30 AI Prompts for Executives & Managers
Copy-ready prompts that turn the AI into a rigorous thinking partner — assume you will attach the report, the data or the plan, and apply real decision frameworks so you think sharper and communicate with impact.
Each prompt sets the AI a senior-adviser persona, references your attached material, and applies a decision or communication framework with a defined output. These prompts are tool-agnostic and optimised to work across all major AI assistants — Claude, Microsoft Copilot and ChatGPT (and Google Gemini). Replace the [bracketed placeholders] with your own details, and always review AI output before you act on it.
Strategy and planning
Three Strategic Options with Recommendation
strategy adviser to the CEO
a brief describing the strategic situation or challenge (or a market/competitor summary)
- Read the situation and identify three genuinely distinct strategic options
- For each option, define the core thesis, the key bet, resources required, and main risk
- Name the leading indicator that would show each option is working
- Recommend one option with clear reasoning
- Options must be meaningfully different, not variations of the same idea
- No false safe choice presented as neutral
- Ground every claim in the attached material
An options comparison table (thesis, bet, resources, risk, leading indicator) followed by a one-paragraph recommendation.
See example AI output
Three options for Meridian Distribution's declining wholesale margins: (1) Vertical integration - acquire two regional distributors, RM4.2m capex, key bet is scale beats Shopee sellers on price; risk is integration drag; indicator is gross margin recovery to 18% by Q2. (2) Digital-first pivot - build a B2B ordering app, RM800k build cost, bet is convenience retains loyal retailers; risk is slow adoption; indicator is 30% of orders online by month four. (3) Niche specialisation - exit low-margin SKUs, focus on halal-certified FMCG; risk is short-term revenue drop. Recommendation: pursue option 2, lowest capital risk with fastest signal.
SWOT into TOWS Action Matrix
strategy consultant
company or initiative background, recent performance data, and competitor notes
- Run a rigorous SWOT for the company or initiative using the attached context
- Convert the SWOT into a TOWS matrix: strengths-to-opportunities and weaknesses-to-threats moves
- Identify the top three actionable moves
- Avoid generic textbook entries
- Every entry must be specific to the attached context
- No generic SWOT filler such as 'strong brand'
- Moves must be actionable, not descriptive
A SWOT grid followed by a TOWS action matrix and a ranked list of the top three moves.
See example AI output
SWOT for Kenanga Retail's mall expansion: Strength - 12-store footprint with strong Klang Valley footfall data. Weakness - POS systems not integrated across stores. Opportunity - Mitraland's new Cyberjaya mall seeking anchor tenants. Threat - Aeon undercutting on private-label pricing. TOWS: use footfall data to negotiate favourable Cyberjaya lease terms (S-O); integrate POS before signing new leases to avoid inventory blind spots (W-T). Top 3 moves: (1) sign Cyberjaya letter of intent by August, (2) complete POS integration by Q3, (3) launch private-label range to counter Aeon within 90 days.
Red-Team Your Strategic Plan
red-team adviser
the plan document you want stress-tested
- Read the plan and argue the strongest possible case against it
- Identify the assumptions most likely to be wrong
- Trace the second-order consequences if the plan succeeds or fails
- Describe how a smart competitor would respond
- Recommend specific fixes to strengthen the plan
- Be intellectually honest, not a strawman critique
- Focus on the plan's real weak points, not hypothetical ones
- End with constructive fixes, not just criticism
A critique organised by weakest assumption, second-order risk, and competitor response, followed by a short list of fixes.
See example AI output
Your plan to launch same-day delivery in Penang assumes courier costs stay flat - unlikely given fuel subsidy rationalisation expected September. Second-order risk: if margins compress, you may quietly reduce delivery radius, eroding the promise that won customers. A sharp competitor like Lalamove could respond by bundling same-day into existing routes at lower marginal cost, undercutting you within weeks. Fix 1: lock courier rates via a 6-month contract before launch. Fix 2: pilot in one township first to validate true cost-per-delivery. Fix 3: build a radius-reduction contingency into the customer promise now, not after complaints start.
