A Strategy Session That Changes How Decisions Are Made

A strategy session often ends in the same way: the participants are tired, the walls are covered with sticky notes, “growth,” “customer centricity” and “digitalisation” have appeared in the presentation, and the leaders’ calendars remain unchanged.

This does not necessarily mean that the team worked badly. More likely, the event was designed for discussion or inspiration but expected to produce a strategic choice.

A session does not become strategic because of an off-site venue, the seniority of the participants or the number of days. It becomes strategic when it helps an organisation:

• make several connected choices under uncertainty;
• reject incompatible directions;
• test the logic behind those choices;
• reallocate attention, money, people and authority;
• establish a way to revisit the strategy as new evidence emerges.

The meeting itself does not create the result. The result comes from a management cycle that starts before the session and continues afterwards.

In one minute

A strong strategy session answers five questions:

1. Which strategic challenge requires a decision now?
2. Which facts, uncertainties and assumptions define the situation?
3. Where will the organisation play, and how does it expect to win?
4. What will it stop doing, and which resources will it genuinely reallocate?
5. How will the team know whether the strategic hypothesis is supported or needs revision?

For a complex cross-functional challenge, a 3+1 format is useful:

• day 1 — shared reality and diagnosis;
• day 2 — alternatives, challenge and choice;
• day 3 — capabilities, portfolio, resources and execution;
• day 4 — integration, rehearsal and launch of the management cycle, but only when it has distinct work to accomplish.

The minimum output is:

• a defined strategic challenge;
• a connected set of strategic choices;
• an explicit “will not do” list;
• a register of critical assumptions;
• a limited portfolio of initiatives;
• resource and ownership decisions;
• outcome measures, leading signals and guardrails;
• a 7/30/60/90-day calendar;
• a rule for continuing, changing or stopping a strategic bet.

What a strategy session is for

A strategy session is a temporary decision-making system. It brings together people, evidence and authority that are normally separated across functions and meetings.

It is especially useful when an organisation needs to:

• choose a growth direction or new business model;
• respond to a change in markets, technology, regulation or customer behaviour;
• align strategy after a change in ownership, leadership or operating model;
• stop competition among numerous disconnected initiatives;
• allocate scarce investment, talent or production capacity;
• define the role of a function or business unit within the wider strategy;
• test the robustness of a decision across several possible futures;
• turn an already selected strategy into an executable system.

Strategy does not remove uncertainty. It makes assumptions and choices explicit so that the organisation knows which signals to watch and when to change course.¹

When a session will not help

Do not schedule a three- or four-day strategy session when:

• the decision has already been made and the leader only wants the appearance of participation;
• the participants have no authority to recommend, decide or enable action;
• the critical facts still require research that has not been completed;
• the problem is operational and needs process analysis rather than strategic choice;
• the conflict concerns one leader’s behaviour and should not be disguised as a group discussion;
• the organisation is unwilling to stop any initiative or reallocate resources;
• no one owns the post-session cycle;
• the objective is to “inspire,” “build cohesion” or “discuss the future” but contains no decision.

In these situations, conduct interviews, research, process analysis, owner negotiations or a short working meeting around a specific decision first.

A strategy session is not a lecture or a budget defence

Four formats are easily confused.

The formats can be connected, but their boundaries should remain clear. When long-term choice is immediately converted into an annual budget defence, short-term financial issues tend to crowd out the strategic conversation.²

The objective: make named decisions, not “develop a strategy”

“Develop the company’s three-year strategy” is too broad to guide event design.

A working objective contains six elements:

By the end of the session, [who] will make or prepare [which decisions] about [the object and horizon], using [the main evidence], within [the constraints], and establish [how the choices will be tested and revisited].

Example:

By the end of day 3, the CEO and product, commercial and operations leaders will select two priority customer segments for 2027–2029, define a how-to-win mechanism for each, stop incompatible initiatives, allocate 80% of investment capacity and approve measures for testing the critical assumptions over the next 90 days.

Weak objective:

Align the vision and create a growth roadmap.

Stronger objective:

Choose among three growth directions, state the conditions required for each to succeed, define the primary bet and fallback option, and decide which projects will not receive resources in the next cycle.

The output contract

Before selecting a venue, define five parts of the contract.

If the leader will make the final decision, say so in advance. Promising consensus where one person actually decides damages trust more than an honest statement of authority.

