Skip to main content

Integration Overview

How CoolR Delivers Data at Scale

Two integration paths. One is right for most enterprise deployments.


CoolR's primary integration model is designed for operational reliability at scale. Data moves between CoolR and your environment via Azure Blob Storage, encoded in Apache Parquet format — a columnar structure optimized for the analytical workloads CPG teams actually run.

What this means in practice:

  • Consistent delivery. Data arrives on a defined schedule. No polling, no dropped callbacks, no dependency on network conditions at the moment of query.
  • Schema stability. Parquet files carry their own schema. Your downstream systems — whether Power BI, Databricks, Snowflake, or a custom pipeline — can consume without fragile parsing logic.
  • Volume-ready. Whether you are processing 5,000 cooler events or 5 million, the same pipeline handles it. The architecture doesn't change with scale.
  • Auditable. Each delivery is a discrete, inspectable file. If something looks wrong, you can open the Parquet and check it — no specialized tooling required.

This approach fits organizations running analytics-grade data infrastructure. If your team works with Azure Data Factory, Databricks, or similar, this is the native pattern.

Delta window

The delta (incremental) export covers the last 30 days on a rolling basis. Even if your pipeline misses a scheduled run, you won't lose data — there is no need for manual tracking or gap recovery.


2. REST API (Available for Point Integrations)

CoolR exposes a REST API for use cases that genuinely require it: real-time triggers, lightweight reads, or integration with systems that cannot consume batch files. It is documented, versioned, and supported.

That said, the API is not the right starting point for most enterprise deployments. Batch pipelines outperform API polling on throughput, cost, and reliability at the data volumes CoolR customers typically operate.

If you're evaluating which path fits your environment, start with the bulk sync documentation. If you have a specific real-time requirement, reach out to your CoolR implementation contact before building against the API.


Key Integrations

IntegrationDescriptionRecommended approach
Ingesting data from CoolRReceive cooler analytics, KPIs, and master data in your environmentBulk (Parquet via Azure Blob)
Push data to CoolRSend master data (outlets, products, assets, users, planograms) to CoolRBulk file import or single-record API
Sending images to CoolRUpload cooler images from third-party apps, mobile devices, or IoT systemsREST API (image upload endpoint)
Accessing images from CoolRRetrieve processed cooler images for embedding in client systemsREST API (image preview endpoint)