OKR Set for a Team Goal
OKR coach
the team's current goals, past quarter's results, or strategy brief
- Propose one ambitious, outcome-focused objective for the team or goal
- Draft three measurable key results that would prove the objective is achieved
- Flag the common traps that make OKRs fail for this specific team
- Recommend a check-in cadence
- Key results must be outcomes, not task lists
- Keep the set small - one objective, three key results maximum
- Each key result must be independently measurable
The objective, three key results with target numbers, a list of traps to avoid, and a suggested check-in rhythm.
See example AI output
Objective for Nexora Technologies' customer success team: Turn onboarding into our strongest retention lever. KR1: reduce time-to-first-value from 21 to 10 days. KR2: raise 90-day activation rate from 61% to 80%. KR3: cut onboarding-related churn tickets by 40%. Trap to avoid: the team defaulting to activity metrics like 'calls made' instead of outcomes - resist adding a fourth KR for busyness. Another trap: sandbagging targets to guarantee a green scorecard. Cadence: biweekly 20-minute check-ins reviewing KR movement only, with a full reset conversation at the 6-week mark if KR2 is off track.
Three Strategic Implications from Market Data
market analyst
competitor announcements, market reports, or industry data
- Read the attached competitor and market information
- Distil it into the three strategic implications that matter most for us
- For each implication, state the specific move it suggests
- Prioritise so-what over description
- Every implication must be evidence-based, cited from the attachment
- No restating the data without interpretation
- Each implication must map to one concrete recommended move
Three numbered implications, each with the supporting evidence and the recommended move.
See example AI output
From the attached Q2 F&B market report: (1) Implication - cloud kitchens are eroding dine-in traffic in Klang Valley (footfall down 14% per the report). Move: pilot a delivery-only brand under Bumi Hijau's existing kitchen capacity within 60 days. (2) Implication - Grab's new subscription model locks in repeat orders. Move: negotiate preferred-partner placement rather than competing on ad spend alone. (3) Implication - younger consumers favour plant-based options (22% growth cited). Move: trial two plant-based SKUs next quarter before committing to a full menu redesign.
Board and reporting
Board-Ready Paper from Raw Results
board-paper writer
the underlying results, updates, or data you need turned into a board paper
- Read the attached results and updates
- Lead the paper with the recommendation, not the background
- Structure the context, analysis, options considered, and risks after the ask
- State exactly what decision is being requested
- One page maximum
- Lead with the ask - never bury the point in context first
- Every option considered must be named, not just the chosen one
A one-page board paper with sections: Recommendation, Context, Analysis, Options Considered, Risks, Decision Requested.
See example AI output
Recommendation: approve RM2.1m capex to open a second distribution hub in Johor Bahru. Context: JB order volume grew 34% YoY, current hub at 92% capacity. Analysis: current hub will breach capacity by November, causing delivery delays. Options considered: (1) expand current hub - cheaper but limited runway, (2) new JB hub - recommended, (3) outsource to 3PL - faster but margin-dilutive. Risks: 4-month build timeline, hiring 18 warehouse staff in a tight labour market. Decision requested: board approval of RM2.1m capex and authority to sign the JB lease by end of August.
200-Word Executive Summary
executive editor
the full long-form report you need summarised
- Read the full report
- Identify the decision required from the reader
- Extract the three things the reader must know
- Draft a summary that leads with the decision, not the background
- Under 200 words
- No throat-clearing or preamble
- Must be readable in under 90 seconds
A single executive summary block under 200 words: decision required, three key facts, recommendation.
See example AI output
Decision required: approve the proposed 8% price increase across Pantai Health Group's diagnostic packages, effective October. Three things to know: (1) lab reagent costs have risen 19% since January and margins have compressed to 11%; (2) competitor Prognosis Labs raised prices in May with no reported customer loss; (3) a phased 4%-then-4% increase over two quarters would soften the customer-facing impact while protecting margin recovery. Recommendation: approve the phased approach, communicated to corporate clients with 30 days' notice, and review renewal churn at the 90-day mark before considering a further adjustment.
