AI Quick Assessment
Integrating AI into real-world processes first requires clarity: where to intervene, what data are needed, and what value is realistic to expect. The AI Quick Assessment helps reduce uncertainty before investing by identifying the most promising opportunities and prerequisites to get off to a structured start.
The Express level is designed for those seeking quick, structured answers to questions such as:
- “Does it make sense to do something with AI in my company/institution, or not?”
- “What are the 2-3 most concrete opportunities?”
- “What are the main obstacles (data, systems, risk) before starting?”
It is the ideal format to start without bureaucracy, align stakeholders and arrive at a clear decision, with an output that can also be used for internal presentations, preparatory to structured planning.
What is included
In a few focused meetings we gather context and constraints, look at the process at the operational level, and build a credible shortlist:
- Clarification of candidate goals and metrics
- Summary map of the chosen process (process/flow)
- Identification of the main applicable IA use cases
- Prioritization with transparent criteria (impact, feasibility, risk, time-to-value)
- “Light” verification of data and systems
- Evidence of critical issues: privacy, operational risk, IT dependencies
Output
A lightweight but “decision-ready” package:
- Slide deck 10-12 pages
- Priority matrix
- Top use cases in the context analyzed
- Final recommendation: what makes sense to pilot and what should be prepared first
What is not included
Light is not a technical project: it does not include quantitative data analysis, mining, development or integration. It is used to quickly decide whether and where to invest.
The Intensive level is for those who want to immediately structure planning and establish a solid foundation for starting a pilot. It is suitable when:
- You already have pressure on an operational problem (clinical, manufacturing, etc.) and want a measurable solution
- Want to understand if the data hold up and what it takes to make it usable
- You need a clear plan to share with IT, management, and risk/compliance
Here the goal is not just to choose a use case, but to arrive at a structured output that puts one in a position to start with the design of the IT/data science solution and its implementation: a concrete pilot plan with metrics, prerequisites and roadmap.
What is included
We work extensively on processes, data and governance:
- Insight into key processes (1-2) and more detailed map of how they work today
- Interviews with the right people (clinical/operations, IT/data, risk)
- Identification of possible use cases and prioritization with clear criteria (impact, feasibility, risk)
- Data verification: sources, access, expected quality, what is missing, and dependencies
- Defining metrics and baseline (what we measure today and what needs to improve)
- Definition of how AI would enter into daily work (target workflow)
- First check on risks and compliance (privacy, accountability, tracking, monitoring)
- Pilot plan (8-12 weeks): activities, roles, prerequisites, and estimated effort/budget
Output
An output that allows unambiguous transition to design and implementation:
- Report + essential technical annexes
- “Pilot-ready” tab of selected use case
- Backlog clinical and technical requirements (what to build, what to integrate)
- Roadmap 90 days to get to the pilot
- Definition of success criteria and potential hostile factors
What is not included
Intensive is not yet development: it does not include model building, full integrations or putting into production. It is the phase that avoids costly mistakes before investing in the project.
