Core concepts at a glance

Every automation in DigitalGenius is built in Flow Builder. Before you start building, it helps to understand the handful of building blocks that every flow is made from. This page gives a short introduction to each one. The pages in the Key Concepts section below cover them in more detail.

Flows

A flow is a visual automation that handles a specific job, such as answering a "where is my order" enquiry or processing a return. It is made up of a sequence of steps that run in order, from the moment the flow is triggered to the moment it finishes.

Most accounts have a main flow that receives incoming tickets or chats from your helpdesk and decides what should happen next, often by handing off to more specific flows.

Activities

Activities are the individual steps inside a flow. Each activity does one thing, and there are four main types:

• Actions: prebuilt steps from the DigitalGenius library that perform a task, such as predicting a customer's intent, looking up an order in Shopify, or updating a ticket in Zendesk.
• Subflows: another flow used as a step. This lets you build a piece of logic once and reuse it across multiple flows.
• Templates: responses with dynamic placeholders that are filled in with live data, such as the customer's name or tracking number.
• Decisions: branching points that send the flow down different paths depending on which conditions are met.

Modules and connections

Modules are the integrations that let actions talk to your other systems, such as your helpdesk, ecommerce platform, returns provider, or carrier. Once a connection is set up, any action for that module can use it.

Flow parameters

Flow parameters are values set at the flow level and available to every activity in that flow. These include the inputs the flow receives when it starts and configuration values like IDs, settings, or thresholds that you want to manage in one place.

How it fits together

When a ticket or chat arrives, your helpdesk sends it to DigitalGenius and a flow starts running. Data enters the flow through its inputs. Each activity takes inputs, does its job, and produces outputs that later activities can use. For example:

1. An AI action predicts the customer's intent from their message.
2. A decision routes the conversation based on that intent.
3. A Shopify action looks up the customer's order.
4. A template builds a reply using the order details.
5. A helpdesk action sends the reply and updates the ticket.

Each run of a flow is recorded, so you can review what happened at every step.