What Hospitality Can Learn from Commandement par intention

Saturday, 3:00 p.m. The resort is running at 94% occupancy. A 120-guest wedding cocktail reception begins at 4:30 p.m., 140 covers are booked in the restaurant, and rain will probably force the outdoor outlet to close. One hostess and two servers are absent. The operation must protect the 6:30 p.m. arrivals, accommodate resident walk-ins, redeploy the outdoor team, and determine how much capacity can still be sold.

This situation often triggers an improvised meeting and several parallel conversations. The restaurant requests additional staff, Banqueting holds on to its commis, Front Office wants tables protected for hotel residents, and each team optimises its own area. The property has the information, but lacks a common operating framework through which to use it.

At a very different scale, armed forces face a comparable challenge: coordinating specialised units with incomplete information, constrained resources, and limited decision time. The French doctrine of commandement par intention (command by intent) and the German Auftragstaktik centralise intent and the essential trade-offs, while extensively decentralising execution.

The point is not to turn a hotel into a barracks; hospitality already has its hierarchy, uniforms, and briefings. The relevant lesson is an organisation designed to prepare decisions, distribute authority, and sustain the operation when no plan unfolds exactly as expected.

From instruction to intent

An instruction prescribes a method: “Add two servers to the restaurant.” Intent defines an effect: “Between 6:15 and 8:00 p.m., protect the arrival and first-contact experience. No confirmed guest should be left waiting without acknowledgement, and sold capacity must remain compatible with four fully staffed sections.”

The second formulation requires the manager to exercise judgement. They may close a peripheral section, consolidate bookings into contiguous stations, block fifteen covers at 6:45 p.m., transfer a server from the outdoor outlet, and assign a manager to the entrance. If conditions change, the manager retains the objective and adapts the method.

Intent defines the purpose, the effort principal (main effort), and the desired end state. In this case: protect the guest journey during the arrival compression; concentrate the effort principal on reception between 6:15 and 8:00 p.m.; and finish with a stabilised dining room and a viable second seating. The director sets capacity, risk, and the acceptable level of service without personally designing the floor plan.

A backbrief does not ask whether the instruction was understood. The manager restates their assessment, plan, resource requirements, and escalation thresholds. This reveals who understands the operation beyond their own station and allows autonomy to be calibrated against demonstrated judgement.

S1–S6: organising collaboration around the operation

Within a NATO staff, the principal management considerations are grouped into six functional areas. The S1–S6 structure is not an organisational chart to be copied, but six perspectives that must be covered. S1 addresses people: attendance, capability, fatigue, and relief. S2 builds the demand picture: occupancy, reservations, groups, weather, and weak signals. S3 runs current operations. S4 evaluate inventory, glassware, equipment, Stewarding, and logistics. S5 prepares events, contingencies, and future capacity. S6 maintains the POS, reservation systems, operating plans, and information flow. 

All of them should be evaluated daily, weekly, and monthly by the top leadership. For example, in S1 (people), the daily brief covers schedule adjustments, the weekly current staffing levels and needs, and the monthly direction and hirings. 

This framework makes it possible to designate an effort principal. A team with seven different main priorities has in reality none. A team with ten different emergencies has none, and a hotel where all guests are VIPs will actually lead to worse service. 

If the restaurant becomes the priority, the outdoor team reinforces it, Banqueting pre-positions the required glass racks, and the Assistant F&B Manager remains mobile between the entrance, kitchen pass, and bar. The other outlets continue to operate, but their local decisions temporarily align with the department’s collective outcome.

Information must lead to a decision down the line

Briefings easily become inventories: occupancy, covers, VIPs, absences, groups, and shortages. These inputs are necessary; reciting them does not constitute command.

Management expects four elements: the current situation, its likely development, the operational impact, and a recommendation. “Three absences, 140 covers, and rain” describes a problem. “The peak cannot be operated across six sections. I recommend consolidating into four sections, transferring two team members, and blocking fifteen covers at 6:45 p.m., with capacity reopened if the replacements arrive” prepares a decision.

Management must also define what requires immediate escalation: an accident, a food-safety risk, payment-system failure, waiting time beyond the agreed threshold, an absence affecting a critical function, an employee-relations issue, or a reputational risk. Everything else should be resolved at the appropriate level and then recorded in a shared main courante (casual report).

An ordre préparatoire provides direction before the final plan is issued: the risk of closing the outdoor outlet, the anticipated effort principal, the checks required, and the time of the next update. Under the one-third/two-thirds rule, the senior level uses no more than one third of the available preparation time and leaves two thirds to the teams. An 80% plan issued early allows each outlet to build and test its execution; a perfect plan delivered at 4:00 p.m. leaves only improvisation. This is especially true in banqueting. 