What to expect

Realistic expectations:

• participants will see contradictions among current decisions;
• disagreement will emerge around markets, risk and organisational capabilities;
• some favoured initiatives will lose priority;
• several important questions will remain hypotheses;
• a decision may require additional validation;
• the leader will have to change their own decisions, measures or resource allocation;
• post-event energy will decline unless the operating system changes.

Do not expect:

• certainty about the future;
• unanimity on every question;
• a finished five-year plan that never needs revision;
• changed behaviour as a result of one strong speech;
• execution of dozens of initiatives without removing existing work;
• a reliable evaluation based only on participant satisfaction.

A useful result is not the absence of doubt. It is the ability to name the choice, its logic, its critical assumptions and the signal that would cause the choice to be revisited.

Readiness check

Rate every statement as yes, partly or no.

Decision

• There is one primary strategic question.
• At least two legitimate alternatives exist.
• The cost of postponing the decision is understood.
• The person with authority will attend the critical blocks.

Evidence

• A baseline view of economics, customers, competitors and internal results exists.
• Facts are separated from interpretations.
• Forecasts are labelled as assumptions.
• Missing or contradictory evidence is visible.

Execution

• At least some resources can be reallocated.
• Initiatives can be stopped or postponed.
• An owner has been appointed for the post-session cycle.
• The first outcome review is already in the calendar.

Participants

• The group includes people with decision authority.
• It represents knowledge of customers, operations, finance and implementation.
• Status differences can be facilitated safely.
• Remote participants have equal access to evidence and influence.

If no dominates the decision or execution categories, a multi-day session is premature.

Preparation: the work starts three to six weeks earlier

Methodfield’s practical recommendation is to treat preparation as the first stage of strategy, not as event logistics.

Week 1: mandate and question

The sponsor and facilitator define:

• the object and horizon of the strategy;
• one to three decisions to be made;
• decisions outside the scope;
• the decision rule;
• the core participant group;
• confidentiality;
• criteria for session success;
• the owner of follow-through.

Weeks 1–2: strategic diagnosis

Conduct 8–15 interviews with participants and key stakeholders. Ask for specific episodes and decisions rather than general opinions:

• Which decision in the past six months exposed a weakness in the current strategy?
• Where are functions optimising different outcomes?
• Which assumption about a customer or market has not been tested for a long time?
• Which initiative is no one willing to stop?
• Which decision repeatedly returns to the leader?
• What could a competitor or new entrant do differently?

Add evidence:

• revenue, margin, share, retention and cash-flow trends;
• economics by segment, channel or product;
• customer events and reasons for choice;
• competitor actions;
• regulatory and technology changes;
• production and talent constraints;
• the current initiative portfolio and actual workload;
• results from the previous strategy and inaccurate forecasts.

Weeks 2–3: evidence pack

Prepare a short pre-read structured around questions rather than departments:

1. What has changed?
2. What is happening to the current strategy?
3. Where do results differ, and why?
4. Which alternatives are already visible?
5. Which assumptions need testing?
6. Which decisions are required in the session?

A useful guide is 8–12 substantive pages plus appendices. Practitioner research on strategic planning suggests that concise material distributed in advance supports discussion better than a long presentation first encountered during the meeting.³

Label every important statement:

• fact — observable and verifiable;
• interpretation — an explanation of a fact;
• assumption — something that must prove true;
• decision — a choice that remains to be made;
• unknown — a material evidence gap.

Week 3: options and criteria

Do not ask one working group to prepare the “right strategy.” Develop two to four genuinely different options.

For each, describe:

• target customer, market or field of play;
• problem or value;
• how-to-win mechanism;
• required capabilities;
• economic logic;
• main risks;
• critical assumptions;
• early signals;
• what the organisation would have to stop doing.

Agree the criteria before evaluating the options. Otherwise, the group may change the criteria after seeing its preferred alternative.

Week 4: design and rehearsal

The facilitator:

• sequences the work from divergence to choice;
• allocates time across evidence, alternatives and decisions;
• agrees the leader’s participation points;
• prepares artefact templates;
• tests technology and accessibility;
• rehearses difficult transitions;
• identifies evidence that must not be shown at an individual level;
• appoints the owner of the decision log.

Five to seven days beforehand, participants receive the pre-read and answer in writing:

1. Which fact changes your view most?
2. Which conclusion do you disagree with?
3. Which choice matters most?
4. Which assumption is most dangerous?
5. What should the organisation stop doing?

