Prompts • Skills • Connectors • Agents • Brand workflows • Tools
AI Playbook
for Consumer Product Brands
For operations, finance, supply chain, and e-commerce leaders at brands that sell through retailers, distributors, and online. A practical guide to demand planning, inventory across channels, retailer chargebacks and EDI, 3PLs, trade spend, and landed cost.
Who this playbook is for
This playbook is for brands that sell through retailers and distributors. It's for consumer product brands in apparel, beauty, supplements, home goods, and pet products that sell through retailers, distributors, marketplaces, and their own websites. It covers the operations and finance side of the business. Food and beverage brands have their own playbook.
- Operations and supply chain leaders
- Finance, controllers, and deductions teams
- Demand and inventory planners
- E-commerce and marketplace leaders
- Brands selling through retailers, distributors, Amazon, and DTC
- Food and beverage brands (see the Food & Beverage playbook)
- Retailers managing their own stores
- Marketing and creative work
- Consumers
Start With Your Role
Choose the role closest to your day-to-day work. Start with one useful prompt, make it repeatable, then add the information it needs.
- Start with: monthly operations review or OTIF root causes
- Save as a skill: metric definitions and review format
- Connect: NetSuite (read-only), 3PL and retailer reports
- First agent: OTIF & chargeback root-cause reporter
- Measure: fines, fill rate, cost to serve
- Start with: forecast review or weeks-of-supply
- Save as a skill: forecast rules and inventory policies
- Connect: ERP, Shopify, POS exports
- First agent: weeks-of-supply watch
- Measure: accuracy, stockouts, excess
- Start with: deduction categorization or SKU-channel margin
- Save as a skill: gross-to-net mapping and dispute rules
- Connect: ERP AR, remittances
- First agent: deduction triage & deadline watch
- Measure: recovery rate, deductions % of sales
- Start with: landed cost build-up or supplier scorecard
- Save as a skill: cost build-up and tariff scenario format
- Connect: PO and freight data
- First agent: landed cost variance monitor
- Measure: cost accuracy, on-time supply
- Start with: DTC weekly summary or listing health
- Save as a skill: weekly report format, claims rules
- Connect: Shopify connector (read-only), marketplace reports
- First agent: marketplace listing health monitor
- Measure: buy box, stockouts, margin after fees
- Start with: sell-through analysis or trade ROI
- Save as a skill: business review format
- Connect: retailer POS exports
- First agent: promotion post-event analyzer
- Measure: ROI, sell-through, retailer scorecard
100 AI Prompts for Consumer Product Brands
Ready-to-use prompts for demand planning, inventory, retailer EDI, chargebacks and deductions, 3PLs, trade spend, landed cost and tariffs, e-commerce, finance, and distributors. Copy one, add your brand's context, and review the result.
Use the reports your team already trusts, quote vendor guides and contracts exactly, and review the result before it reaches a retailer, supplier, or 3PL.
Prompts for forecasting by SKU and channel, reading retailer sell-through, and planning launches. AI drafts the forecast; planners own the number.
Forecasts drift when promotions and new doors aren't reflected. This catches mismatches before purchase orders go out.
Risks: History can mislead after a channel change. Control: Demand planner and channel owners agree before updating.
Sell-in without sell-through creates future deductions and returns. This shows the real demand.
Risks: POS data may lag or be incomplete. Control: Confirm with the retailer portal before acting.
Analog-based forecasts give launches a defensible starting point instead of a guess.
Risks: New products are uncertain. Control: S&OP team approves the buy; plan a fast reorder path.
Promotions are the biggest source of forecast error. History-based lift estimates reduce stockouts and excess.
Risks: Past lift may not repeat. Control: Compare to actuals after the event and update the model.
Measuring bias shows whether the team is consistently too optimistic or too cautious.
Risks: Metrics can be gamed. Control: Use a consistent, documented method.
A focused pre-read turns the meeting into decisions, not data review.
Risks: Numbers must reconcile across teams. Control: Finance confirms financial impact.
Seasonal peaks need decisions months earlier than they feel urgent.
Risks: Past seasons may not repeat. Control: Review against current retailer commitments.
Tail SKUs consume working capital and planning time.
Risks: Some SKUs anchor retailer relationships. Control: Sales and leadership decide.
Early signals can show demand shifts before sales do.
Risks: Signals are noisy. Control: Treat as input, not forecast.
Clear collaboration reduces surprise orders and stockouts.
Risks: Commitments to retailers matter. Control: Sales leader approves anything that sounds like a commitment.
