35+ AI Prompts for Supply Chain Management

Why Supply Chain Needs Structured AI Prompts

Supply chain management is the most constraint-dense professional domain we measure on AI Prompt Architect. In our analysis of 2,400 supply chain prompts processed between January and June 2026, prompts with explicit constraint parameters — lead times, MOQs, safety stock levels, carrier capacity — produce 3.2x more actionable outputs than unconstrained prompts (AI Prompt Architect platform data, Q2 2026). The delta is larger than any other vertical we track.

Supply chain prompts represent 7% of all professional prompts on our platform, but they carry the highest average complexity score of any industry vertical: 74 out of 100. That complexity stems from the multi-variable, constraint-heavy nature of the domain. A demand planning prompt must account for seasonality, lead time variability, supplier reliability, and demand sensing signals simultaneously. A logistics prompt must balance vehicle capacity, time windows, driver hours, and cost constraints. When any of these parameters are missing, the output degrades from actionable to academic.

The most common failure mode is surprisingly basic. Our analysis reveals that 63% of supply chain prompts fail to specify units of measure — pallets versus cases versus eaches — leading to outputs that require complete rework. The difference between "order 500" (pallets? cases? eaches?) is the difference between a replenishment order and a warehouse crisis. Precision matters more in supply chain prompting than in any other domain we measure.

The STCO framework (Situation, Task, Context, Output) addresses these failure modes systematically. In supply chain applications, the Situation defines your operational context and data environment; the Task specifies your analytical objective; the Context encodes your constraints, units, and business rules; and the Output defines the exact deliverable format. When applied consistently, STCO-structured supply chain prompts produce outputs that are not merely informative but operationally implementable.

Model Benchmarks for Numerical Tasks

Not all models handle supply chain calculations equally. Our benchmarking across numerical supply chain tasks reveals meaningful performance differences:

These benchmarks are based on our standardised supply chain test suite of 340 prompts. For routine analytical tasks, Gemini 2.5 Pro offers the strongest balance of accuracy and speed. For complex multi-step calculations where latency is acceptable, o3 leads on raw numerical accuracy. For nuanced risk narratives and scenario analysis, Claude 4.8's constraint handling is strongest. As detailed in our complete prompting tools guide, model selection should be driven by task type, not brand preference.

Demand Planning Prompts

Demand-forecasting prompts that include 24 or more months of historical data context and specify seasonality patterns produce forecasts with 22% lower mean absolute percentage error (MAPE) in our validation testing (AI Prompt Architect benchmark, n=560). In a domain where a 1% forecast improvement can translate to millions in reduced inventory costs, prompt quality is directly P&L relevant.

Prompts that request confidence intervals alongside point forecasts are rated "decision-ready" by supply chain planners 2.8x more often than prompts that generate point estimates alone. The reason is operational: planners need to set safety stock levels, and a point forecast without uncertainty quantification forces them to guess at variability. This holds for English-language, single-market forecasts; multi-market or multilingual datasets require additional calibration.

1. Demand Forecast Synthesis

Situation: You are a demand planning analyst for a [consumer goods / industrial / retail] company operating in [market/region]. You have access to [24-60] months of weekly sales history for [product category], with data stored in [units: cases/eaches/pallets].

Task: Generate a 12-month rolling demand forecast with monthly granularity, decomposing the signal into trend, seasonality, and residual components.

Context: Key seasonality drivers include [list: e.g., summer peak, back-to-school, holiday]. Known demand disruptions in the historical period: [list any outliers to exclude or weight]. Forecast should account for [price elasticity / promotional calendar / new distribution points].

Output: A monthly forecast table with columns: Month, Point Forecast ([units]), Lower Bound (90% CI), Upper Bound (90% CI), YoY Growth %. Include a narrative summary of the top 3 forecast risks.

2. Seasonal Decomposition

Situation: You are analysing the seasonal pattern of [SKU/category] sales across [number] years of [weekly/monthly] data in [units].

Task: Decompose the time series into trend, seasonal, and irregular components using [additive/multiplicative] decomposition.

Context: Seasonal indices should be normalised to a base of 1.0. Flag any periods where the irregular component exceeds 2 standard deviations — these likely represent promotional lifts or supply disruptions rather than organic demand.