Who should participate

Group size should be determined by the work, not by hierarchy.

Decision core: usually 6–10 people

The core includes:

• the sponsor or final decision owner;
• owners of the business outcome;
• leaders who control critical resources;
• people with essential customer, market, technology and execution knowledge;
• the strategy lead or decision secretary.

A smaller group makes genuine strategic conversation easier. A wider population can contribute through selected blocks, interviews and option testing.²

Extended group: usually 12–24 people

An extended group is justified when:

• execution requires many functions;
• material knowledge is distributed;
• implications must be tested across regions or products;
• participants will own execution streams.

Retain a smaller core with the authority to close decisions.

Roles

One person may hold more than one role, but the leader should not simultaneously be the main speaker, sole source of content, facilitator and judge of their own idea.

When the session includes an external speaker

An external expert is useful as a source of evidence, a different view of the world or a professional challenge. The expert should not replace strategic work.

Give the speaker:

• one strategic question;
• organisational context without unnecessary confidential detail;
• three to five assumptions to challenge;
• the desired form: evidence, mechanisms and contradictions rather than a motivational story;
• limits on commercial presentation.

A working block:

Do not place the speech immediately before a final vote: a vivid narrative can receive disproportionate weight. Extract the evidence first, then have participants independently record their implications.

When the speaker is the leader

The leader’s speech establishes the status and boundaries of the conversation. Its influence must therefore be designed with particular care.

A 12–15-minute opening

The leader answers six questions:

1. Why is the decision required now?
2. What has changed in the external or internal system?
3. Which facts cannot be ignored?
4. Which decisions are genuinely open?
5. What is non-negotiable, and why?
6. How has the leader personally contributed to the current situation, and what are they prepared to change?

Strong opening:

We promised acceleration, expanded the product portfolio and stopped none of the existing initiatives. I continued approving exceptions and therefore reinforced the conflict among priorities. Segment, portfolio and product-capacity decisions are open in this session. The constraint is that we will not compromise safety commitments or current contracts.

Weak opening:

I want you to think more boldly. All ideas are welcome. At the end, I will tell you which ones are realistic.

During diagnosis, the leader speaks later

Participants first:

• record conclusions individually;
• exchange evidence in small groups;
• expose differences;
• only then hear the leader’s view.

Groups tend to spend more time on information that everyone already holds and do not always surface unique evidence held by individual participants.⁴ An early view from a high-status leader can narrow the conversation further.

The leader should leave selected blocks

This is appropriate when the group needs to generate:

• candid evidence about senior-leadership behaviour;
• a map of decisions that return to the top unnecessarily;
• risks that people are reluctant to raise;
• feedback on current incentives and sanctions.

The facilitator does not report individual statements. They return themes, frequency, examples of mechanisms and questions that require decisions.

Research on leader inclusiveness links behaviours that invite contribution with psychological safety and engagement in improvement within teams marked by status differences.⁵ In practice, the leader must not merely permit disagreement but demonstrate how it can influence the decision.

At the point of choice, the role changes again

The leader:

• clarifies the criteria;
• checks that alternatives were genuinely considered;
• identifies which advice was accepted or rejected;
• makes the decision under the stated rule;
• explains the logic;
• records what evidence may change the decision later.

After the session

Within seven days, the leader:

• confirms resource decisions;
• responds to open proposals;
• communicates what was accepted, delayed or rejected;
• removes at least one named obstacle;
• does not launch new initiatives outside the agreed portfolio.

The strongest post-session signal is not a thank-you message. It is a changed decision.

Principles for building a three- or four-day programme

1. Move from questions to decisions, not slides to discussion

Every block should produce one of five outcomes:

• fact;
• conclusion;
• alternative;
• decision;
• open question with an owner and date.

2. Diverge before converging

The team needs time to:

1. form a view independently;
2. expose differences;
3. create alternatives;
4. challenge them;
5. choose;
6. translate the choice into action.

Trying to “align immediately” often conceals useful disagreement.

3. Keep facts, interpretations and bets visible at the same time

Use separate areas:

4. A decision includes rejection

If every function keeps all current projects and receives new ones, the strategy has not changed resource allocation.

5. Choose the model for the question

Do not begin with “let’s do a SWOT.” Define the decision first, then select the smallest useful set of models.