A visible log makes it clear why the forecast is what it is.
Risks: Notes may be incomplete. Control: Owners confirm their assumptions.
12 Claude-Ready Consumer Brand Operations Skills
These are downloadable Claude Skill packages, not just longer prompts. Each includes a SKILL.md your team can tailor with your retailers, rules, and formats. A copy-and-paste version works in ChatGPT and other AI tools.
Connect Your Work
AI can only help with the information it can see. Start with one system, see what it can do, then add more when you are ready.
Easy start
Start with a limited, low-risk connection your team can test quickly.
- Access: an official Claude connector for Shopify that uses Shopify's MCP servers; it can read orders, customers, and analytics, and with write scopes can edit products, prices, inventory, and discounts
- Useful for: DTC performance questions, stockout checks, and order lookups
- Setup & limit: grant read-only scopes first; add write scopes only for a specific, approved task
- Access: Gmail, Drive, Sheets, and Calendar connections in Claude and ChatGPT
- Useful for: vendor guides, rate cards, and planning sheets
- Setup & limit: start with operations folders; keep customer data out of scope
- Access: Outlook, SharePoint, OneDrive, Excel, and Teams connections, or Microsoft Copilot in your tenant
- Useful for: planning files, contracts, and S&OP materials
- Setup & limit: admin controls access; start with one site
- Access: the official Slack connection for Claude and ChatGPT
- Useful for: S&OP and operations channel summaries
- Setup & limit: exclude channels with personnel or confidential deal discussions
Needs an administrator
Useful, but someone needs to set access and permissions first.
- Access: NetSuite's MCP-based connector for Claude; requires an admin to enable features and set a limited role
- Useful for: inventory, order, and financial questions from your ERP
- Setup & limit: use a narrow, read-focused role; the connector can't run as Administrator
- Access: AI built into the SPS network that flags EDI anomalies and suggests fixes in advisory mode; no outside AI connector
- Useful for: catching EDI exceptions inside your existing SPS workflows
- Setup & limit: a person approves any transaction; check plan availability
- Access: AI access varies by product, plan, and country
- Useful for: margin and AR questions for smaller brands
- Setup & limit: keep postings and payments with finance
Integration project
Plan the use case, source data, permissions, and owner before connecting.
- Access: no official direct Claude or ChatGPT connection is confirmed here; Amazon offers APIs to approved developers
- Useful for: scheduled report exports (inventory health, fees, listing health) into a shared folder
- Setup & limit: use approved apps or exports; never share account credentials with AI tools
- Access: Walmart Retail Link/Luminate, Target Partners Online, and others are portals, not AI connections
- Useful for: downloaded POS and scorecard reports
- Setup & limit: follow each retailer's data-sharing terms
- Access: 3PL portals and WMS systems vary; some offer APIs
- Useful for: activity and invoice data for audits and scorecards
- Setup & limit: agree on data feeds in your 3PL contract
Good to know
Limits and honest gaps.
- Access: customs entries and classifications are handled by licensed customs brokers
- Useful for: broker-provided rates for landed cost and tariff models
- Setup & limit: AI never classifies products or decides duty treatment
- Access: food and beverage brands have separate regulatory and supply chain needs
- Useful for: not covered here; see the Food & Beverage playbook
- Setup & limit: this playbook covers apparel, beauty, supplements, home goods, and pet products
Tools
Choose the AI tools and operations software that fit your brand. The goal is to make your existing systems easier to use, not replace everything at once.
AI Assistants
5ERP & Commerce
5EDI & Retail Compliance
6Fulfillment & 3PL
5Finance
4Collaboration
412 AI Agents for Consumer Product Brands
An agent is an assistant for one recurring job. It gathers the right information, follows your process, and prepares the work for someone to review and decide.
A good first agent handles one task your team already does, reads data without changing it, never contacts partners, and hands its work to a named person.
# Forecast Exception Monitor ## Job to be done Catch forecast mismatches before they become purchase orders. ## When it runs Weekly after POS and sales data refresh. ## Approved context Sales history, forecast, promotion calendar, retailer POS. ## Operating loop Compare forecast to trend and sell-through; flag SKUs over threshold and promotions not reflected; draft owner questions. ## What it delivers A forecast exception list. ## Human owner Demand planner. ## What it must never do Never change the forecast or publish numbers. ## How to measure it Forecast accuracy, bias, exceptions resolved. ## Operating rules 1. Use only approved sources and follow the linked brand operations skill. 2. Cite source records and separate facts from assumptions. 3. Route material variances, dispute deadlines, and policy exceptions to the human owner. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when contract terms, retailer rules, or approval authority are unclear. 6. Use connectors read-only unless the owner has explicitly approved a write action.