Output: A table of seasonal indices by period, a trend equation, and a list of flagged irregular periods with magnitude and probable cause.

3. New Product Launch Forecast

Situation: You are forecasting demand for a new product launching in [market] on [date]. No direct sales history exists. Comparable products in the portfolio include [list analogues with their launch trajectories].

Task: Generate a 6-month launch forecast using analogue-based forecasting, adjusting for [distribution breadth, price point differential, marketing spend differential].

Context: Analogue product A achieved [X units] in month 1, scaling to [Y units] by month 6. Distribution for the new product will be [Z]% of analogue A's initial distribution. Marketing investment is [ratio] of analogue.

Output: Monthly forecast in [units], with adjustment factors documented. Include an optimistic (+20%), base, and pessimistic (-30%) scenario.

4. Promotional Uplift Modelling

Situation: You are modelling the demand uplift for a [promotion type: price reduction / BOGO / display / feature] on [product] in [retailer/channel].

Task: Estimate the incremental volume uplift and post-promotion dip, factoring in [cannibalisation of adjacent SKUs / pantry loading / forward buying behaviour].

Context: Historical promotions of this type have generated [X-Y]% uplift with a [Z]-week post-promotion trough of [W]%. Base volume is [N units/week].

Output: Week-by-week forecast covering pre-promo baseline (2 weeks), promotion period, and post-promo recovery (4 weeks). All figures in [units]. Include net incremental volume after cannibalisation.

5. Demand Sensing from POS Data

Situation: You have access to [daily/weekly] point-of-sale data from [number] retail locations for [product category]. Current statistical forecast (generated [date]) shows [X units] for the next [period].

Task: Compare POS sell-through velocity against the current forecast and recommend adjustments to the near-term (4-week) demand signal.

Context: POS data covers [X]% of total distribution. Known biases: [list, e.g., over-representation of urban stores]. Current inventory position is [X weeks of cover].

Output: Adjusted 4-week demand signal in [units/week], variance to current forecast (%), and a confidence assessment (high/medium/low) with rationale.

6. Consensus Forecast Reconciliation

Situation: You are facilitating a consensus forecast review for [product line] covering [period]. Three forecast inputs exist: Statistical forecast ([X units]), Sales team input ([Y units]), Marketing input ([Z units]).

Task: Analyse the variance between inputs, identify the key assumptions driving each divergence, and recommend a consensus number with documented rationale.

Context: Historical forecast accuracy by source: Statistical = [MAPE]%, Sales = [MAPE]%, Marketing = [MAPE]%. Known upcoming events: [list].

Output: Reconciliation table showing each input, variance, key assumption, and recommended consensus figure. Include a bias-adjustment note if any source has a consistent directional bias.

7. Forecast Accuracy Retrospective

Situation: You are conducting a quarterly forecast accuracy review for [business unit/category] covering [quarter]. Actual sales data and corresponding forecasts are available at [SKU/category] level in [units].

Task: Calculate MAPE, bias (over/under forecast percentage), and weighted MAPE by revenue contribution. Identify the top 5 SKUs with the largest absolute forecast error.

Context: Acceptable MAPE threshold for this category is [X]%. Previous quarter's MAPE was [Y]%.

Output: Summary table with aggregate metrics, top 5 error drivers with root-cause hypotheses (demand shift, supply disruption, promotion timing, data quality), and 3 specific recommendations for improving next quarter's accuracy.

8. ABC-XYZ Demand Segmentation

Situation: You are segmenting [number] SKUs by revenue contribution (ABC) and demand variability (XYZ) for [product line/business unit]. Revenue and demand variance data covers [12-24] months.

Task: Classify each SKU into one of 9 segments (AX, AY, AZ, BX, BY, BZ, CX, CY, CZ) and recommend differentiated planning strategies per segment.

Context: ABC thresholds: A = top 80% cumulative revenue, B = next 15%, C = remaining 5%. XYZ thresholds: X = CV < 0.5, Y = CV 0.5-1.0, Z = CV > 1.0.

Output: Segmentation matrix with SKU counts per cell, recommended forecast method per segment, recommended safety stock approach per segment, and a list of CZ items flagged for SKU rationalisation review.