6. Create the artefact inside the block

Update a live document after every block. Do not rely on sorting sticky-note photographs later.

7. Energy does not replace cognitive work

Working intervals will usually last 60–90 minutes. A difficult choice should be followed by a break, movement or a change of mode. Evening participation should be voluntary. Do not make critical decisions late at night or after alcohol.

Recommended 3+1 architecture

The base format is three shared days. Add day 4 only when there is a demonstrated need.

When day 4 is needed

• The choice spans several business units.
• Significant resource reallocation is required.
• The portfolio contains interdependent initiatives.
• New decisions or processes need rehearsal.
• The choice must be cascaded to teams and regions.
• The cost of misinterpreting the strategy is high.

When day 4 is not needed

• It repeats presentations from the earlier days.
• It exists only because the venue has been booked.
• The participants do not own execution.
• The decisions remain open.
• There is no integration material to work on.
• Existing management meetings can handle the work.

Run it immediately or later

Choose one:

• four consecutive days — when participants have travelled, decisions can close quickly and day 4 performs real integration work;
• three days plus day 4 after 7–14 days — when teams need to validate dependencies, evidence and resource implications;
• three days without a fourth — when the decision is clear and execution can run through the existing management cycle.

Detailed agenda: day 1 — shared reality

The purpose is not to agree an attractive picture. It is to identify the challenge that requires a strategic response.

“The strategy we actually run” block

Do not start by asking about the mission. Take 10–15 recent decisions:

• where investment went;
• who was hired;
• what was launched;
• which customers received a “yes”;
• which exceptions were approved;
• what was stopped;
• which measures were rewarded.

Ask participants to infer the actual strategy from these decisions. Then compare it with the stated strategy.

Strategic-challenge formula

We seek [outcome], but [change or constraint] makes the current approach insufficient. Evidence [facts] suggests [diagnosis]. We therefore need to choose among [classes of alternatives] without violating [constraints].

Do not embed the solution in the diagnosis.

Detailed agenda: day 2 — alternatives and choice

The purpose is to create genuine alternatives, test them under uncertainty and close the critical choices.

A genuine alternative

“Grow quickly,” “grow moderately” and “grow cautiously” are not different strategies.

Alternatives should differ on at least two dimensions:

• target customer or market;
• value proposition;
• source of advantage;
• revenue model;
• channel or delivery approach;
• required capability system;
• nature of risk;
• resource allocation.

Decision rule

Choose one mode before the discussion:

• the sponsor decides after consultation;
• the group makes a recommendation and the board approves;
• the group decides by consent within the agreed criteria;
• two alternatives enter bounded experiments.

Voting reveals preference but does not demonstrate strategy quality. Do not use a total of sticky-dot votes as the sole basis for a significant decision.

Detailed agenda: day 3 — strategy execution

The purpose is to change the management system, not only its language.

A constrained portfolio

Give the group 100 units of investment capacity. Every initiative consumes:

• budget;
• leadership attention;
• scarce specialist capacity;
• operational change capacity.

Adding an initiative is possible only after another is removed or reduced. Research on strategic initiative portfolios highlights the need to manage initiatives as an interdependent system — across timing, scope, resource allocation, interfaces and feedback cycles — rather than as a collection of independent projects.⁸

From strategy to OKRs

OKRs do not replace strategic choice. They help make the expected result observable.

The sequence is:

Strategic challenge
→ connected set of choices
→ critical capabilities
→ limited initiative portfolio
→ outcomes and signals
→ OKR or another management cycle

If a team starts by writing Key Results before deciding where to play and how to win, it measures activity without a shared strategic logic.

Day 4 — integration lab

Day 4 may involve a smaller group: stream, function, data and critical-resource owners.

Teach-back

Without access to the final slide, each function explains:

• the choice that was made;
• why it was made;
• what it changes;
• what the function will stop doing;
• which assumption it will test.

Differences reveal where the strategy has not yet become a shared working model.

Decision rehearsal

Use a real situation that will occur within the next two weeks:

• a major customer requests an exception;
• a new idea needs a scarce team;
• a competitor cuts its price;
• a measure falls behind;
• a functional target conflicts with the shared outcome.

Participants make the decision using the new strategy. Observers record:

• who decided;
• which criteria were used;
• which information was missing;
• where the old logic returned;
• which rule needs to change.

