The note to the executive team
Proposed decision: adopt a doctrine of proof common to all speeches on industrial transformation.
The problem to solve
An investment announcement, an intention to decarbonize and an actually measured improvement do not designate the same state of the project. However, they are often found united in the same promise. The risk appears when employees, local residents or customers understand an acquired result where the company describes a conditional trajectory.
Recommended arbitration
The steering committee must validate three levels of formulation: what has been achieved, what is committed and what remains targeted. Each public commitment receives an operational manager, dated proof and a review condition. The communications department organizes this consistency; it does not certify technical data alone.
What management should decide
Give a coordination mandate to communication with operations, purchasing, finance and public affairs. Allow key dependencies to be published without revealing sensitive information. Plan a decision point before each structuring announcement and follow-up on commitments after publication.
The transition takes place in the dependencies
An organization rarely controls the calendar it communicates on its own. Credibility begins with a precise description of what it controls.
An investment does not erase a vulnerability
The IEA 2026 panorama highlights the continued concentration of refining and the importance of trade restrictions for critical mineral chains [1]. For communication, the lesson is concrete: an announcement of future capacity must be compared with the real availability of inputs, equipment and skills. This sectoral observation does not allow us to conclude that a company is exposed without examining its own contracts. [1]
Three calendars to make compatible
The industrial calendar follows studies, authorizations, construction and commissioning. The financial calendar tracks investment decisions and their conditions. The public calendar is triggered as soon as a territory, a client or a media interprets an announcement. When these rhythms diverge, communication must explain the stages, not artificially shorten the duration of the project.
The proof must follow the perimeter
A drop observed in a process does not necessarily apply to an entire product, a site or the group. An improvement in intensity can coexist with an increase in total volume. Before constructing a story, it is therefore necessary to specify the unit, the scope, the period and the method. Presenting these limits protects understanding; it takes nothing away from the value of demonstrated progress.
Build the commitment register
A communication board is not enough. There needs to be a chain of accountability that links each formulation to a decision and evidence.
Start with the stated statements
Identify the commitments visible in management speeches, press kits, the website, commercial presentations and recruitment documents. Prioritize those that combine high exposure and low demonstration capacity. The number of formulations counts less than their importance for the decisions of the audiences concerned.
Making validation workable
For each statement, record an owner of the data, the exact wording authorized, the source, the date and the events that make an update necessary. A validation must result in a clear state: publishable, publishable under conditions, or to be suspended. Avoid collective validations where no one can then explain the reasoning.
Prepare for internal disagreements
If operations deems a schedule uncertain and business development wants a firm date, the committee must explicitly arbitrate the acceptable level of commitment. The communication sets out possible interpretations and proposes a formulation. It does not turn an unresolved disagreement into a deliberately vague sentence.
| Project status | Minimal evidence | Useful wording |
|---|---|---|
| Realized | Measured result and scope | Describe the observed result |
| Engaged | Decision, means, steps | Describe what is being launched |
| Aimed | Assumptions and dependencies | Present a conditional ambition |
Test three scenarios before the announcement
These scenarios are illustrative exercises. They do not describe an actual company or an industry forecast.
Scenario 1 — A supplier becomes unavailable
Critical equipment arrives late. The teams know that the announced commissioning is compromised, but no revised schedule is yet reliable. The right answer starts with the confirmed impact, continuity measures and the next reassessment deadline. The committee decides who should be informed first, which contractual commitments are affected and when public communication becomes necessary.
Scenario 2 — The territory contests the benefit
A project presented as creating value is questioned about water, nuisance or local employment. Responding solely with one's national interest leaves the lived question intact. It is necessary to specify the expected effects at the territorial level, document the uncertainties and identify what can still change. The consultation must focus on real choices; the CNDP distinguishes public influence on the decision and final responsibility of the decision-maker [2]. [1]
Scenario 3 — A claim is disputed
A third compares an environmental promise to data published elsewhere by the group. The response unit reconstitutes the perimeters before contesting the criticism. If the ambiguity comes from the company, it corrects the wording and explains the change. If the perimeters differ, it shows this difference without automatically attributing a hostile intention to the interlocutor.
