Data ingestion

Data ingestion is the process of making your external activity data available to Zenskar for billing and revenue operations. Zenskar functions as an active aggregator rather than a static billing store. Zenskar works directly with your existing data infrastructure, either by connecting to it in place or by replicating it into Zenskar's own servers.

The ingestion philosophy: no custom transformation required

Most billing integrations require a transformation step: your raw activity data must be mapped, reshaped, or reformatted to conform to the vendor's proprietary schema before it can be loaded. This mapping work is custom to each integration and a common source of maintenance overhead and data quality issues.

Zenskar eliminates this requirement. Rather than defining a proprietary schema that your data must conform to, Zenskar works with your data in its existing structure — either by querying it directly in your source system or by replicating it into an equivalent instance of the same database system you already use. Either way, there is no vendor schema to learn and no mapping layer to maintain.

Data source connector modes

When connecting a database or data warehouse as a source, Zenskar provides two distinct modes. The right choice depends on the size of your data and how frequently it changes.

Directly from Source

Zenskar queries your source database directly without replicating any data to Zenskar's servers. Your tables are available in the Data Navigator during aggregate creation, and queries are executed against your source at billing time.

• Ideal for databases larger than 30 GB.
• Ideal for real-time data where freshness is critical.
• No sync waiting time.

flowchart LR
    subgraph Customer ["Your infrastructure"]
        DB[(Source database)]
    end

    subgraph Zenskar ["Zenskar"]
        Billing[Billing engine]
    end

    DB -->|Query at billing time| Billing

Sync to Zenskar

Zenskar automatically replicates your source data to its own servers on a periodic schedule. The replicated data is then used during aggregate creation for invoice generation.

• Ideal for databases smaller than 30 GB.
• Ideal for periodic data ingestion where real-time freshness is not required.

flowchart LR
    subgraph Customer ["Your infrastructure"]
        DB[(Source database)]
    end

    subgraph Zenskar ["Zenskar"]
        Replica[(Zenskar replica)]
        Billing[Billing engine]
    end

    DB -->|Periodic sync| Replica
    Replica --> Billing

Ingestion methods

Beyond data source connectors, Zenskar provides three additional methods for getting data into the platform.

Aggregator versus source

Traditional billing platforms are designed as sources: your application pushes transaction records to them, and you extract reports or data from their proprietary schema. This model works well for payment processing but breaks down for usage-based billing, where the raw activity data lives in your own infrastructure.

Zenskar is designed as an aggregator: it connects to where your data already lives — querying it in place or replicating it on your terms — preserving your schema and eliminating the need to conform to a vendor data model.

Next steps:

• To understand the specific data structure required for ingestion, see the [Usage event] concept page.
• Ready to start? See the [Get started with usage ingestion] tutorial.

How-to guides