Which models to use

Do not use every model. One diagnostic model, one choice model, one uncertainty model and one execution model are usually enough.

Scenarios are not forecasts. They allow teams to explore several plausible future conditions, rehearse decisions and recognise early signals.⁹

How to use SWOT

SWOT is useful as a late synthesis tool when:

• factors are supported by evidence;
• they relate to a specific strategic question;
• the team connects internal and external conditions;
• the output is alternatives and action.

Do not begin with four empty quadrants. Participants will quickly produce a broad list, but not a strategic choice.

Business simulations and exercises

An exercise is justified when it makes a work mechanism visible and ends with transfer into the real operating system.

For every exercise, define:

Work problem
→ exercise mechanism
→ observable behaviour
→ debrief
→ work artefact
→ post-session test

1. Strategy auction

Purpose: reveal real priorities and resource conflicts.

Scenario:

• participants receive 100 notional units of resource;
• options require budget, people and leadership attention;
• a new constraint appears after the first allocation;
• every additional choice requires an explicit exclusion.

Observe:

• whether functions protect their own projects;
• whether dependencies are discussed;
• whether management capacity is considered;
• which criteria are changed to protect a preferred result.

Debrief:

1. What were we protecting: the strategy or our territory?
2. Where did the constraint change the choice?
3. What are we willing to stop in the real portfolio?
4. Which resource decision must the leader make?

Artefact: draft start / continue / change / stop portfolio.

2. Wargame: competitor response

Purpose: test whether the strategy depends on passive competitors.

Scenario:

• team A defends the selected strategy;
• team B plays a strong incumbent competitor;
• team C plays a new entrant or substitute;
• team D plays a critical customer, partner or regulator.

Each team receives 45 minutes and a resource constraint.

Debrief:

1. Which response were we unprepared for?
2. Which advantage is easy to copy?
3. Which capability must exist before launch?
4. Which signal will show that a competitor has started responding?

Artefact: response and early-signal map.

3. Premortem

Purpose: identify failure causes before participants become publicly attached to the plan.

Scenario:

Eighteen months have passed. The strategy has been declared a failure. Which specific events and decisions caused it?

Participants write independently, then cluster the causes and select the most dangerous. Gary Klein proposed the premortem as a way to use imagined failure to expose threats to a plan.¹⁰

Artefact: risks, warning signals, owners and actions.

4. Scenario wind tunnel

Purpose: test the strategy under different external conditions.

Scenario:

• select three to four meaningfully different scenarios;
• evaluate the strategic option within each;
• mark it as works, needs adaptation or becomes unacceptable;
• identify no-regret moves and conditional decisions.

Debrief:

1. What remains sensible across every scenario?
2. Which decision should wait for a signal?
3. Which small move preserves the right to play?
4. Which trigger switches the organisation to fallback logic?

Artefact: robustness and trigger table.

5. Stop-doing market

Purpose: turn priority into the real cessation of work.

Every leader brings:

• one initiative they are prepared to stop;
• one recurring activity that can be simplified;
• one measure that reinforces the old behaviour;
• one request to another leader.

The group accepts an exclusion only when it has an owner and stop date.

Artefact: stop-doing list and released capacity.

6. Hidden information

Purpose: demonstrate how functions lack a complete picture for a strategic decision.

Different groups receive different parts of the market, customer, operational and risk evidence. A strong decision is possible only when unique information is combined.

Important:

• use a fictional, neutral case;
• do not evaluate participants’ intelligence;
• do not turn the debrief into blame of a function;
• transfer the learning into the real evidence-sharing process.

Artefact: requirements for the evidence pack and discussion rules.

7. Press conference from the future

Purpose: test the clarity of the strategic logic, not create a slogan.

One group presents the organisation three years from now. The others play:

• a customer;
• an investor;
• an employee;
• a partner;
• a critical journalist.

Questions should reveal:

• who received value;
• why they chose the organisation;
• what the company did differently;
• which price it paid for focus;
• which capability enabled the result;
• which evidence supports the story.

Artefact: testable strategic narrative.

8. Decision-rights simulation

Purpose: test the new management system.

Give the team three to five real decisions with incomplete evidence and limited time. Before starting, participants must name:

• decision owner;
• who provides mandatory input;
• who is consulted;
• who executes;
• deadline;
• escalation rule.

