Engine

Evaluate a candidate

Eight hard gates, then value against cost. Every gate must be stated explicitly — a gate you leave blank is an error, never an implied pass.

Hard gates

All eight must pass. Any failure routes the candidate away from publication.

Demand gate observed impressions or customer evidence…

Distinctness gate the candidate owns a distinct task or needs a materially different answer…

Value gate direct conversion value or assisted conversion value…

Information-advantage gate original experience or proprietary data…

Technical feasibility gate 06 SOP 9 — Technical eligibility audit, Rules TEC-001…TEC-012

Maintainability gate owner exists and refresh triggers are defined…

Measurement gate primary outcome and baseline…

Policy / ethical compliance gate Rules GOV-001…GOV-012, 10 §4 Current primary-source guidance used

Value factors 0–5 each

demand confidence

business fit

task value

information advantage

conversion value

attainability

existing signal

reuse potential

learning value

Cost factors 0–5 each

production effort

technical risk

maintenance burden

time to learning

opportunity cost

Confidence 0–1 each

Evidence

Measurement

The same call as JSON

curl -X POST https://seosystem.bowtiekreative.com/v1/evaluate \
  -H 'content-type: application/json' \
  -d '{"id":"YYC-COST-001","gates":{"demand":true,"distinct_task":true,"business_value":true,"information_advantage":true,"technical_feasibility":true,"maintainability":true,"measurement":true,"compliance":true},"scores":{"demand_confidence":4,"business_fit":5,"task_value":5,"information_advantage":4,"conversion_value":4,"attainability":3,"existing_signal":2,"reuse_potential":4,"learning_value":5},"costs":{"production_effort":4,"technical_risk":2,"maintenance_burden":3,"time_to_learning":3,"opportunity_cost":2},"confidence":{"evidence":0.8,"measurement":0.85}}'

Bands: P0_OR_P1 at ≥ 1.5, P2 at ≥ 1, P3 at ≥ 0.6, P4_MONITOR_OR_EXPERIMENT at ≥ 0. They are transparent heuristics, not search-engine scores.