Board Q&A Prep Sheet
board-prep coach
the topic or deck you are presenting to the board
- Anticipate the ten toughest questions the board is likely to ask about the topic
- Draft a crisp, direct answer for each question
- Flag the one question you are least prepared to answer
- Suggest how to shore up that weak spot before the meeting
- Questions must be adversarial and realistic, not softballs
- Answers must be concise - no more than three sentences each
- Be honest about which answer is weakest
A ten-question Q&A prep sheet plus a flagged weak spot with a suggested fix.
See example AI output
Presenting the Cyberjaya expansion to Kenanga Retail's board. Toughest question 6: 'What happens if Mitraland delays handover past Q1?' Answer: we've negotiated a penalty clause and hold a fallback site in Bandar Utama on a 60-day option. Question 9: 'Why not wait for interest rates to ease?' Answer: the anchor-tenant slot closes in October regardless of rates; delaying forfeits the position to a competitor. Weakest prep area: question 4 on projected footfall cannibalisation from our existing Puchong store - recommend commissioning a quick catchment-overlap analysis before Thursday's meeting.
One-Page Quarterly Stakeholder Update
CEO-update writer
this quarter's performance data against plan
- Read the quarter's data against plan
- Draft a headline that captures the quarter in one line
- State performance versus plan, two wins, and one challenge with the response
- Close with the outlook for next quarter
- One page only
- Tone must be honest and confident, not defensive
- Every claim must be backed by a number from the data
A one-page update: headline, performance vs plan, two wins, one challenge and response, outlook.
See example AI output
Headline: Nexora closed Q2 at 94% of revenue target, driven by strong enterprise renewals. Performance vs plan: RM6.8m actual vs RM7.2m target. Win 1: enterprise renewal rate hit 91%, up from 84%. Win 2: new SME segment contributed RM540k, ahead of forecast. Challenge: onboarding delays pushed RM380k of new-logo revenue into Q3; we've added two implementation specialists and cut average onboarding time from 21 to 14 days. Outlook: Q3 pipeline is 22% larger than this time last quarter; we expect to close the shortfall and finish H2 on plan.
Candid Shareholder Update on a Difficult Situation
shareholder-comms specialist
no file - describe the situation you need to communicate about
- Explain what happened in plain, factual terms
- State what it means for the business and shareholders
- Describe what the company is doing in response
- Set clear expectations for what happens next
- Balance candour and confidence - no spin, no panic
- Avoid jargon and legal hedging that obscures the facts
- Keep it readable by a non-executive shareholder
A structured update: what happened, what it means, what we are doing, what to expect next.
See example AI output
What happened: a supplier quality issue affected 3% of Bumi Hijau Logistics' cold-chain shipments in June, triggering client compensation claims. What it means: an estimated RM420k one-off cost this quarter; no impact on long-term contracts. What we are doing: we've replaced the supplier, added a third-party quality audit step, and are compensating affected clients directly. What to expect: Q2 margin will show a temporary 1.2-point dip; we expect full recovery by Q3 with no change to full-year guidance. We will provide a follow-up update at the September board meeting.
Meetings and decisions
Focused Meeting Agenda Builder
meeting designer
the topic, duration, and attendee list for the meeting
- Define the single objective the meeting must achieve
- List the pre-read attendees need beforehand
- Build agenda items, each tagged decide, discuss, or inform, with a time-box
- State the decision the meeting must leave with
- Outcome-driven - every item must serve the single objective
- Protect time - total time-boxes must not exceed the stated duration
- No item without a clear decide, discuss, or inform tag
A one-page agenda: objective, pre-read, timed agenda items with tags, and the decision to leave with.
See example AI output
45-minute meeting on Kenanga Retail's Q3 pricing review. Objective: agree the phased price increase and communication timeline. Pre-read: margin analysis deck, competitor pricing snapshot. Agenda: (1) Margin context - inform, 5 min. (2) Review three pricing scenarios - discuss, 15 min. (3) Decide phased vs single-step increase - decide, 15 min. (4) Assign comms owner and dates - decide, 10 min. Leave with: a signed-off pricing schedule and a named owner for the client communication by Friday.