# Weeks-of-Supply Watch ## Job to be done Spot stockout and excess risk across channels early. ## When it runs Daily after inventory feeds update. ## Approved context Inventory by location, forecasts, in transit, on order. ## Operating loop Compute weeks of supply by SKU and channel; flag stockout and excess risk; suggest rebalancing and reorders. ## What it delivers A daily inventory risk list. ## Human owner Inventory planner. ## What it must never do Never create POs or transfers. ## How to measure it Stockout days, excess inventory, turns. ## Operating rules 1. Use only approved sources and follow the linked brand operations skill. 2. Cite source records and separate facts from assumptions. 3. Route material variances, dispute deadlines, and policy exceptions to the human owner. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when contract terms, retailer rules, or approval authority are unclear. 6. Use connectors read-only unless the owner has explicitly approved a write action.
# Retailer PO & ASN Checker ## Job to be done Stop PO and ASN problems before they become chargebacks. ## When it runs When a new PO arrives; before each ASN is sent. ## Approved context EDI documents, price list, inventory, vendor guide rules. ## Operating loop Check PO price, SKU, quantity, window, and ship-to; check ASN against pick data; flag exceptions. ## What it delivers A PO and ASN exception queue. ## Human owner Order management lead. ## What it must never do Never acknowledge POs or send ASNs. ## How to measure it ASN error rate, PO exceptions, compliance fines. ## Operating rules 1. Use only approved sources and follow the linked brand operations skill. 2. Cite source records and separate facts from assumptions. 3. Route material variances, dispute deadlines, and policy exceptions to the human owner. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when contract terms, retailer rules, or approval authority are unclear. 6. Use connectors read-only unless the owner has explicitly approved a write action.
# Deduction Triage & Deadline Watch ## Job to be done Recover invalid deductions before dispute windows close. ## When it runs Daily when remittances post; weekly deadline review. ## Approved context Remittance and deduction data, POs, PODs, ASNs, promotion agreements, dispute rules. ## Operating loop Categorize new deductions; assess validity; assemble evidence; list deductions near deadlines. ## What it delivers A deduction worklist with dispute packages. ## Human owner Deductions manager. ## What it must never do Never submit disputes or write off balances. ## How to measure it Recovery rate, deductions past deadline, days to dispute. ## Operating rules 1. Use only approved sources and follow the linked brand operations skill. 2. Cite source records and separate facts from assumptions. 3. Route material variances, dispute deadlines, and policy exceptions to the human owner. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when contract terms, retailer rules, or approval authority are unclear. 6. Use connectors read-only unless the owner has explicitly approved a write action.
# OTIF & Chargeback Root-Cause Reporter ## Job to be done Show where compliance fines come from so they stop. ## When it runs Monthly. ## Approved context Chargeback data, OTIF reports, 3PL data. ## Operating loop Group fines by retailer, type, DC, and facility; trend; identify top causes; draft corrective actions. ## What it delivers A root-cause report. ## Human owner VP operations. ## What it must never do Never contact retailers or 3PLs. ## How to measure it Fines as % of sales, repeat causes, OTIF. ## Operating rules 1. Use only approved sources and follow the linked brand operations skill. 2. Cite source records and separate facts from assumptions. 3. Route material variances, dispute deadlines, and policy exceptions to the human owner. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when contract terms, retailer rules, or approval authority are unclear. 6. Use connectors read-only unless the owner has explicitly approved a write action.
# 3PL Invoice Auditor ## Job to be done Catch 3PL billing errors every month. ## When it runs When each 3PL invoice arrives. ## Approved context Invoice, activity data, rate card. ## Operating loop Match each line to rate card and activity; flag errors; total the disputed amount; update scorecard. ## What it delivers An audit report and scorecard. ## Human owner Logistics manager. ## What it must never do Never dispute, approve, or pay invoices. ## How to measure it Billing errors recovered, audit time. ## Operating rules 1. Use only approved sources and follow the linked brand operations skill. 2. Cite source records and separate facts from assumptions. 3. Route material variances, dispute deadlines, and policy exceptions to the human owner. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when contract terms, retailer rules, or approval authority are unclear. 6. Use connectors read-only unless the owner has explicitly approved a write action.
