Getting a recurring task automated usually means a whole IT project with an outside developer, at the end of which nobody in the business still understands exactly what is running in the background.
Here, someone in the business simply describes the recurring task in chat, right inside their own application. That turns into a visible flow of individual steps, such as querying data, filtering it, creating a document, exporting it, sending it by mail or updating a record. Each step is accepted or rejected on its own before it is saved, then the flow runs on demand or on a schedule, with a full history and downloadable results.
Already in use in two completely different applications for two different customers, a field-service planner and a club-membership system, running on the same engine, each one tailored to its own data.
What ports as-is
- The underlying structure that keeps every task, step, connection, run and chat log straight
- The individual step types and how they run in the right order
- Execution with a status per step, including for a step that turns out not to be possible
- The schedule for automatic runs in the background
- The chat that turns a description into a flow built for your own data
- The planning view and the history of every run
What we build for you
- The description of your own data that the planner reasons over
- Which step types are available at all
- The layout for documents and exports
- Schedules and triggers
- Where someone has to approve before anything gets written or sent