Artefact: map of recurring strategic decisions.

Exercises to avoid

Do not use an exercise simply because it is popular.

Avoid:

• trust falls and physically risky formats;
• public ranking of personality or “courage”;
• games based on humiliation, secret exclusion or forced disclosure;
• competitions with no connection to the work system;
• childish metaphors for an audience that has not been given a rationale;
• exercises in which the leader already knows the “correct answer”;
• voting presented as proof of strategy;
• mandatory evening activities;
• simulations that produce no work artefact.

Even a strong exercise is useless without a debrief.

How to run the debrief

Use five levels:

1. What happened? Observable actions only.
2. What were we thinking at the time? Participant interpretations.
3. Which mechanism appeared? Status, incentive, information, resource or decision rule.
4. Where does this happen in our work? A specific episode.
5. What will we change? Rule, artefact, role, measure or forum.

Example:

In the simulation, the commercial group did not disclose a risk until the final round. This does not prove “low trust.” Possible mechanisms include an incentive to defend volume, the absence of a mandatory risk field and late operations involvement. In real work, we will review the last three major proposals and change the readiness criteria.

What reactions to expect from participants

Reactions change as the session moves from analysis to exclusion.

Resistance does not always mean disengagement. It may indicate that the session has reached a genuine redistribution of status, budget or freedom to act.

Hybrid and distributed format

When some participants join remotely:

• everyone must have the same digital access to materials;
• remote participants work in the same small-group process rather than watching a broadcast;
• a separate facilitator monitors their queue and questions;
• decisions and document changes are visible in real time;
• do not rely on physical walls that remote participants cannot see;
• collect positions simultaneously and in writing;
• schedule breaks for time zones;
• prioritise sound quality over decorative video.

If a critical decision owner or holder of unique evidence cannot participate fully, reschedule the critical block.

Artefacts that should remain

Strategy on one page

• strategic challenge;
• desired outcome and horizon;
• where to play;
• how to win;
• critical capabilities;
• management system;
• main assumptions;
• measures and triggers;
• what the organisation will not do.

Decision log

Strategic-bet register

Portfolio

For every initiative:

• link to a strategic choice;
• expected outcome;
• critical assumption;
• owner;
• allocated capacity;
• dependency;
• next evidence;
• continue, change or stop criterion.

How to measure the result

Asking only “How much did you like the session?” measures reaction, not strategy quality.

Use five levels.

Three types of measure

For each bet, define:

• outcome — what must change for the customer or business;
• leading signal — an early observation linked to the logic;
• guardrail — what must not deteriorate while pursuing the outcome.

Example:

Bet: mid-sized logistics companies will choose a standardised service over a custom implementation.

• outcome: share of the target segment with positive contribution margin after 12 months;
• leading signal: share of qualified customers accepting the standard process without customisation;
• guardrail: retention and number of critical exceptions;
• trigger: if fewer than 25% of 20 qualified proposals accept the standard process, revisit the value proposition or segment.

Do not promise a universal percentage improvement from the session. Define the baseline, direction, period, data source and decision rule.

What to do during the first 90 days

The continuation strategy is part of event design.

Within 48 hours

Send a decision package, not photographs of flip charts:

• strategy on one page;
• decision log;
• open-question list;
• bet register;
• start / continue / change / stop portfolio;
• resource decisions;
• owners and dates;
• review calendar.

Do not rewrite uncomfortable language in a way that removes exclusions and objections.

By day 7

The sponsor:

• confirms resource decisions;
• answers deferred questions;
• stops the agreed initiatives;
• communicates the strategy to the next levels;
• makes the first real decision using the new logic.

Day 14

Run a 45–60-minute review:

1. Where has the strategy already been used?
2. Which decision was made differently?
3. Which old rule returned?
4. Which resource was not provided?
5. Which question requires a sponsor decision?

Day 30

Run the first full strategy review:

• update the facts;
• test critical assumptions;
• assess early signals;
• revisit the portfolio;
• remove constraints;
• record changes to the logic.

Day 60

Review:

• initiative interactions;
• actual use of scarce resources;
• customer and operational consequences;
• decision quality below the top team;
• new external signals;
• the need to stop or strengthen bets.

Day 90

Answer:

1. Which assumptions received support?
2. Which were contradicted?
3. Which outcome started moving?
4. What deteriorated on the guardrails?
5. Which initiatives should continue, change or stop?
6. Does the choice need revision, or only the execution approach?