AI-generated forecasts are inputs to your S&OP process, not replacements for demand planner judgement. These templates produce structured starting points that reduce preparation time — they do not eliminate the need for market intelligence that only your planning team possesses.

Inventory Management Prompts

Safety stock calculation prompts that specify service-level targets (e.g., 98.5% fill rate) and demand variability parameters produce outputs 47% more aligned with actual reorder points than prompts without these constraints (AI Prompt Architect platform data, Q2 2026). Inventory-optimisation prompts that include carrying-cost percentages generate recommendations saving an estimated 14% on holding costs compared to prompts without cost context.

1. Safety Stock Calculator

Situation: You are calculating safety stock for [SKU/category] with the following parameters: average daily demand = [X units], demand standard deviation = [Y units/day], average lead time = [Z days], lead time standard deviation = [W days].

Task: Calculate the required safety stock to achieve a [target]% service level (cycle service level), using the combined demand and lead time variability formula.

Context: Current safety stock is [N units]. Storage capacity constraint: maximum [M units]. Units of measure: [eaches/cases/pallets]. Z-score for target service level: [value].

Output: Safety stock quantity in [units], the formula used with all intermediate calculations shown, reorder point, and a sensitivity table showing safety stock at 95%, 97%, 98.5%, and 99% service levels.

2. EOQ Analysis

Situation: You are optimising order quantities for [SKU] with: annual demand = [D units], ordering cost = [S per order], holding cost = [H per unit per year], unit cost = [C].

Task: Calculate the Economic Order Quantity, annual ordering cost, annual holding cost, and total inventory cost. Compare against current order quantity of [Q units].

Output: EOQ in [units], cost comparison table (current vs. EOQ), number of orders per year, average inventory level, and potential annual savings.

3. ABC Classification

Situation: You are classifying [number] SKUs for [warehouse/business unit] by annual consumption value. Data includes annual unit sales and unit cost for each SKU.

Task: Perform ABC classification and recommend cycle counting frequencies and inventory control policies per class.

Output: Classification summary table (class, SKU count, % of SKUs, % of value), recommended policies per class (review frequency, counting method, reorder approach).

4. Dead Stock Identification

Situation: You are reviewing inventory for [warehouse] to identify dead and slow-moving stock. Inventory data includes: SKU, current stock quantity [units], last sale date, average monthly sales (trailing 12 months), unit cost, total inventory value.

Task: Flag all SKUs with zero sales in the last [90/180/365] days and calculate the carrying cost of dead inventory.

Output: Dead stock list sorted by inventory value (descending), total dead stock value, carrying cost estimate (using [X]% annual holding cost rate), and recommended disposition strategy per SKU (liquidate, return to supplier, donate, scrap).

5. Multi-Echelon Optimisation

Situation: You are optimising inventory placement across a [number]-echelon network: [e.g., factory warehouse, regional DC, local hub]. Service-level target: [X]% at the customer-facing echelon.

Task: Recommend inventory allocation by echelon that minimises total system inventory while meeting service targets, considering [lead times between echelons, demand variability at each node, transportation costs].

Output: Recommended stock levels per echelon in [units], total system inventory vs. current state, expected service level per echelon, and estimated carrying cost reduction.

6. Inventory Turns Analysis

Situation: You are analysing inventory turnover for [business unit] across [number] SKUs. Data covers [period]. COGS = [value], average inventory value = [value].

Task: Calculate overall inventory turns, days of inventory on hand, and identify the bottom 20% of SKUs by turnover rate.

Output: Aggregate metrics, distribution of turns across SKU base, bottom 20% list with individual turn rates and recommendations, and benchmark comparison against industry average of [X] turns.

7. SKU Rationalisation

Situation: You are evaluating [number] SKUs for potential rationalisation in [category/business unit]. Data includes: revenue contribution, margin, inventory turns, demand variability (CV), customer concentration, and substitutability score.

Task: Apply a weighted scoring model to identify the bottom [10-20]% of SKUs for rationalisation review, balancing financial performance against strategic considerations.

Output: Scored SKU list, rationalisation candidates flagged, estimated inventory reduction, estimated margin impact of discontinuation, and recommended substitution mapping.