Territorial dialogue is decision-making work
The quality of a relationship is not measured by the number of meetings organized, but by the way in which the questions received are handled.
Identify expectations that cannot be replaced
Employees ask what is changing in the work and skills. Communities look at infrastructure and local effects. Residents question the nuisances and remedies. Customers expect continuity and demonstrable characteristics. A general presentation can open the discussion, but it must lead to appropriate responses, based on the same factual basis.
Show room for maneuver
Distinguish between what has been decided, what is still open and what depends on another authority. This distinction avoids promising a negotiation on a fixed element or, conversely, presenting an option that is still debatable as irreversible. It gives public affairs a role in understanding institutions and constraints, without confusing dialogue with automatically obtaining support.
Close the response loop
Keep a record of questions, answers given and commitments made. Unresolved topics should remain visible with a deadline and a person responsible. A change of project can improve the relationship if it is explained; prolonged silence on a known issue can deteriorate it even when the work is progressing correctly. The committee receives the blocked points, not just a report of the meetings.
Media, governing voice and visibility in AI
The same industrial reality must remain intelligible when it becomes an interview, a web page or a response generated by an AI.
Give the media verifiable objects
Prepare a sheet per announcement: decision taken, timetable, scope, resources committed, dependencies and technical contact. A site visit or an exchange with an expert is useful when it allows you to verify a specific statement. It does not replace the documentation. The manager's words must explain the choices and their consequences, without replacing technical qualification.
Build a coherent public base
The GEO program here consists of making reference information accessible, assigned and updated. The pages must distinguish between groups, subsidiaries, sites and projects, indicate their dates and link announcements to updates. An old ambition should not remain the only document consultable after a change of calendar. SEO and useful content remain relevant foundations in Google's AI functionalities [3]. [1]
Measure accuracy rather than presence alone
Create a small set of questions representative of employees, partners and territories. Keep the date, the service questioned, the language and the response obtained. Identify scope errors, obsolete data and cited sources. The results are limited observations: they do not predict all future responses and do not guarantee any citations. The correction first concerns the controllable information.
A 90-day work program
The proposed calendar is a method of organization. It must be adapted to the size of the project, applicable obligations and available resources.
Days 1-30 — Establishing the Base
Communications and operations map priority promises and identify missing evidence. Purchases describe sensitive dependencies; finance confirms the status of investments. Public affairs maps administrative decisions and interlocutors. The expected result is a validated register, not a new slogan.
Days 31 to 60 — Testing the Word
Organize an exercise on a project delay and a public protest. Have different teams answer the same question, then compare the facts, dates and commitments expressed. Correct discrepancies in reference sheets and the validation process. Also test the circuit for escalating questions from the field: useful information must reach the decision-maker before becoming a media surprise.
Days 61 to 90 — Publish and install tracking
Update the most exposed pages, prepare spokespersons and publish any evidence that can be published. Establish a regular review of commitments and a correction mechanism. The first report to the committee presents the inconsistencies resolved, those that persist and the expected decisions. It specifies the limits of the measurement and the means necessary to maintain the method over time.
What the committee should follow
The dashboard should show the strength of the word and the decisions still needed.
Four signals to look at
Monitor commitment register coverage, evidence that has become obsolete, response time to sensitive questions and wording gaps between channels. Present the volumes and their calculation basis: a percentage without scope can mask a weakness in a major commitment.
What these measurements do not prove
Accurate content does not automatically lead to membership. Local disagreement is not necessarily a communication failure. The measures are used to detect a problem in understanding, consistency or response; industrial choices and their impacts remain to be discussed in substance.
The decision to be recorded in the minutes
Validate the register manager, the review of priority commitments and the public correction rule. Ask the next committee for an example of a consolidated promise and an example of voluntarily revised wording. These two cases make the change in practice observable.
Sources and benchmarks
- IEA — Global Critical Minerals Outlook 2026, Executive summary [1]
- National Commission for Public Debate — Frequently asked questions [1]
- Google Search Central — AI features and your website [1]
Belief System analysis and recommendations. The scenarios are illustrative. The sources illuminate the context; they do not constitute validation of our recommendations.