# Trade Deduction Matcher ## Job to be done Match every trade deduction to an approved promotion. ## When it runs Weekly. ## Approved context Deductions, approved promotions, accruals. ## Operating loop Match deductions to promotions; flag over-deductions and unmatched items; update accrual view. ## What it delivers A trade reconciliation report. ## Human owner Trade finance manager. ## What it must never do Never approve deductions or post accruals. ## How to measure it Unmatched trade dollars, accrual accuracy. ## Operating rules 1. Use only approved sources and follow the linked brand operations skill. 2. Cite source records and separate facts from assumptions. 3. Route material variances, dispute deadlines, and policy exceptions to the human owner. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when contract terms, retailer rules, or approval authority are unclear. 6. Use connectors read-only unless the owner has explicitly approved a write action.
# Promotion Post-Event Analyzer ## Job to be done Measure every promotion's ROI after it ends. ## When it runs Four weeks after each promotion ends. ## Approved context Sales and POS data, promotion terms, spend, margin. ## Operating loop Estimate baseline and lift; calculate incremental profit and ROI; summarize lessons. ## What it delivers A post-event ROI summary. ## Human owner Trade marketing manager. ## What it must never do Never change trade plans. ## How to measure it Promotions evaluated, ROI trend. ## Operating rules 1. Use only approved sources and follow the linked brand operations skill. 2. Cite source records and separate facts from assumptions. 3. Route material variances, dispute deadlines, and policy exceptions to the human owner. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when contract terms, retailer rules, or approval authority are unclear. 6. Use connectors read-only unless the owner has explicitly approved a write action.
# Landed Cost Variance Monitor ## Job to be done Flag when actual landed cost drifts from standard. ## When it runs When POs are received and freight and duty invoices post. ## Approved context PO costs, freight and duty invoices, standard costs, broker rates. ## Operating loop Calculate actual landed cost; compare to standard; flag variances by cause; flag tariff rate changes from broker updates. ## What it delivers A variance report. ## Human owner Cost accountant. ## What it must never do Never change standard costs or classify products. ## How to measure it Variance detected, standard cost accuracy. ## Operating rules 1. Use only approved sources and follow the linked brand operations skill. 2. Cite source records and separate facts from assumptions. 3. Route material variances, dispute deadlines, and policy exceptions to the human owner. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when contract terms, retailer rules, or approval authority are unclear. 6. Use connectors read-only unless the owner has explicitly approved a write action.
# Supplier Delivery Watch ## Job to be done Warn early when supplier shipments will miss need dates. ## When it runs Weekly; when a supplier updates a ship date. ## Approved context Open POs, production updates, transit times, need dates. ## Operating loop Compare expected arrival to need dates; flag late POs; estimate impact on weeks of supply. ## What it delivers A late-PO risk list. ## Human owner Sourcing manager. ## What it must never do Never contact suppliers or change POs. ## How to measure it On-time supplier delivery, stockouts from late POs. ## Operating rules 1. Use only approved sources and follow the linked brand operations skill. 2. Cite source records and separate facts from assumptions. 3. Route material variances, dispute deadlines, and policy exceptions to the human owner. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when contract terms, retailer rules, or approval authority are unclear. 6. Use connectors read-only unless the owner has explicitly approved a write action.
# Marketplace Listing Health Monitor ## Job to be done Keep marketplace listings live, competitive, and profitable. ## When it runs Daily. ## Approved context Listing health reports, buy box data, inventory, fees. ## Operating loop Flag suppressed listings, buy box loss, low inventory, and SKUs unprofitable after fees. ## What it delivers A daily listing health report. ## Human owner Marketplace manager. ## What it must never do Never change listings, prices, or ads. ## How to measure it Buy box %, suppressed listings, margin after fees. ## Operating rules 1. Use only approved sources and follow the linked brand operations skill. 2. Cite source records and separate facts from assumptions. 3. Route material variances, dispute deadlines, and policy exceptions to the human owner. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when contract terms, retailer rules, or approval authority are unclear. 6. Use connectors read-only unless the owner has explicitly approved a write action.
# Compliance Document Expiry Tracker ## Job to be done Keep product compliance documents current for retailers and regulators. ## When it runs Monthly; 60 days before expiry. ## Approved context Document inventory, requirements, SKU list. ## Operating loop Find missing and expiring documents; draft reminders; summarize by retailer request. ## What it delivers A document status report. ## Human owner Quality or regulatory lead. ## What it must never do Never create or alter compliance documents. ## How to measure it Expired documents, retailer requests met on time. ## Operating rules 1. Use only approved sources and follow the linked brand operations skill. 2. Cite source records and separate facts from assumptions. 3. Route material variances, dispute deadlines, and policy exceptions to the human owner. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when contract terms, retailer rules, or approval authority are unclear. 6. Use connectors read-only unless the owner has explicitly approved a write action.