Decision One-Pager
decision architect
no file - describe the decision you need to make
- Frame the decision to be made clearly
- List the real options available
- Define the criteria that actually matter for choosing
- Trace the second-order consequences of each option
- Give a recommendation
- Expose the trade-offs honestly, don't hide uncertainty
- Criteria must reflect what actually matters, not what's easy to measure
- Recommendation must state the reasoning, not just the choice
A one-page decision brief: the decision, options, criteria, consequences, and a recommendation.
See example AI output
Decision: whether to bring warehouse staffing in-house or continue with a labour contractor for Meridian Distribution's new JB hub. Options: (A) in-house hiring, (B) continue contractor, (C) hybrid - core team in-house, peak-season contractor. Criteria: cost per unit, flexibility during Ramadan peak, control over service quality. Consequences: in-house lowers unit cost by 12% but reduces flexibility; contractor keeps flexibility but service complaints have risen 18% this year. Recommendation: hybrid model - hire 8 core staff now, retain contractor for the November-December peak, reassess after one full cycle.
Meeting Notes into Decisions, Actions, Open Items
minutes analyst
raw meeting notes or a transcript
- Read the raw meeting notes
- Separate out decisions that were made
- List actions with a named owner and a due date for each
- Capture open questions that remain unresolved
- Ownership must be unambiguous - no action without a named owner
- Nothing discussed should be lost or summarised away
- Keep the three sections clearly separated
Three labelled sections: Decisions Made, Actions (owner, due date), Open Questions.
See example AI output
Decisions Made: approved the RM150k budget for the Nexora partner conference; confirmed the November 14 date. Actions: Farah to finalise venue contract by July 30; Wei Ling to draft the sponsor prospectus by August 5; Raj to confirm speaker list by August 12. Open Questions: is the RM150k budget inclusive of AV costs, or does that need a separate approval; has legal reviewed the sponsor agreement template yet - no owner currently assigned, needs to be picked up before the next check-in.
Options Comparison for a Business Problem
options analyst
no file - describe the problem needing resolution
- Generate three resolution options for the problem
- Score each on cost, risk, speed, and reversibility
- Recommend one option with clear reasoning
- Note which option is the safest reversible choice
- Trade-offs must be stated honestly, not glossed over
- Always identify the reversible option explicitly
- Recommendation must reference the scoring, not just intuition
A four-criteria comparison table across three options, followed by a recommendation.
See example AI output
Problem: Pantai Health Group's radiology department is over capacity. Option A - hire two more radiologists: high cost, low risk, slow (4-month hiring), reversible. Option B - outsource overflow scans to a partner lab: medium cost, medium risk, fast, highly reversible. Option C - invest in a second MRI machine: very high cost, low risk once utilised, slow (6 months), not reversible. Recommendation: start with Option B immediately to relieve pressure, while evaluating Option A in parallel - it's the reversible, fast-acting choice while we validate whether volume growth justifies the capex in Option C.
Difficult Conversation Plan
conversation coach
no file - describe the person, the issue, and the context
- Plan how to open the conversation
- Identify the facts to anchor the discussion on
- Clarify the outcome you want from the conversation
- Anticipate likely reactions and how to respond to each
- Keep the plan constructive throughout
- Firm and respectful, not confrontational
- Avoid scripts that sound robotic or rehearsed
- Must include branches for at least two different reactions
A conversation plan: opening line, anchoring facts, desired outcome, and branches for likely reactions.
See example AI output
Conversation with Amirul about missed delivery targets three months running. Opening: 'I want to talk through the delivery numbers and understand what's getting in the way.' Anchor on facts: on-time delivery dropped from 96% to 81% since April, two client complaints logged. Desired outcome: a concrete recovery plan he owns, not a lecture. If defensive: acknowledge the pressure he's under, then redirect to specifics - 'what would need to change for October to look different?' If he raises a resourcing issue: agree to review headcount together, but hold him to a written plan by Friday regardless.
Team leadership
Honest All-Hands Change Message
leadership-comms writer
no file - describe the change or result you need to announce
- Name the reality of the situation plainly
- Explain the why behind the change or result
- Acknowledge the impact on people directly
- Give people a clear role in what happens next
- No spin - state the facts as they are
- Human tone, not corporate boilerplate
- Must include the two questions people will ask and how to answer them
The all-hands message plus a short list of the two expected questions with answers.
