Skip to content
Retour à tous les templates

Artefact Prova

Template d’audit de workflow IA pour équipes marketing

Cartographie un workflow marketing et décide ce que l’IA doit automatiser, augmenter, protéger ou refuser avant que l’équipe s’engage sur un pilote.

Champs du template

  • Workflow stage
  • What AI handles today
  • What AI should handle in six months
  • What stays human and why
  • Ethical or operational risk: low, medium, high
  • Action: automate, augment, protect, decline
  • ICE score for pilot candidates

Exemple rempli

Les exemples remplis ci-dessous restent en anglais, car Prova évalue les artifacts soumis en anglais.

Client reporting workflow for a paid media account.

Data extraction from ad platforms
Metric reconciliation
Anomaly detection
Insight drafting
Strategist review
Client-ready narrative
Follow-up action list

Version faible vs version solide

Version faible

Workflow stageReporting
AI handles todayAI writes reports
Stays humanStrategy
RiskMedium
ActionAutomate

Pourquoi cela échoue

  • "Reporting" is too broad to audit.
  • "Strategy" does not name the judgment point.
  • It does not identify source systems, reconciliation risk, or who approves the client-facing narrative.
  • Automating the whole workflow is unsafe because the work includes interpretation and client trust.

Version solide

Workflow stageWeekly paid media variance explanation for Meta, Google, and LinkedIn
AI handles todayDrafts first-pass variance notes from exported spend, revenue, CPA, and campaign-change logs
AI should handle in six monthsFlags likely causes, drafts three audience-specific summaries, and proposes follow-up checks
Stays humanStrategist approves causal claims and client recommendation because platform data can be delayed or misleading
RiskHigh for client narrative, medium for internal diagnosis
ActionAugment, with protected human approval before client send

Pourquoi cela fonctionne

  • The stage is specific.
  • Source systems are named.
  • The human judgment point is explicit.
  • Risk differs by use case.
  • The action is narrow enough to pilot.

Ce que Prova vérifie et que l’IA générique rate souvent

  • Whether the workflow scope is specific enough to audit
  • Whether "stays human" reasons are concrete or just polite language
  • Whether high-risk stages are incorrectly classified as automate
  • Whether ICE scores are supported by operational reality
  • Whether the selected pilot is narrow enough to test
  • Whether the next sprint should be measurement, rollout, or foundation repair

Prochaine étape

Tu veux un retour sur ta version? Prova commence par une courte évaluation pour aligner le standard de revue avec ton rôle, ton objectif et ta première audience. Ensuite, tu entres dans le sprint adapté à ton travail actuel.

Prova est actuellement disponible uniquement en anglais.

Avant d’envoyer: retire les noms de clients, les chiffres confidentiels et tout élément que ton équipe ne voudrait pas stocker dans un système de formation ou de coaching.

Continuer vers Prova en anglais