AI for Demand & Inventory Planning
Use AI to review forecasts, read sell-through, and position inventory across channels. Planners own the numbers.
- Prompts: SKU-channel forecast review, sell-through analysis, multi-channel inventory position
- Skills: demand forecast review; multi-channel inventory & replenishment
- Connect: NetSuite AI Connector (admin setup), Shopify connector, retailer POS exports, planning tool
- Agent: forecast exception monitor; weeks-of-supply watch
- What AI does: Flags SKUs where forecast and trend diverge
- Helps: Fewer bad POs
- Control: Planners decide
- What AI does: Compares sell-in to POS
- Helps: Avoid future returns and deductions
- Control: Confirm in retailer portals
- What AI does: Uses analog curves for new SKUs
- Helps: Defensible initial buys
- Control: S&OP approves
- What AI does: Weeks of supply by channel and location
- Helps: Fix inventory in the wrong place
- Control: Check transfer costs
- What AI does: Calculates order quantities and dates
- Helps: Consistent buying
- Control: Buyers approve POs
- What AI does: Shows trade-offs for constrained SKUs
- Helps: Faster shortage decisions
- Control: Leadership decides
Planning Implementation Checklist
WorkflowBefore You Start
- Export SKU-channel history and forecasts
- Collect retailer POS access
- Document safety stock and allocation policies
- Baseline: forecast accuracy, stockouts, excess
- Set connectors read-only
After You Start
- Track accuracy and bias monthly
- Review exceptions weekly
- Compare launch forecasts to actuals
- Update assumptions log
- Review connector permissions quarterly
- Ownership: Planners and S&OP own the forecast; AI flags and drafts.
- Data agreements: Retailer POS data may be restricted by data-sharing terms.
- Purchases: AI never places or edits POs.
- Write access: Keep ERP and store connectors read-only unless an owner approves a specific write action.
- Accuracy: Verify every recommendation against the system of record.
AI for Retailer Orders, EDI & Chargebacks
Use AI to summarize vendor guides, catch PO and ASN errors, and recover deductions. Your team submits disputes and approves write-offs.
- Prompts: vendor guide summary, deduction categorization, dispute package
- Skills: retailer EDI & vendor compliance; deductions & chargeback recovery
- Connect: EDI provider reports (SPS Commerce MAX is built into SPS), retailer portals, ERP AR
- Agent: retailer PO & ASN checker; deduction triage & deadline watch
- What AI does: One-page rules by retailer
- Helps: Avoid common fines
- Control: Check current version
- What AI does: Flags errors before acknowledgment and shipment
- Helps: Fewer chargebacks
- Control: Order management approves
- What AI does: Sorts deductions by type and validity
- Helps: Systematic recovery
- Control: Verify before disputing
- What AI does: Assembles evidence in retailer format
- Helps: Win more disputes
- Control: Staff submit
- What AI does: Traces fines to facilities and processes
- Helps: Stop repeat fines
- Control: Confirm with 3PL
- What AI does: Matches deductions to approved promotions
- Helps: Stop margin leakage
- Control: Trade and AR review
Retail Compliance Checklist
ControlsBefore You Start
- Collect current vendor guides
- Export 12 months of deductions
- Organize POD and BOL access
- List dispute deadlines by retailer
- Baseline: deductions % of sales, recovery rate
After You Start
- Review dispute packages before submission
- Track deadlines weekly
- Report root causes monthly
- Measure recovery by category
- Update vendor guide summaries quarterly
- Submissions: AI never submits disputes or acknowledges POs.
- Honesty: Dispute only with real evidence; never alter documents.
- Write-offs: Controller approves.
- Confidentiality: Retailer portal data follows retailer terms.
- Partners: Corrective requests to 3PLs come from your operations leader.
AI for 3PLs & Fulfillment
Use AI to audit 3PL invoices, score performance, and plan peak. Your logistics team owns the relationship.