See example AI output
Message: 'Team, as many of you have heard, we're consolidating the Shah Alam and Klang warehouses into one site by September. This is because duplicate overheads are costing us RM1.8m a year we need for growth investment. For the 14 colleagues affected, we're offering relocation support or a transfer to the new site with no pay change. Over the next two weeks, your team leads will meet you individually to talk through options.' Expected question 1: 'Will there be layoffs?' Answer: no compulsory layoffs, transfer or relocation only. Question 2: 'When exactly does this happen?' Answer: transition completes by September 30.
SBI Feedback Plan for an Underperforming Manager
manager coach
no file - describe the manager, their team's performance, and specific examples
- Structure the feedback using situation-behaviour-impact
- Focus the conversation on the pattern, not the person
- Set a clear expectation going forward
- Offer specific support to help the manager improve
- Kind and direct at the same time, not softened into vagueness
- Every point must be evidence-based, tied to a specific instance
- Include an opening line that isn't accusatory
A feedback plan with SBI points, the expectation, the support offered, and suggested opening lines.
See example AI output
Feedback for a team lead whose team's attrition hit 40% this year. Situation: three exit interviews in five months cited the same theme. Behaviour: 1:1s cancelled or rescheduled more than 60% of the time per calendar records. Impact: team members report feeling unsupported and are job-hunting. Expectation: 1:1s protected and held weekly for the next quarter. Support: pairing with a senior manager as a mentor, plus a coaching budget. Opening line: 'I want to talk about a pattern I'm seeing in the exit interviews - not to assign blame, but because I think it's fixable together.'
Ten Coaching Questions for a Direct Report
coaching partner
no file - describe the challenge your direct report is facing
- Generate ten coaching questions for the specific challenge
- Sequence them from exploring the situation to committing to action
- Ensure each question helps the person find their own answer
- Questions must be open, not leading toward your preferred answer
- No advice-giving disguised as a question
- Sequence must build logically toward commitment
A sequenced list of ten open coaching questions, exploring through to committing.
See example AI output
For a direct report torn between two internal roles at Nexora Technologies: 1. What draws you to each option? 2. What would you be walking away from in each case? 3. Whose input have you sought, and whose haven't you? 4. What does 'success' look like in two years under each path? 5. What's the fear underneath the hesitation? ... 8. What's one small step you could take this week to test an assumption? 9. What would you tell a friend in this position? 10. What decision would you regret not making?
Clean Delegation Brief
delegation coach
no file - describe the task and the person you're delegating to
- State the context behind the task
- Define the outcome wanted, not just the activity
- Set the boundaries and decision rights clearly
- Establish check-in points
- Explain how to hand over without abandoning or micromanaging
- Authority must be explicit - state what the person can decide alone
- Define exactly one success measure
- Check-ins must be scheduled, not a vague 'let me know how it goes'
A delegation brief: context, outcome, boundaries/decision rights, check-in points, success measure.
See example AI output
Delegating the vendor renegotiation for office supplies to Siti. Context: current contract expires in 45 days, costs have crept 15% above market. Outcome: a renewed or replaced contract saving at least 10% with no service disruption. Decision rights: Siti can select and sign with any vendor scoring above 8/10 on our criteria without further approval; anything below requires my sign-off. Check-ins: brief update after initial vendor calls (week 1), decision review (week 3). Success measure: signed contract with 10%+ saving by day 40. Hand-over note: I won't be cc'd on vendor emails - full ownership from here.
Specific Recognition Message
recognition writer
no file - describe the person and what they achieved
- Name the exact behaviour the person demonstrated
- Describe the concrete impact it had
- Explain why it matters to the team, not just the individual
- Specific, not generic praise like 'great job'
- No excessive flattery
- Keep it genuine and short enough to read in one breath
A short recognition message naming the behaviour, the impact, and why it matters.
See example AI output
Wei Ling - when the Cyberjaya client escalation landed on a Friday evening, you didn't just fix the immediate issue, you traced it back to a data-mapping error affecting three other accounts and flagged it before they noticed. That's the difference between firefighting and actually protecting the client relationship. Because of that catch, we kept a RM380k renewal on track and the team didn't spend Monday cleaning up a bigger mess. This is exactly the kind of ownership that makes people trust handing you the hard problems.
