Skip to content
Praestria

Advisory offer — AI & change management

Integrating AI starts with transforming teams

We help energy players move from scattered experiments to AI usage that is governed, adopted by teams and measured over time.

The challenge

AI almost never fails for technical reasons

The models work, the tools exist, the integrations get done. What fails is adoption: licences deployed but unused, teams fearing for their jobs, unofficial usage proliferating without governance, skills leaving with the vendor at the end of the project.

In energy, the paradox is striking. The sector handles considerable volumes of data — load curves, pricing, regulatory flows — and concentrates very high-value AI use cases. Yet organizations move cautiously, between sensitive data, regulatory requirements and teams already under pressure.

Our conviction: AI is a human transformation issue before it is a technological one. That's why we approach it as a consultancy — through usage, skills and organization — not as a vendor, through the tool.

Our approach

Five phases, from diagnosis to anchoring

  1. 01

    AI maturity assessment

    Two to three weeks to objectively establish your starting point: existing usage (including informal), data quality and accessibility, available skills, team appetites and fears, regulatory constraints. Board-level restitution with a maturity matrix and three actionable recommendations.

  2. 02

    Business use-case mapping

    Team-by-team workshops to identify use cases in your vocabulary: tender analysis, regulatory watch, load curves, credit scoring, sales administration, disputes. Each case is assessed on three axes — value, feasibility, risk — and prioritization is arbitrated with you, not for you.

  3. 03

    Framed pilot

    A restricted scope, volunteer users, success criteria defined before starting. Six to eight weeks in the field with before/after measurement: time spent, quality produced, user sentiment. The pilot is as much about learning as convincing.

  4. 04

    Roll-out

    Industrializing what has proven itself: integration into everyday tools (CRM, email, documents), access and data management, progressive scale-up team by team. Nothing is deployed without its enablement component.

  5. 05

    Anchoring

    The full change-management programme — detailed below — so usage outlives the project: training, champions, measurement, governance.

Change management at the core

Five workstreams to convert the try

Most players “do AI” by selling technology. Very few commit to adoption. Yet that is where everything plays out — and that is where we put our resources.

1

Awareness

Demystification workshops tailored to each population — executive committee, managers, operational staff. What AI does, what it doesn't, and what it concretely changes in each person's job, with demonstrations on your own cases. Goal: replace diffuse fear with practical understanding — the precondition for everything else.

2

Role-based training

No generic training: a sales administrator, a pricer and a lawyer don't need the same reflexes. Short, differentiated paths built on each role's real usage, systematically including the limits: bias, hallucinations, sensitive data, output verification.

3

Champions network

In each team, we identify and train volunteer champions: first users, field relays, friction detectors. We animate them during the engagement, then hand over the animation. This is the mechanism that makes the transformation durable after we leave.

4

Adoption measurement

What isn't measured doesn't stick. We track simple, honest indicators: real usage rate per team, autonomy, time saved on targeted tasks, quality produced. The adoption dashboard is a systematic deliverable — it tells you the truth about your transformation.

5

Governance

A clear usage charter (what is allowed, with which data, with what verification), a light validation process for new use cases, an AI committee meeting at the right frequency. Governance reassures leadership, legally secures usage — and prevents the return of shadow AI.

Deliverables

Concrete deliverables, transferred to your teams

AI maturity matrix

Your starting point, objectified: data, tools, skills, usage, governance.

Use-case mapping

Cases identified per team, assessed (value / feasibility / risk) and prioritized in three waves.

AI roadmap

Sequencing, workloads, technical and organizational prerequisites, budget — arbitrated at board level.

AI charter & governance

Usage charter, data rules, validation process, AI committee.

Role-based training plan

Differentiated paths per role, materials included, roll-out calendar.

Adoption dashboard

Usage, autonomy and value indicators — maintained during the engagement, then transferred.

Proof by usage

« We practise what we preach »

AI is not a conference topic for us: it is built into our platform — regulatory text analysis, automated watch, scoring — and into our internal processes. Our consultants use it every day, with its strengths and its limits.

So you are not buying a conviction in principle, but operational experience: what works, what disappoints, how long it really takes, and what it changes in a team. Few consultancies can make that claim.

Where does your organization stand?

The AI maturity assessment is the natural entry point: two to three weeks, a compact format, a board-level restitution. You'll know where you stand — and where to start.

All our advisory services