Task, Scope and Purpose: providing direction without prescribing the method

Management by intent requires a clear distinction between what must be achieved and why it matters. For an immediate operation, Task and Purpose will usually be sufficient. The Task defines the observable result assigned to the manager; the Purpose identifies the effect that result must produce within the wider operation.

“Reorganise the floor plan” remains an incomplete instruction. “Remove fifteen covers from the 6:45 p.m. inventory and consolidate the dining room into four sections, in order to keep waiting time below ten minutes and protect the second seating” links a task to its purpose. If replacement staff arrive or bookings change, the Restaurant Manager can adjust the configuration while preserving the intended effect. The Task directs action; the Purpose provides the criterion by which that action may be adapted.

For a broader mandate, Scope and Purpose define the decision space. The Scope establishes what is included in the assignment, what is excluded, the resources available, the constraints, and the level of delegated authority. The Purpose states the business or operational change expected. A cocktail-menu redesign could therefore cover twelve drinks, retain the currently listed spirits, and require no additional headcount; its Purpose would be to shorten guest decision time, encourage a second order, and improve gross profit per cover. The manager remains free to shape the offer but cannot quietly extend the project into glassware, procurement, or bar organisation.

This distinction creates a straightforward escalation rule. As long as circumstances change but the Purpose remains achievable within the authorised Scope, the manager adapts the plan. When the Purpose can no longer be achieved, or the proposed solution falls outside the Scope, the decision is escalated. Commandement par intention therefore does not provide less direction. It defines the outcome, rationale, and boundaries precisely enough for execution to remain autonomous.

Planning for situation dégradée

In F&B, the live operation is almost always operating under some degree of degradation: an incomplete delivery, a table that lingers, a slowing glasswasher, or a late absence. Management must therefore determine when the offer will be reconfigured and under which rules.

In a situation dégradée, the military preserves essential capabilities despite the loss of an asset and temporarily reduces secondary activity. A hotel can do the same. With two absences, the restaurant closes one section and blocks a reservation slot. During a POS outage, it uses numbered manual checks and a single order-entry point. If the main wash-up falls behind, Banqueting protects its dedicated stock and the bar simplifies its glassware range. The degradation becomes explicit, prioritised, and reversible; the operation no longer promises nominal service with inadequate resources.

Each threshold has a pre-authorised response: if projected waiting time exceeds fifteen minutes, reduce capacity; if the entrance is not covered by 5:30 p.m., redeploy a manager; after ten minutes of system downtime, activate the manual process. The manager acts without seeking approval that has already been granted in advance through business continuity plans.

Management must also retain an operational reserve: a supervisor without a fixed section, a cross-trained employee, pre-positioned equipment, or an available Assistant. Assigning a senior manager to a station at 5:00 p.m. adds one pair of hands but deprives the department of its decision capability.

Managing the crisis and closing the learning loop

When an incident crosses outlet boundaries or threatens business continuity, a cellule de crise (crisis management cell) replaces fragmented conversations. It remains small: the decision-maker, the operations lead, the functions directly concerned, and the custodian of the main courante. The group maintains a single operating picture, timestamps decisions, assigns actions, and schedules the next update. Other managers remain in their operations.

Communication follows two distinct time horizons. During execution, feedback is brief and corrective: the expected standard, the observed deviation, its consequence, and the instruction. Afterwards comes the retour d’expérience: the original plan, the actual timeline, the decisions taken, and their effects. The facilitator examines causes and interactions; disciplinary matters are handled separately so that they do not distort the operational analysis.

The RETEX follows a three-part logic: what was expected to happen, what actually happened, and what must now become standard. A retour d’expérience must produce an owner, a deadline, and a verifiable change: revised glassware PAR levels, a closure threshold, a wet-weather plan, a POS offline drill, or cross-training. Teams then receive four pieces of information: what is changing, why it is changing, when the change takes effect, and who will verify implementation. Otherwise, the debrief remains a ritual.

The owner of an Action Plan following a crisis should be thought well beforehand. Accountability, fault, and responsibility should not be part of the decision, and management should keep focusing on the Action Plan, not the owner. The question is not “who should do it” or “who should be taking care” but rather “who is the better person to achieve the stated plan”. Motivation and ownership will steer the decision, but management should be able to evaluate each criteria as a system.  

Less management, more command

A complex hotel needs stable intent, a shared operating picture, clearly identified support functions, and thresholds that allow managers to act at the appropriate level. This system recognises that resources will be imperfect, some capabilities will have to be degraded, and priorities will conflict. It makes those trade-offs explicit before they become improvisations.

Senior management must understand the operation in sufficient detail to formulate useful intent, decide what will be protected, allow managers to build the method, and intervene when agreed thresholds no longer hold. The evidence of effective command is not that the senior leader personally rescued the service; it is that the organisation reconfigured itself before needing them to do so.