industry-department-solutions
latest
false
Supply Chain & Retail Solutions API guide
- Overview
- API Resources
Data Onboarding Checklist
Checklist for tracking progress through data preparation and submission stages when onboarding to the Supply Chain & Retail Solutions Data Ingestion API.
Use this checklist to track your progress as you prepare and submit data to the Supply Chain & Retail Solutions ingestion API. Complete each item before moving on to the next stage.
Before you begin
- Read the Getting Started guide
- Confirm you have access to platform.peak.ai
- Generate a Personal Access Token (PAT) from the Access Tokens page in the platform
Gather your solution details
Submit a support ticket if you do not have these details already.
- Obtain your solution name (e.g.,
QP_OOTB) - Obtain your prefix (e.g.,
QP_) - Obtain your suffix (e.g.,
_OOTB) - Identify the target schema name (e.g.,
STAGE) - Identify the app name and app version of the standard schema you will roll out
- Confirm your API base URL (if your tenant is in the Ireland region, use the default base URL
https://ingestion.peak.ai; if your tenant is in any other region, use your region-specific base URL — for example,https://ingestion.peaksydney.peak.aifor Sydney — and raise a support ticket if you have not been given one. See API host and reference for details.)
Roll out the solution tables
This step creates the warehouse tables (and matching <table_name>_failed_rows tables) you will ingest into. See Schema lifecycle for the full reference.
- Confirm a standard schema has been saved for your
appName+appVersion - Send a
POST /api/v2/schema/rolloutrequest for yoursolutionName— see the linked reference for the full payload, example request, and response - Verify the response lists every expected table under
successfulObjectsand thatfailedObjectsis empty - (Optional) Use
GET /api/v2/schema/solutionsto confirm the rollout is registered for your tenant
Prepare your data
- Review the data model for your solution module:
- Retail
- Merchandising
- Review the supported Data Types and ensure your data matches the expected formats
- Identify the primary key columns for each object you plan to ingest
- Confirm all required fields are present in your dataset
- Split your data into batches of 2000 rows or fewer per request (1 MB payload-size cap)
Submit your data
- Choose your operation type:
UPSERT(insert or update) orAPPEND(insert only) - Construct your API request using the ingestion URL pattern:
POST {base-URL}/api/v2/objects/{prefix}{OBJECT_NAME}{suffix}— use your tenant's base URL (see API host and reference) - Stay within the ingestion limits: up to 2000 rows per request and 50 requests per second per tenant
- Send a test request with a small data sample (consider
dryRun: truefor a no-write smoke test) and verify a successful response - Review the API Guide for error codes if any validation errors are returned. A
207 Multi-Statusresponse means part of the batch was accepted and part rejected. - Ingest your full dataset
A 200 response confirms your rows were accepted; they are typically queryable within seconds.
Monitor your ingestion
- Confirm aggregate outcomes (records ingested, pass rate, top error codes) in the Data Quality Dashboard in your Peak tenant
- Review any failed rows in the dashboard's failed-rows detail page
- To investigate a specific batch, note the
requestIdfrom each ingestion response and use it to filter the Data Quality Dashboard
Deploy the application
- Confirm the selected application and version are deployed to the Library in your Peak tenant, making it available for use
Optional
- Add custom attributes to extend the schema for your business needs
- Review the default scheduled ingestion schedule (every 30 minutes) and request a custom cron for any tables that need a different one
- When your application moves to a new version, upgrade your rolled-out schema (additive changes only — new tables and columns). Preview it first with
dryRunset totrue. See Schema lifecycle