- Prompts: 3PL invoice audit, scorecard, peak readiness plan
- Skills: 3PL performance & invoice audit
- Connect: 3PL portal exports, WMS reports, parcel data
- Agent: 3PL invoice auditor
- What AI does: Matches charges to rate card and activity
- Helps: Recover billing errors
- Control: Manager approves disputes
- What AI does: Measures accuracy, speed, and fines caused
- Helps: Factual reviews
- Control: Agree on data
- What AI does: Models proposals on your volumes
- Helps: True cost comparison
- Control: Validate volumes
- What AI does: Capacity vs. demand by week
- Helps: Avoid peak failures
- Control: 3PL commits in writing
- What AI does: Return rates and dispositions
- Helps: Fix product and listing issues
- Control: Inspect samples
- What AI does: Dim weight and zone analysis
- Helps: Lower shipping cost
- Control: Review with carriers
3PL Management Checklist
WorkflowBefore You Start
- Collect contracts and rate cards
- Agree on activity data feeds
- Define scorecard targets
- Baseline: cost per order, accuracy
- Name the relationship owner
After You Start
- Audit every invoice
- Hold monthly reviews
- Track chargebacks caused by the 3PL
- Plan peak 4 months out
- Review contract annually
- Disputes: The logistics manager disputes invoices; AI prepares.
- Contracts: Rate cards and SLAs govern; quote them exactly.
- Regulated items: Hazmat and lot-controlled products need qualified partners.
- Data: Customer addresses stay in fulfillment systems.
- Relationships: Keep reviews collaborative and data-based.
AI for Trade Spend & Promotions
Use AI to plan trade calendars, measure ROI, and reconcile trade deductions. Sales and finance own spend decisions.
- Prompts: trade promotion ROI, trade deduction matching, gross-to-net waterfall
- Skills: trade spend planning & ROI; channel margin & gross-to-net
- Connect: trade promotion management tool, ERP, retailer POS
- Agent: trade deduction matcher; promotion post-event analyzer
- What AI does: Incremental profit per promotion
- Helps: Repeat what works
- Control: Consistent baselines
- What AI does: Calendars and spend by retailer
- Helps: Stay within budget
- Control: Leadership approves
- What AI does: Matches trade deductions to agreements
- Helps: Stop over-deductions
- Control: Trade finance reviews
- What AI does: Checks trade accruals at month-end
- Helps: Fewer margin surprises
- Control: Controller approves
- What AI does: Flags cross-channel price conflicts
- Helps: Protect retail relationships
- Control: Counsel on pricing policy
- What AI does: Waterfall by channel and retailer
- Helps: See revenue leakage
- Control: Finance confirms mapping
Trade Spend Control Checklist
ControlsBefore You Start
- Collect promotion agreements
- Set baseline method
- Document approval levels
- Map gross-to-net categories
- Baseline: trade % of gross sales
After You Start
- Evaluate every major promotion
- Reconcile deductions weekly
- Review accruals monthly
- Report ROI quarterly
- Update policy annually
- Approvals: Trade spend follows approval levels.
- Pricing law: Resale price and MAP policies have legal limits; counsel advises.
- Accruals: Controller approves entries.
- Retailer data: Use under data-sharing terms.
- Commitments: Only approved promotions are communicated to retailers.
AI for Landed Cost, Sourcing & Tariffs
Use AI to build landed costs, model tariff exposure, and manage suppliers. Customs classification stays with licensed brokers.
- Prompts: landed cost build-up, tariff exposure, supplier scorecard
- Skills: landed cost & tariff analysis; supplier & sourcing management; product compliance documentation
- Connect: ERP PO and cost data, freight forwarder reports, customs broker entries
- Agent: landed cost variance monitor; supplier delivery watch
- What AI does: Builds full cost per unit
- Helps: True margins
- Control: Broker-provided duty rates
- What AI does: Models exposure by SKU and country
- Helps: Plan price and sourcing moves
- Control: Trade counsel advises
- What AI does: Delivery, quality, and cost trends
- Helps: Better supplier reviews
- Control: Sourcing lead decides
- What AI does: Single-source and country exposure
- Helps: Plan diversification
- Control: Quality qualifies suppliers
- What AI does: All-in cost and transit trade-offs
- Helps: Better freight buys
- Control: Check quote validity
- What AI does: Tracks required product documents
- Helps: Ready for retailer requests
- Control: Regulatory advisor confirms
Sourcing & Cost Checklist
ControlsBefore You Start
- Get HTS codes and duty rates from your broker
- Collect freight and fee data
- Define standard cost process
- Map suppliers and countries
- Baseline: landed cost accuracy
After You Start
- Update landed cost on every receipt
- Review tariff scenarios when rates change
- Score suppliers quarterly
- Track compliance document expiry
- Update standards with controller approval
- Customs: AI never classifies products or determines duty; licensed brokers and counsel do.
- Product compliance: Requirements vary by category (CPSIA, MoCRA, supplement GMP, pet food registration); a regulatory advisor confirms.