8. Warehouse Slotting Optimisation

Situation: You are optimising pick-slot assignments for [warehouse] with [number] active SKUs across [number] pick locations. Current pick data shows average pick path of [X metres/feet] per order.

Task: Recommend slot reassignments based on pick frequency (velocity), product affinity (frequently co-picked items), and ergonomic considerations (weight, size).

Output: Top 20 recommended slot moves, estimated reduction in pick path distance, before/after slot map for the highest-velocity zone, and implementation sequence (move heavy movers first during low-volume shift).

Always validate AI inventory recommendations against your ERP's actual demand variability data. These templates produce analytically sound starting points, but real-world constraints — supplier reliability shifts, promotional calendars, warehouse capacity limits — require human calibration before implementation.

Procurement Prompts

Supplier evaluation prompts using weighted-criteria matrices score 41% higher on decision-quality metrics than unstructured "evaluate this supplier" prompts (AI Prompt Architect platform data, Q2 2026). RFQ-generation prompts that include specification tolerances and quality requirements produce documents requiring 35% fewer clarification rounds with suppliers.

Security Notice: Our Security Scanner flags commercial supplier data in 28% of procurement prompts. Never share actual supplier pricing, contract terms, or proprietary cost structures in AI prompts. Use indexed figures, anonymised supplier codes, and relative cost ratios. The analytical quality of the output is equivalent; the data-leakage risk drops to near zero.

1. Supplier Scorecard

Situation: You are evaluating [number] suppliers for [commodity/category]. Evaluation criteria include: quality (defect rate), delivery (OTIF %), cost competitiveness (indexed to market benchmark), capacity flexibility, financial stability, and ESG compliance.

Task: Create a weighted scorecard with the following weights: Quality [X]%, Delivery [X]%, Cost [X]%, Capacity [X]%, Financial [X]%, ESG [X]%. Score each supplier on a 1-5 scale per criterion.

Output: Scorecard table with weighted scores, overall ranking, gap analysis for the lowest-scoring criterion per supplier, and recommended actions (develop, maintain, phase out, single-source risk flag).

2. RFQ Generator

Situation: You are preparing a Request for Quotation for [commodity/service] for [company type]. Annual volume requirement: [X units]. Delivery locations: [list]. Contract period: [duration].

Task: Generate a structured RFQ document including specification requirements, quality standards, delivery terms, pricing structure requirements, and evaluation criteria.

Output: Complete RFQ document with: technical specifications (with tolerances), quality requirements (inspection criteria, certification requirements), delivery schedule, pricing table template (unit price, tooling, NRE, logistics), terms and conditions summary, and supplier response deadline.

3. Total Cost of Ownership Analysis

Situation: You are comparing [number] sourcing options for [component/material]. Unit prices are [Supplier A: indexed at 1.00, Supplier B: indexed at X, Supplier C: indexed at Y].

Task: Calculate the total cost of ownership including: unit price, inbound logistics, customs/duties, quality costs (inspection, scrap rate), inventory carrying costs, and supplier management overhead.

Output: TCO comparison table with all cost elements, TCO per unit, TCO ranking (which may differ from unit-price ranking), and a sensitivity analysis on the top 2 cost drivers.

4. Make vs. Buy Framework

Situation: You are evaluating whether to manufacture [component] internally or source externally. Internal capacity is [available/constrained]. Current internal cost: [indexed]. Best external quote: [indexed].

Task: Conduct a make vs. buy analysis considering: variable costs, fixed cost absorption, quality control, IP protection, supply continuity, and strategic importance.

Output: Decision matrix with quantitative (cost comparison) and qualitative (risk, strategic) factors, recommendation with rationale, and transition plan if recommending a change from current state.

5. Contract Risk Assessment

Situation: You are reviewing a [contract type] with [supplier code — anonymised] for [commodity/service]. Contract value: [indexed]. Duration: [period]. Key terms include [summary of material terms].

Task: Identify the top 10 risk factors in this contract arrangement, assess probability and impact of each, and recommend mitigation clauses.

Output: Risk register table (risk, probability, impact, risk score, mitigation clause recommendation), overall contract risk rating (low/medium/high), and 3 priority negotiation points.

6. Supplier Diversification Strategy

Situation: You are assessing single-source risk for [number] commodities currently sourced from sole suppliers. Total annual spend on single-sourced items: [indexed value].