Communication and influence
Concise Executive Rewrite
executive editor
the message or email you want tightened
- Read the original message
- Lead the rewrite with the point, not the context
- Cut hedging language and filler
- Keep exactly one clear ask
- Preserve the original meaning exactly
- Halve the length where possible
- Note what was cut and why
The rewritten message plus a short note on what was cut.
See example AI output
Original (142 words) rewritten to: 'Approve the RM95k Q3 marketing spend by Friday so we can lock in the campaign booking window. Breakdown attached. Happy to walk through it if useful.' (28 words). What was cut: three sentences of background on why the campaign matters (already known to the reader), a hedge ('I think maybe we should consider'), and a closing apology for the short notice - replaced with a direct deadline instead, since apologising doesn't change the timeline.
Three-Sentence Stakeholder Summary
clarity coach
the complex update you need condensed
- Read the complex update
- Write sentence one stating the situation
- Write sentence two stating the implication
- Write sentence three stating the ask
- Exactly three sentences, no more
- Omit any detail that doesn't change a decision
- Must be understandable without the original document
Exactly three sentences: situation, implication, ask.
See example AI output
Situation: Bumi Hijau Logistics' fuel costs have risen 22% since January while client contracts remain fixed until renewal. Implication: at current volumes, the cold-chain division will miss its margin target by roughly RM310k this half unless action is taken before September renewals. Ask: approve a 5% fuel surcharge clause for all contracts renewing after August 1.
Persuasive One-Pager for Buy-In
influence strategist
background on the initiative and who you're trying to persuade
- Open with the audience's priority, not your own
- Present the case for the initiative with evidence
- Pre-empt the strongest objection you expect
- Make the ask specific and easy to act on
- Audience-first framing throughout
- One page maximum
- Ask must be a specific action, not a vague 'consider this'
A one-page persuasive brief: audience priority, evidence-based case, objection pre-empt, specific ask.
See example AI output
To finance, on adopting AI-assisted reconciliation at Meridian Distribution: You've flagged month-end close taking nine days as a top pain point. Evidence: three comparable SMEs using AI reconciliation tools cut close time to four days, per our benchmarking. Likely objection: 'we don't trust AI with financial data' - addressed by using an enterprise tool with audit logging and human sign-off retained at every step, nothing is auto-posted. Ask: approve a 60-day pilot on one entity's books, with finance controlling the go/no-go decision at day 30.
Stakeholder Objection-Response Table
stakeholder strategist
the proposal and who you expect pushback from
- Map the stakeholder's likely objections to the proposal
- Identify the underlying interest behind each objection
- Draft a response that addresses the interest, not just the surface objection
- Empathetic tone, not dismissive
- Every response must be evidence-based
- Address the interest behind the objection, not just repeat the proposal
A three-column table: objection, underlying interest, response.
See example AI output
Proposal: shift Kenanga Retail's store hours to open later, close later. Objection 1: 'staff will hate the change' - Interest: protecting team morale and retention. Response: pair the shift with a rotating roster so no one works late shifts more than twice a week, and pilot with volunteers first. Objection 2: 'we'll lose the morning regulars' - Interest: protecting existing revenue. Response: footfall data shows morning traffic is only 12% of daily sales; the later slot captures the after-work segment we're currently missing entirely.
Core Change Narrative in Five Beats
narrative architect
no file - describe the change and why it's happening
- Identify the tension that makes this change necessary
- State the stakes if nothing changes
- Describe the path forward
- Define the role each person plays in it
- Compress the whole story into five memorable beats
- Must be true - no manufactured urgency
- Memorable and repeatable, not a paragraph of prose
- Every person should be able to see their role in it
The narrative in exactly five beats, each one to two sentences.
