities; sanity-check with last-click and MMM-lite views.

Risk assessment

• Saturation/creative fatigue; tracking noise; supply constraints; brand safety; learning-phase resets.

Optimization iteration

• Pick top scenario; run weekly reallocation with 10–20% move limits until variance stabilizes.

Implementation strategy

• Phase 1 (Weeks 1–2): small reallocations, A/B creative refresh, tighten UTMs & pixels
• Phase 2 (Weeks 3–6): scale winners; cap underperformers; launch 1 pilot channel
• Phase 3 (Weeks 7–8): consolidate gains; renegotiate rates; prep Q4 plan

Success metrics

• Primary: blended CAC or MER target; ROAS ≥ X; LTV:CAC ≥ Y
• Leading: CTR, CVR, CPA trend; creative fatigue index; share of spend in top-quartile cohorts
• Ops: time to implement reallocations; % budget within guardrails

Reusable Artifacts (for both scenarios)

• 1-page Decision Record: Context • Criteria & weights • Option scores • Final choice • Risks & mitigations • Owner & date
• Dashboard: leading indicators (weekly) + outcome metrics (monthly)
• Stop conditions: proceed when confidence ≥ threshold and no critical risk unresolved; else run one more iteration with a new hypothesis.