Task: Prioritise commodities for supplier diversification based on: supply risk (sole source, geographic concentration), business impact (revenue at risk if supply fails), switching cost, and qualification timeline.

Output: Prioritised diversification roadmap, estimated timeline and qualification cost per commodity, recommended dual-sourcing ratio (e.g., 70/30), and risk-reduction quantification.

7. Should-Cost Model

Situation: You are building a should-cost model for [component/assembly]. Bill of materials includes [number] elements. Manufacturing process involves [list key processes].

Task: Estimate the should-cost based on: raw material costs (indexed to commodity indices), labour content (hours × rate), manufacturing overhead, SG&A margin, and profit margin.

Output: Cost breakdown structure (material, labour, overhead, SG&A, profit), should-cost estimate, comparison to current purchase price (indexed), and negotiation leverage points where current price exceeds should-cost by more than [X]%.

8. Sustainable Procurement Criteria

Situation: You are developing sustainability criteria for [commodity/category] procurement in alignment with [company sustainability targets / regulatory requirements]. Current Scope 3 emissions for this category: [data if available].

Task: Define weighted sustainability evaluation criteria covering: carbon footprint (Scope 3), water usage, waste reduction, labour practices, circular economy contribution, and certification requirements.

Output: Sustainability scorecard template, minimum qualification thresholds, preferred certification list (e.g., ISO 14001, EcoVadis, SA8000), and integration guidance for incorporating sustainability scores into the existing supplier evaluation process.

Logistics and Distribution Prompts

Route-optimisation prompts that include constraint parameters (vehicle capacity, time windows, driver hours) produce solutions 2.4x closer to optimal versus unconstrained prompts, as validated against OR-Tools benchmarks (AI Prompt Architect benchmark, n=280). Last-mile delivery prompts specifying customer density and time-window constraints generate 31% more feasible route plans. AI route optimisation is a starting point — real-world constraints (traffic, weather, driver preferences) require human adjustment.

1. Route Optimisation Brief

Situation: You are planning delivery routes for [number] stops across [area/region]. Fleet: [number] vehicles, each with [capacity in units/weight/volume]. Operating hours: [start]-[end]. Driver break requirements: [regulation].

Task: Generate optimised route assignments that minimise total distance while respecting all constraints: vehicle capacity, customer time windows, driver hours regulations, and [any additional constraints].

Output: Route plan per vehicle (stop sequence, estimated arrival times, load utilisation %), total distance and time per route, utilisation summary, and flagged stops that could not be feasibly scheduled within constraints.

2. Carrier Selection Matrix

Situation: You are evaluating [number] carriers for [lane/region/mode]. Shipment profile: [volume/frequency/weight characteristics]. Service requirements: [transit time, temperature control, handling requirements].

Task: Create a weighted carrier evaluation matrix covering: rate competitiveness (indexed), on-time performance, damage/claims rate, capacity availability, technology capabilities (tracking, EDI), and sustainability credentials.

Output: Scored matrix with overall rankings, recommended primary and backup carrier per lane, and estimated annual cost at each carrier's rate (indexed to lowest bidder = 1.00).

3. Warehouse Network Design

Situation: You are evaluating the optimal warehouse network configuration for [company type] serving [number] customers across [geography]. Current network: [describe current DCs]. Demand pattern: [centralised/dispersed].

Task: Analyse the trade-offs between [number]-DC and [number]-DC configurations, considering: transportation costs, inventory carrying costs (safety stock duplication), facility costs, and service level (order-to-delivery time).

Output: Comparison table of network configurations with total landed cost estimates (indexed), service level impact, inventory investment implications, and a recommended configuration with rationale.

4. Last-Mile Optimisation

Situation: You are optimising last-mile delivery for [number] daily deliveries in a [urban/suburban/rural] area. Customer density: [orders per square km]. Delivery time windows: [X]% require specific slots.

Task: Recommend a last-mile delivery model (own fleet, gig economy, hybrid, parcel locker, click-and-collect) based on cost per delivery, customer satisfaction scores, and environmental impact.

Output: Model comparison table with cost per delivery (indexed), estimated customer satisfaction impact, carbon footprint per delivery, and recommended model with implementation roadmap.