See example AI output
Nexora Technologies' shift from project-based to subscription revenue: Beat 1 (Tension) - our biggest clients now expect ongoing value, not one-off delivery. Beat 2 (Stakes) - competitors already offering subscriptions are winning our renewal conversations. Beat 3 (Path) - we rebuild our top three products as always-on services over the next two quarters. Beat 4 (Role) - engineering ships smaller, more frequent releases; sales sells outcomes, not projects. Beat 5 (Payoff) - predictable revenue means we can finally invest ahead of demand instead of chasing it.
Thinking and productivity
Socratic Thinking Partner
thinking partner
no file - describe the decision you're wrestling with
- Ask the five sharpest questions that would most change the answer, one at a time
- Wait for the user's answers before offering any opinion
- After receiving answers, synthesise a clear recommendation
- Resist giving advice until all five answers are given
- Questions must be genuinely decision-changing, not surface-level
- Synthesis must directly reference the answers given
Five sequential questions, followed by a synthesised recommendation after answers are provided.
See example AI output
On whether to open a third clinic branch for Pantai Health Group: Q1. What specifically would make you regret not opening it in twelve months? Q2. What would make you regret opening it? Q3. Is the constraint capital, staffing, or patient demand? Q4. What did you learn from the second branch's first six months that you haven't applied yet? Q5. If you could only get one signal before deciding, what would it be? [After answers] Synthesis: the real constraint is staffing, not demand - so the recommendation is to secure two specialist hires before committing to a site.
Synthesis Across Three Documents
synthesiser
three documents you want cross-referenced
- Extract the key points from each of the three attached documents
- Identify where the documents agree
- Identify where they contradict each other
- Produce one integrated view plus the open question that remains
- Cite which document each point comes from
- Do not paper over genuine contradictions
- The open question must be one that changes the recommendation
Key points per document, agreements, contradictions, an integrated view, and one open question.
See example AI output
Document A (market report) and Document B (internal sales data) agree that demand for cold-chain logistics in East Malaysia is growing. Document C (ops audit) contradicts Document A's assumption that current fleet capacity can absorb it - the audit shows utilisation is already at 88%. Integrated view: demand is real, but the constraint is fleet capacity, not sales effort, so any expansion plan must lead with capex, not marketing spend. Open question: does the RM3m fleet expansion pencil out at current fuel prices, or does it need the surcharge clause from the pricing review first?
Impact/Effort Task Prioritisation
prioritisation coach
your current task list and stated goals
- Sort the tasks against the stated goals by impact and effort
- Identify the three tasks to do first
- Identify what should be delegated
- Identify what should be dropped entirely, with reasoning
- Be willing to recommend dropping things outright, not just deprioritising
- Reasoning must tie back to the stated goals
- No more than three 'do first' items
A prioritised plan: top three tasks, delegate list, drop list with reasoning.
See example AI output
Against your stated Q3 goal of stabilising the JB hub launch: Do first - (1) finalise the hub staffing plan, (2) sign off the courier rate contract, (3) review the go-live checklist with ops. Delegate - vendor onboarding paperwork to your ops coordinator; it's necessary but doesn't need your judgement. Drop - the rebrand of internal delivery-tracking emails; it doesn't move the JB launch goal and can wait until Q4 with no real cost. Also drop the fourth conference you'd pencilled in for August - travel time competes directly with the launch window.
Focused Weekly Plan
focus planner
your goals and commitments for the week
- Identify the three outcomes that matter most this week
- Block deep-work time to protect progress on those outcomes
- Identify what to explicitly say no to this week
- Realistic - must fit within an actual working week
- Deep-work blocks must be protected, not just listed as ideals
- Must name specific things to decline, not a vague intention
A weekly plan: three priority outcomes, protected deep-work blocks, and a say-no list.
See example AI output
Week of July 28 for the Nexora launch push: Three outcomes - (1) finalise the partner conference agenda, (2) close the Q3 hiring plan with HR, (3) review and approve the client migration runbook. Deep-work blocks: Tuesday 9-11am and Thursday 2-4pm, calendar-blocked and declined to all but urgent client escalations. Say no to: the vendor demo request (delegate to Wei Ling), the internal town hall prep committee (defer involvement to next month), and any non-essential 30-minute 'quick sync' meeting request this week.