- Supplier confidentiality: Don't share one supplier's pricing with another.
- Standards: Controller approves standard cost changes.
- Claims: Product claims need substantiation and review.
AI for E-Commerce & Marketplaces
Use AI to monitor DTC and marketplace operations, listings, reviews, and fees. E-commerce leaders approve changes.
- Prompts: DTC weekly summary, listing health review, review theme analysis
- Skills: e-commerce & marketplace operations
- Connect: Shopify connector for Claude (start read-only), marketplace reports, review exports
- Agent: marketplace listing health monitor
- What AI does: DTC performance and stockout impact
- Helps: Catch problems fast
- Control: Look at trends
- What AI does: Buy box, suppression, and rating flags
- Helps: Protect sales
- Control: Verify in Seller Central
- What AI does: Product and fulfillment issues from reviews
- Helps: Fix root causes
- Control: Combine with returns
- What AI does: Margin after marketplace fees
- Helps: Find unprofitable SKUs
- Control: Verify fee schedules
- What AI does: MAP and cross-channel price gaps
- Helps: Protect retail partners
- Control: Counsel on enforcement
- What AI does: Consistent customer responses
- Helps: Faster service
- Control: Match policies
E-Commerce Operations Checklist
WorkflowBefore You Start
- Connect Shopify read-only
- Export marketplace reports
- Collect approved product claims
- Document MAP policy
- Baseline: conversion, stockouts, fees
After You Start
- Review listing health daily
- Analyze reviews monthly
- Check margin after fees quarterly
- Approve any write access deliberately
- Update templates with policy changes
- Write access: Store connectors start read-only; product, price, and discount changes need owner approval.
- Claims: Health, safety, and performance claims need regulatory approval.
- Customer data: Follow your privacy policy; minimum necessary.
- Marketplace rules: Verify current policies before acting.
- Pricing: MAP enforcement follows counsel's guidance.
Set Up & Get Running
If you already use prompts, here is how to turn one into a repeatable skill and then a useful first agent. Start with a task that already works.
From prompt to first agent
- Choose a recurring task
- Use real data
- Good first choices: deduction categorization, 3PL invoice audit, or weeks-of-supply review
- Write down inputs, output, and reviewer
- Test on three real cycles
- Improve when results are unclear
- Connect Shopify or NetSuite read-only
- Add exports for retailer and 3PL data
- Keep write access off
- Start with a checking task
- Set when it runs; a person reviews
- Track dollars recovered and time saved
It's useful, measurable, and easy to check. It categorizes new deductions, assembles evidence, and flags dispute deadlines. It doesn't submit disputes or write anything off.
Copy these builders
Use the first after a prompt works. Use the second only after the skill has been tested on real work.
Days 1-30: Foundation
- Name an AI lead in operations
- Write data and approval rules
- Set up business accounts
- Connect Shopify or NetSuite read-only
- Pick 1 pilot (deductions or 3PL audit)
- Record baseline
Days 31-60: Expand
- Save the pilot as a skill
- Add a 2nd workflow (forecast exceptions or landed cost)
- Set up retailer and 3PL report exports
- Draft the AI use policy
- Measure dollars recovered and time saved
Days 61-90: Standardize
- Add a 3rd workflow (trade matching or inventory position)
- Adopt policy
- Pilot one review-only agent
- Write SOPs
- Report results to leadership
- Plan next wave
Implementation Success Measures
Measurement30-Day Targets
- Pilot running weekly
- Baseline recorded
- Connectors read-only
- No unapproved actions
60-Day Targets
- Recovered dollars or errors caught tracked
- Second workflow live
- Exports in place
- Draft policy reviewed
90-Day Targets
- 3 workflows with SOPs
- Policy adopted
- First agent piloted
- Measurable margin or time impact
- Decision on next wave
How the Pieces Fit Together
Start with a prompt. Save the ones that work. Add the systems and files they need. Then use an agent for recurring work, with someone checking the result.
- Ask AI to do one clear task
- Use it for a quick draft, analysis, or explanation
- Check the result before you use it
- Save the step-by-step instructions for a task your team repeats
- Add your policies, examples, and review steps
- Update it when your process or regulations change
- Bring the right systems, files, and reports into the task
- Stop copying and pasting information by hand
- Start read-only, with only the sources the task needs
- Run a skill on a schedule or when something happens
- Gather the information, prepare the work, and flag issues
- Leave POs, pricing, disputes, and partner messages with people
Where AI Can Help Consumer Brand Operations
Use AI to get routine operations work into a clear first draft. Keep spending, pricing, and partner decisions with your team.