5. Freight Audit Analysis

Situation: You are auditing [number] freight invoices from [period] totalling [indexed value]. Carrier contracts specify rates of [rate structure]. Historical overcharge rate: [X]%.

Task: Define the audit methodology, identify common overcharge categories, and estimate the recovery opportunity.

Output: Audit checklist (rate verification, accessorial validation, weight/dimension accuracy, duplicate detection), top 5 overcharge categories by estimated frequency, projected recovery value, and recommended audit frequency.

6. Cross-Docking Assessment

Situation: You are evaluating whether [product line/flow] is suitable for cross-docking at [facility]. Current process: receive → putaway → pick → ship. Average dwell time: [X days].

Task: Assess cross-docking feasibility based on: demand predictability, supplier delivery reliability, order profiles (full pallet vs. mixed), and facility layout constraints.

Output: Feasibility assessment (high/medium/low), estimated dwell time reduction, labour savings estimate, required supplier performance thresholds, and implementation prerequisites.

7. Reverse Logistics Planning

Situation: You are designing a reverse logistics process for [product type] with a current return rate of [X]%. Annual returns volume: [quantity]. Return reasons: [breakdown by category].

Task: Design a returns processing workflow covering: return initiation, transportation, inspection/grading, disposition (refurbish, resell, recycle, scrap), and value recovery tracking.

Output: Process flow diagram (text-based), disposition decision tree, estimated value recovery rate by return category, KPI framework (return processing time, recovery rate, cost per return), and technology requirements.

8. 3PL Performance Scorecard

Situation: You are evaluating the performance of [3PL provider] operating [facility/service] on your behalf. Contract KPIs include: [list current KPIs]. Review period: [quarter/year].

Task: Create a comprehensive 3PL scorecard covering: operational accuracy (order accuracy, inventory accuracy), service levels (OTIF, processing time), cost management (cost per unit handled, variance to budget), and strategic contribution (continuous improvement initiatives, technology adoption).

Output: Scorecard template with KPI definitions, measurement methodology, target vs. actual framework, weighted overall score, and governance rhythm (daily, weekly, monthly review cadence).

Risk and Resilience Prompts

Geopolitical risk prompts that specify affected trade corridors and alternative sourcing regions produce mitigation plans rated 3.1x more implementable by procurement directors in our evaluation (AI Prompt Architect platform data, Q2 2026). Supply chain disruption prompts that reference specific risk frameworks — ISO 31000, Supply Chain Risk Management (SCRM) — score 38% higher on completeness than framework-agnostic prompts.

1. Disruption Scenario Planning

Situation: You are developing disruption scenarios for [supply chain network]. Critical dependencies: [list top 5 dependencies — suppliers, routes, facilities]. Current business continuity plan was last updated [date].

Task: Develop 3 disruption scenarios (moderate, severe, catastrophic) with trigger events, impact duration estimates, and cascading effect analysis across the supply chain.

Output: Scenario table with: trigger event, probability assessment, affected nodes, estimated revenue impact (indexed), recovery timeline, and top 3 mitigation actions per scenario.

2. Geopolitical Risk Assessment

Situation: You are assessing the geopolitical risk exposure of [supply chain] with sourcing from [countries/regions] and primary trade corridors through [routes/ports].

Task: Identify the top 5 geopolitical risk factors affecting this supply chain, assess their probability and impact over a [12/24]-month horizon, and recommend diversification strategies.

Output: Risk matrix (factor, probability, impact, risk score), affected trade corridors, alternative sourcing regions with qualification timeline estimates, and a monitoring dashboard of leading indicators to track.

3. Supplier Risk Heat Map

Situation: You are mapping risk across [number] tier-1 suppliers for [business unit]. Risk dimensions: financial stability, geographic concentration, single-source dependency, regulatory compliance, and cyber security posture.

Task: Generate a supplier risk heat map categorising each supplier into red/amber/green across all dimensions.

Output: Heat map table, aggregate risk score per supplier, priority action list for red-rated suppliers, and recommended monitoring frequency per risk level.

4. Business Continuity Plan

Situation: You are updating the business continuity plan for [facility/function]. Critical processes: [list]. Maximum acceptable downtime: [hours/days]. Last disruption event: [describe].

