A scientific and iterative approach
A strategy is not a document; it is a process. Our method brings together four stages that respond to one another, so that every decision stems from a diagnosis and every action informs the next.
Four stages, one standard
Drawn from the methods of applied research and complex project management, this cycle ensures coherence between what we know, what we decide and what we do.
Strategic diagnosis
Establishing an objective reading of the situation and its dynamics.
Strategic design
Setting out options, deciding and defining a clear trajectory.
Operational deployment
Translating strategy into organisation, message and action.
Evaluation and adjustment
Measuring effects, learning and changing course where needed.
The four stages of our approach
Diagnosis, design, deployment, evaluation: our approach explained step by step.
Emplacement vidéo : collez l’adresse YouTube ou Vimeo dans le widget Vidéo ci-dessous et gardez « Superposition d’image » activée. Cette section reste masquée sur le site tant qu’aucune vidéo n’est renseignée.
Understand before acting
The diagnosis establishes a rigorous baseline: electoral history, socio-demographic structure, existing cleavages, perceptions of actors, organisational strengths and weaknesses. It tests quantitative data against qualitative insight gathered in the field, in order to move beyond received wisdom.
Guiding question: what, objectively, is the situation — and what do we still not know?
Read the detailed pageWork carried out
- Fine-grained analysis of previous electoral results
- Study of cleavages and turnout
- Qualitative interviews and field observation
- Audit of organisation and resources
Typical deliverable
Diagnosis report and map of issues, with explicit assumptions.
Choosing a trajectory
Building on the diagnosis, we construct several contrasting scenarios, assess their feasibility and risks, and support the decisions between them. The chosen strategy is organised around positioning, message architecture, territorial priorities and a reasoned allocation of resources.
Guiding question: of the possible options, which is most coherent with the project and the context?
Read the detailed pageWork carried out
- Building and comparing scenarios
- Defining positioning and the core message
- Prioritising areas and audiences
- Stakeholder mapping
Typical deliverable
Strategic plan, message house and prioritised roadmap.
Bringing the strategy to life
A strategy is only as good as its execution. We support team structuring, the design of communication arrangements, the engineering of public meetings and field coordination. Every event — community meeting, tour, major rally — is designed as an operation: objective, audiences, logistics, safety, message consistency and measurement of its impact.
Guiding question: is the organisation able to carry the strategy, everywhere and over time?
Read the detailed pageWork carried out
- Team training and structuring
- Communication plan and editorial calendar
- Event planning and engineering
- Operational tracking dashboards
Typical deliverable
Operational plan, backward schedule, protocols and steering tools.
Learning continuously
At regular intervals, we measure the effects of actions against the objectives set. Gaps are not failures but information: they lead us to revisit the diagnosis’s assumptions and adjust the design. This constant feedback is what makes the approach iterative.
Guiding question: what have we learned, and what must we change?
Read the detailed pageWork carried out
- Tracking indicators defined in advance
- Periodic strategic reviews
- Analysis of gaps and weak signals
- Structured lessons learned
Typical deliverable
Evaluation notes, adjustment recommendations and final review.
Every figure has a status, and we say so
Confusing a target, an estimate and a result is one of the most common sources of error in political debate. In all our work, every piece of data is explicitly labelled.
| Status | Definition | How we present it |
|---|---|---|
| Target | An aim set to guide action. | Presented as an intention, never as an achievement or a forecast. |
| Calculated estimate | A value derived from a model, a sample or an aggregation. | Always accompanied by its method, sources and uncertainty interval. |
| Measured result | A value observed and verifiable after the fact. | Published only with its source and scope of measurement. |
Understanding citizens, not manipulating them
The tools of computational social science are powerful. Using them carries a responsibility towards citizens and towards the integrity of the democratic process.
- Our behavioural analyses and monitoring serve to understand expectations and adapt a message transparently.
- We neither create nor spread misleading information, and never conceal the origin of a message.
- Adversarial scenario simulations are preparation exercises, never research intended to harm individuals.
- We comply with applicable legal frameworks on personal data and electoral communication.
- Data is collected, stored and processed securely, and limited to what is strictly necessary.
Questions about our method
The full approach offers the greatest coherence, but each stage can be the subject of a targeted engagement — for example an independent diagnosis or a mid-course evaluation — provided the starting assumptions are clearly established.
Because an estimate without a margin of uncertainty creates an illusion of precision. An interval honestly shows the range of plausible values and allows robust decisions in the face of several possibilities.
Each corrects the other. Data reveals trends that intuition does not perceive; the field gives meaning to figures that, in isolation, could be misinterpreted. Our diagnoses systematically combine quantitative and qualitative approaches.
Start with a diagnosis
Every sound strategy begins with an accurate reading of the situation. Let’s talk about yours.