- Forecast exception reviews
- Sell-through vs. sell-in
- Weeks of supply by channel
- Replenishment and allocation options
- Vendor guide summaries
- PO, ASN, and OTIF checks
- Deduction categorization and disputes
- 3PL invoice audits
- Landed cost and tariff scenarios
- Trade ROI and deduction matching
- Gross-to-net and channel margin
- Marketplace fee analysis
- POs, allocations, and pricing
- Disputes and write-offs
- Customs and product compliance
- Retailer and supplier relationships
AI Capabilities Explained
No jargon. What each kind of AI does, and what it looks like in brand operations.
Claude, ChatGPT, Gemini, and Copilot write, summarize, and follow instructions.
In operations: vendor guide summaries, dispute letters, S&OP pre-reads.
AI that reads your systems through connectors.
In operations: asking Claude about orders and inventory through the Shopify or NetSuite connector.
Projects demand from history, promotions, and signals.
In operations: SKU-channel forecasts, launch curves, promotion lift.
Compares records to rules and lists exceptions.
In operations: POs vs. price lists, 3PL invoices vs. rate cards, ASNs vs. vendor guides.
AI built into your existing platforms.
In operations: SPS Commerce MAX flagging EDI anomalies.
Finds themes in large amounts of text.
In operations: review themes, return reasons, deduction descriptions.
Guardrails: Data, Approvals & Product Compliance
Simple rules for using AI in brand operations. Retailers, suppliers, and customers trust you with their data; regulators expect accurate products and claims.
- Retailer POS and pricing under data-sharing terms
- Supplier pricing and contracts
- Customer personal and payment data
- Use minimum necessary data
- Purchase orders and allocations
- Disputes and write-offs
- Price changes and promotions
- Retailer and supplier communication
- Start every connector read-only
- Write access only for approved tasks
- Review scopes quarterly
- Log what agents read and produce
- Customs classification by licensed brokers
- Product claims substantiated and reviewed
- Category rules confirmed by advisors
- Pricing policies reviewed by counsel
- Figures tie to ERP and source reports
- Vendor guides and contracts quoted exactly
- Hypotheses labeled as hypotheses
- Verify before acting
- Forecast sign-off
- Supplier and retailer negotiations
- Regulatory decisions
- Financial postings
Brand Operations AI Readiness Checklist
ControlsLeadership & Policy
- Executive sponsor named
- AI use policy adopted
- Business accounts with training off
- Connector permission policy written
- Teams trained
Practice & Review
- Connectors start read-only
- Human approval on POs, disputes, and prices
- Retailer data terms followed
- Quarterly connector review
- Exceptions and errors tracked
Scope: Operations, supply chain, finance, and e-commerce teams.
Approved tools: [e.g., Claude Team with Shopify and NetSuite connectors (read-only by default); Microsoft Copilot; built-in AI in our EDI provider]
Approved uses: forecast and inventory analysis, exception reviews, deduction research, invoice audits, landed cost models, drafting internal and partner communications for review.
Data rules: Retailer POS and pricing data only as permitted by data-sharing terms. No customer payment data. Supplier pricing kept confidential.
Approvals: A named owner approves POs, allocations, disputes, write-offs, price changes, promotions, and any message to retailers, suppliers, 3PLs, or customers.
Connectors: Read-only by default. Write scopes require owner approval for a specific task and are reviewed quarterly.
Compliance: Customs classification by licensed brokers; product claims reviewed by our regulatory advisor.
Approval: [COO / CFO], [date]
AI Maturity Model for Consumer Brand Operations
See where your operations team is today. Choose a realistic next step.
Curious + Ad Hoc
- Individuals use AI on spreadsheets
- No data or approval rules
- No connectors
- No shared prompts
Shared Prompts & Rules
- Data and approval rules
- Business accounts
- Prompt library
- Read-only connector to one system
Embedded Workflows
- Skills for deductions, 3PL audits, forecasting, landed cost
- Adopted policy
- Regular exports from retailers and 3PLs
- Measured results
Connected & Measured
- Review-only agents with owners
- Governed write access for specific tasks
- Margin impact tracked
- Quarterly permission reviews
Maturity Self-Assessment
AssessmentLeadership
- Sponsor
- Policy
- Training
- Connector rules
Practice
- Prompt library
- Skills
- SOPs
- Exports
Protection
- Read-only default
- Human approvals
- Retailer data terms
- Customs and claims rules
Measurement
- Baseline
- Dollars recovered
- Errors caught
- Quarterly review