Enterprise use case · Shopify + Amazon Seller + OpenAI + NetSuite
Product matching across channels
The same product appears under different titles, codes, units or bundle descriptions across commerce channels and the ERP. Catalog owners receive proposed links between channel listings and ERP items, with ambiguous matches kept out of operational use.
Systems involved
What each system contributes
39 actions
- Products
10 actions
- FBA5
- FBM5
25 actions
- Chat Completions8
38 actions
- Items10
- Base8
Amazon Seller pre-built coverage is FBA and FBM. Additional listing or catalog context requires the Amazon Selling Partner API or a Rules + Code step. The model step runs against the AI provider you choose, including Anthropic, Google Gemini or a model you host yourself. Anything outside a pre-built action runs through an API, protocol or database call, or as Rules + Code in Python, Bash, Go, Java, .NET, PHP, PowerShell or JavaScript/TypeScript, with the same approvals and audit.
Architecture
One governed run
How it starts: A new or changed Shopify or Amazon Seller listing is released for catalog matching.
Step by step
How the process runs
- 01
Gather product context
Read the relevant Shopify listing, available Amazon Seller listing or fulfillment identifiers and NetSuite item candidates through permitted interfaces.
- 02
Apply exact constraints
Rules compare reliable identifiers and reject incompatible units, pack sizes or variants before semantic matching.
- 03
Propose likely matches
The selected model compares descriptions and available attributes, ranks candidates and explains the distinctions that support or weaken a match.
- 04
Review uncertain pairs
The catalog owner checks ambiguous variants, bundles and near-matches against the source records.
- 05
Publish approved mappings
A separate step records the approved cross-reference in the designated NetSuite item fields or authorized mapping dataset for downstream workflows.
Control
Human decision and exception path
Human decision
The named catalog owner approves proposed mappings before they are used to route inventory, orders or fulfillment. Price changes, stock adjustments and posted-document changes are separate actions with their own approvals.
Exception path
- Conflicting pack sizes or variant attributes block a match even when descriptions look similar.
- Missing authoritative identifiers leaves the proposal unresolved rather than creating a forced link.
- A changed listing or item invalidates the affected mapping for review before downstream reuse.
Evidence
What gets audited
- Shopify, Amazon Seller and NetSuite references with compared attributes.
- Exact-match constraints and model-ranked candidate explanations.
- Catalog-owner decision and approved mapping version.
- Mapping publication request and destination response.
Nexus runs on premises, in private or sovereign cloud, or air-gapped. Credentials stay in the vault, roles decide who can approve, and the execution record stays inside your deployment boundary.
Sovereignty and deploymentRelated
Patterns, connectors and next steps
- Shopify-NetSuite integration
Integration pattern · Storefront orders and inventory stay aligned with ERP invoicing and items.
- Shopify connector
39 pre-built actions
- Amazon Seller connector
10 pre-built actions
- OpenAI connector
25 pre-built actions
- NetSuite connector
38 pre-built actions
- Enterprise automation
Governed processes across the systems you already run.
- Order to cash across CRM, ERP and billing
Another enterprise use case.
FAQ
Questions about this use case
- Does Nexus replace any of these systems?
- No. Each system keeps its own records. Nexus runs the process between them and records each decision.
- How does the run start?
- A new or changed Shopify or Amazon Seller listing is released for catalog matching.
- What if a step is not covered by a pre-built action?
- Amazon Seller pre-built coverage is FBA and FBM. Additional listing or catalog context requires the Amazon Selling Partner API or a Rules + Code step. The model step runs against the AI provider you choose, including Anthropic, Google Gemini or a model you host yourself. Anything outside a pre-built action runs through an API, protocol or database call, or as Rules + Code in Python, Bash, Go, Java, .NET, PHP, PowerShell or JavaScript/TypeScript, with the same approvals and audit.
- Does a similar product name establish a match?
- No; unit, pack size, variant and authoritative identifiers can disqualify a candidate despite a similar description.
Use case review
Discuss this use case
Bring your version of the process. We will map the systems, approvals and evidence it needs.