Task: Generate a BCP covering: impact assessment, recovery priorities, alternative sourcing/production options, communication protocols, and testing schedule.

Output: BCP document outline with: process criticality ranking, recovery time objectives per process, alternative resource mapping, RACI for crisis response, and annual testing calendar.

ESG and Sustainability in Supply Chain

ESG/sustainability prompts that include Scope 3 emissions context and supplier-tier visibility generate 2.6x more specific decarbonisation recommendations than prompts without this context. As regulatory requirements tighten, sustainability prompting is shifting from voluntary to essential.

5. ESG Supply Chain Audit

Situation: You are conducting an ESG audit of [supply chain segment]. Scope: [tier 1 / tier 1-2 / full supply chain]. Regulatory requirements: [CSRD / SEC climate disclosure / other]. Current Scope 3 data coverage: [X]% of spend.

Task: Design an ESG audit framework covering: environmental (carbon emissions, water, waste), social (labour practices, health & safety, community impact), and governance (ethics, transparency, anti-corruption).

Output: Audit questionnaire template, scoring methodology, minimum compliance thresholds, data collection requirements per tier, and a roadmap for closing Scope 3 data gaps.

6. Single-Source Risk Analysis

Situation: [Component/material] is currently single-sourced from [supplier code — anonymised]. Annual spend: [indexed]. Lead time: [weeks]. No qualified alternative exists.

Task: Quantify the business risk of single-source dependency and develop a qualification roadmap for an alternative supplier.

Output: Risk quantification (revenue at risk, probability-weighted impact), qualification timeline and estimated cost, recommended dual-source ratio, and interim risk mitigation measures.

7. Climate Risk Impact Assessment

Situation: You are assessing climate-related physical and transition risks for [supply chain] with operations/sourcing in [regions]. Climate scenarios: [RCP 4.5 / RCP 8.5 or equivalent].

Task: Identify the top climate risks (physical: flooding, heat stress, water scarcity; transition: carbon pricing, regulation, technology shifts) and their potential impact on supply chain operations.

Output: Climate risk register, impact timeline (short/medium/long term), affected supply chain nodes, adaptation strategies, and TCFD-aligned disclosure recommendations.

8. S&OP Integration Framework

Situation: You are designing the risk integration layer for the [company] S&OP process. Current S&OP maturity: [stage 1-5]. Risk inputs are currently [ad-hoc/structured].

Task: Define how supply chain risk signals should be integrated into the monthly S&OP cycle, including: risk data inputs, scenario triggers, decision escalation criteria, and executive reporting format.

Output: S&OP risk integration process map, risk KPI dashboard specification, escalation matrix, and monthly risk review agenda template.

Geopolitical risk models degrade rapidly as conditions change. Refresh your risk assessment prompts quarterly at minimum, and update trade corridor specifications whenever significant regulatory or political changes occur.

Connecting Prompts to the S&OP Cycle

The templates above gain compound value when chained into your Sales and Operations Planning cycle. Rather than using each prompt in isolation, map them to your S&OP rhythm:

The compounding effect is measurable. Teams that chain supply chain prompts across S&OP phases produce more consistent planning outputs than teams that prompt ad hoc. The key is maintaining data consistency across the chain — the demand signal that feeds inventory planning must use the same units, time horizons, and confidence levels throughout.

How to Score and Optimise Your Supply Chain Prompts

Supply chain prompts show the highest improvement delta of any vertical after STCO restructuring: a 27-point average score increase on our 100-point scale, versus a 19-point cross-vertical average (AI Prompt Architect platform data, Q2 2026). The more constrained the domain, the more structured prompting pays dividends.

Use our Prompt Scorer to evaluate your supply chain prompts across four domain-specific dimensions:

The scoring dimensions are calibrated for operational decision-making, not academic analysis. A prompt that scores well on these four dimensions produces outputs your planning team can act on, not just read. After one STCO-guided iteration, prompts improve by a median of 19 points — and supply chain prompts improve by 27 points on average, reflecting the high-constraint nature of the domain. As covered in our complete business prompts guide, this iterative scoring approach applies across all professional verticals, but delivers the largest returns in supply chain.