Supercharged data import

We rebuilt WasteNot's import pipeline so data sources are no longer treated as one large, opaque sync. Each source is broken into visible streams, imported with more parallelism, and connected to clearer event reporting inside the app.

Parallel data import

Each data source can now be split into streams: individual feeds like orders, refunds, subscriptions, messages, or email events. Most streams import at the same time, so one slow feed does not hold up the rest of the source.

That means fresher customer activity reaches audience rules faster, especially for sources with several event types running at once.

Stream remapping

Events inside a source stream can be mapped into the WasteNot event that best matches how your team uses the data. For example, if you want a refund-like feed to count toward an existing audience rule, you can adjust that mapping without changing the rest of the source.

That gives teams more control over how imported activity appears in WasteNot. New stream types can be added, mapped, enabled, or paused without turning the whole source into a single all-or-nothing import.

Data-source UI

The data-source UI now reflects this stream model. Streams are visible on the source page, with event names, external source names, enabled status, sync status, and per-stream actions.

We also added event counts and a chart for recent activity, so teams can see whether a source is actually producing usable events. When attention is needed, the same page surfaces the source status and reconnect action.

Sync windows

Streams can now be capped at a specific sync end date when a source needs a controlled import window. That gives teams a safer way to backfill or test a source without continuing to pull newer activity past the intended cutoff.