G/01AI readiness checklist
Ten questions on data, processes, people and security to answer before any AI project.
Steps
- 1Inventory your data sources and their owners
- 2List the three most repetitive workflows
- 3Name the KPI each project should move
- 4Check privacy and compliance constraints (Law 25, sector rules)
G/02Discovery playbook
Our four-meeting method, from company goals to a prioritized roadmap.
Steps
- 1Discovery: goals, challenges, operations
- 2Process analysis: workflows, tools, inefficiencies
- 3Optimization: improvements, integrations, automation, AI
- 4Recommendations: priorities, roadmap, next steps
G/03Automation audit
Find the manual work worth automating and estimate the return before building.
Steps
- 1Log every manual handoff between systems for a week
- 2Time each task and multiply by frequency
- 3Score feasibility: API available, rules clear, exceptions rare
- 4Automate the top three, measure, repeat
G/04Data strategy roadmap
From scattered spreadsheets to a governed platform that feeds models and dashboards.
Steps
- 1Define the unified data model for your core entities
- 2Centralize sources into one warehouse or lake
- 3Add quality checks, lineage and access control
- 4Expose governed metrics and forecasts to teams
G/05Security & compliance for AI
Keep data private and auditable when deploying models, on-premise or in a sovereign cloud.
Steps
- 1Classify data and decide what may leave your perimeter
- 2Prefer on-premise or sovereign hosting for sensitive workloads
- 3Log prompts, outputs and access for audit
- 4Review vendors against ISO 27001 / SOC 2 practices
G/06Chatbot launch playbook
How we take a support assistant from pilot to 24/7 production without losing the human touch.
Steps
- 1Pick the three question types that fill your queue
- 2Connect catalogue, orders and help desk so the bot can act
- 3Define escalation rules and hand off with full context
- 4Measure response time, satisfaction and cost weekly
