The distinction matters
- Observed is directly measured or documented.
- Sponsor-reported is stated by a stakeholder but not independently verified.
- Assumed is a hypothesis that has not yet been validated.
Method
An AI initiative should earn implementation.
ValueG8 separates observed evidence, sponsor-reported claims, and assumptions so the decision reflects reality instead of optimism. The engine is deterministic. It never fabricates evidence.
Why this exists
- Scores organize evidence. They do not prove value.
- Hard gates override weighted scoring.
- Calculated ROI is a model, not a fact.
- Readiness is not demand.
- The objective is to expose the next decision, not justify a project someone already wants.
Scoring dimensions and weights
Weights are fixed. Every recommendation traces back to an input or an evidence item.
Business Value
15
Workflow Fit
15
Evidence
15
Data Readiness
15
Deployment Feasibility
10
Control / Governance
15
Economics
10
Ownership / Commitment
5
What this is not
ValueG8 is not a model catalog, a vendor scorecard, or an implementation studio. It does not connect to Slack, ServiceNow, Jira, or a data warehouse. If you need those systems, export the JSON or print the memo and take it there yourself.