The output is a decision, not a report.

How to sustain the result

Weekly: execution

Use an existing management forum:

• one outcome;
• one risk;
• one dependency;
• one decision or request.

Do not create another meeting if this can be embedded in the current cadence.

Monthly: strategic bets

For each bet:

• what was assumed;
• what is being observed;
• what it means;
• which decision follows;
• when the next signal will arrive.

Quarterly: portfolio and resources

Review:

• links between initiatives and strategy;
• total workload;
• conflicts and dependencies;
• released and newly required resources;
• initiatives with no new evidence;
• the need to stop work.

Research on strategy implementation describes it not as the transmission of a finished plan but as work across structure and processes, resources, monitoring, framing and negotiation.¹¹

Every six months: strategic logic

Revisit:

• market and system changes;
• where-to-play and how-to-win choices;
• capabilities and management systems;
• assumptions and scenario triggers;
• the quality of previous forecasts.

At a trigger: out-of-cycle review

Do not wait for the calendar when:

• customer behaviour changes sharply;
• a competitor changes the rules of the game;
• material regulation appears;
• a critical capability becomes unavailable;
• bet economics move outside the defined range;
• a critical technology is acquired or lost;
• ownership or the management model changes.

A one-year development strategy

Do not schedule the next multi-day session automatically one year later. It is needed when a new strategic question emerges. A quarterly review may be sufficient; a major environmental change may require a new session after three months.

After Action Review as a learning mechanism

At the end of each day and after the first significant decisions, use an AAR:

1. What did we expect?
2. What happened?
3. Why was there a difference?
4. What will we retain?
5. What will we change in the next action?

Formal AAR guidance stresses not only discussion but follow-up changes to procedures, retraining and integration of lessons.¹²

For strategy, this means that a strong review changes a rule, resource, measure, decision or hypothesis. It does not end with “we need to communicate better.”

Common failures

An agenda that is too broad

Three days cannot equally deeply refresh the mission, select markets, redesign the organisation, create a culture, write OKRs and resolve every operational conflict.

Presentations instead of work

When day 1 consists of functional presentations, participants defend the past and become tired before reaching the strategic question.

A single option

Discussing one plan may improve the presentation, but it does not test the choice.

False consensus

The group votes, but the leader still decides using unstated criteria.

SWOT without evidence

Participants produce dozens of general factors without connecting them to a mechanism or decision.

Initiatives without capacity

The strategy adds work but does not stop existing work.

The leader gives the answer too early

Other participants start improving the leader’s idea rather than exposing alternatives.

Mixing strategy and budget

Short-term target defence displaces the choice about the future.

Lagging indicators only

The team discovers failure too late because it did not define early signals for its assumptions.

Communication without management change

The strategy is published, but investment, incentives, criteria and decision rights remain the same.

Repeating the session instead of building a system

Six months later, the organisation meets again “to align” without reviewing the previous cycle’s decisions.

Quality checklist

Before the session

• A specific strategic question exists.
• Open and closed decisions are named.
• The decision rule is agreed.
• Distinct alternatives have been prepared.
• The pre-read separates facts from assumptions.
• The group has the necessary authority and knowledge.
• The leader’s role is agreed.
• The day-30 review is in the calendar.

During the session

• Positions are formed before the leader gives their view.
• Arguments connect to evidence.
• Customer and competitor responses are considered.
• The choice includes an exclusion.
• The resource constraint is real.
• Decisions are recorded immediately.
• Open questions have owners and dates.
• Every day ends with an AAR.

After the session

• The decision package is sent within 48 hours.
• The sponsor confirms resources.
• The stop-list begins to take effect.
• The first real decision uses the new logic.
• Bets have signals and triggers.
• Days 30/60/90 are designed for decisions, not status updates.

Practical next step

Before discussing the venue, complete seven lines:

1. Strategic challenge: which change requires a choice?
2. Decision: what must be selected by the end?
3. Alternatives: which two genuinely different paths are legitimate?
4. Exclusion: what might the organisation stop doing?
5. Authority: who closes the decision?
6. Evidence: which fact or experiment would change the choice?
7. Continuation: which meeting will make the next decision after 30 days?

If lines 2, 4, 5 and 7 cannot be completed, it is too early to design the exercises.

Sources / Источники