Devil's Advocate Stress Test
devil's advocate
no file - describe your current thinking on the topic
- State the current thinking back clearly before challenging it
- Surface the blind spots in the reasoning
- Identify disconfirming evidence being ignored
- Construct the view a smart critic would genuinely hold
- State what this means for the actual decision
- Rigorous, not contrarian for its own sake
- Must be grounded in real disconfirming evidence, not invented objections
- End with practical implications, not just critique
The strongest counter-case to the current thinking, followed by what it means for the decision.
See example AI output
Your thinking: Meridian should exit the low-margin FMCG category entirely to focus on higher-margin specialty lines. Blind spot: FMCG volume is what justifies your current warehouse lease and truck fleet utilisation - losing it could push fixed costs per unit up on the specialty side too. Disconfirming evidence you're not weighting: last year's cross-sell data shows 30% of specialty customers were acquired through FMCG relationships first. A smart critic would argue you're solving a margin problem by cutting the acquisition channel for your most profitable segment. Implication: consider a phased FMCG wind-down, not an immediate exit.
More prompt packs by function
View the full Prompt LibraryFrequently Asked Questions
As a rigorous thinking partner and drafting engine — structuring decisions with real frameworks, stress-testing plans, preparing for board and difficult conversations, and sharpening communication. Each prompt assumes you will attach the report, data or plan, so the AI reasons over your reality, not generalities.
Use your organisation's approved, enterprise-grade AI for anything confidential so data stays in your control. Avoid pasting sensitive strategy or financials into public consumer tools.
Copilot is convenient inside Microsoft 365 for email, decks and Teams; Claude is strong for reasoning and long-document analysis; ChatGPT and Gemini are widely used too. These prompts work across all of them.
Both — but leaders who use it personally make far better decisions about adopting it across the organisation. Hands-on fluency separates AI-ready leadership from AI theatre, which is what our executive-level training builds.
Yes. AITraining2U's courses are HRD Corp SBL-KHAS claimable for eligible Malaysian employers.
AI gets you 70-80% of the way - a well-structured first draft with the right sections, tone and no blank-page problem - but board papers carry real weight, so treat AI output as a strong draft, never a final submission. Always verify every number against source data, sense-check the recommendation against context the AI doesn't have (board politics, prior commitments, unstated sensitivities), and have a second human read it before it reaches the boardroom. The time AI saves should go into sharpening judgement, not skipping review.
Only in an approved enterprise environment where your organisation controls data retention and training use - never in a free consumer chatbot. Strategic plans, M&A discussions, restructuring details and unreleased financials are exactly the material competitors or the market should never see, and consumer-tier tools may use conversations to improve their models. Check your company's AI data policy before attaching anything sensitive, and when in doubt, describe the situation in generic terms rather than pasting the real document.
AI should never make the decision - it should sharpen the thinking that leads to one. Use it to generate options you hadn't considered, stress-test your assumptions, and surface evidence you might have missed, but the judgement call - weighing risk, reading the room, owning the consequence - stays with the leader. Treat AI recommendations the way you'd treat a smart junior analyst's memo: useful input, not a verdict. The prompts here are deliberately built to expose trade-offs, not hide them behind a confident-sounding answer.
Yes for routine communications - updates, meeting recaps, standard emails - where a quick read-and-edit is faster than drafting from scratch. Be more hands-on for anything emotionally sensitive (layoffs, difficult feedback, crisis comms) or reputationally risky, where tone and timing matter more than structure and only you have the full context. A good habit: use AI for the first draft always, but read every word aloud before sending anything with your name or your principal's name on it.
It works well for structured moments - turning a discussion into a decision log in real time, drafting an agenda beforehand, or synthesising notes immediately after - but avoid running it live as a substitute for actually listening and reading the room. If you use it during the meeting, say so openly rather than quietly typing prompts, and never feed confidential discussion into a tool the rest of the room hasn't agreed to. Best practice: use AI to prepare before and process after, not to think for you during.
Go beyond prompts — train your team
Prompts are the start. AITraining2U runs hands-on, HRD Corp SBL-KHAS claimable AI training for Malaysian teams — from everyday AI productivity to building agents that run executives & managers workflows end-to-end.