Building Your First Workflow: From Blank Canvas to Saved Report

The blank canvas problem

The first time you open DataQloo's Visual Designer, there's a toolbox on one side, an empty canvas in the middle, and no obvious first move. Every visual tool has this moment — the gap between "I understand what this is for" and "I've actually built something with it."

The fastest way past that gap isn't reading about every node available. It's building the smallest workflow that still does something real, so the next one is a small step from something familiar instead of another blank canvas.

The smallest useful workflow

Three nodes: connect to data, ask one narrow question of it, look at the answer. Here, that question is "show me only the West region's orders" — deliberately small, so the point is the shape of the workflow, not the complexity of what it does.

Step 1 — place an Input node

Drag an Input node onto the canvas and select a dataset — a saved PostgreSQL connection, or a file already imported through the Import Wizard. This node is always the starting point: everything else on the canvas reads from whatever it's connected to.

Step 2 — add a Filter node

Connect a Filter node to it, and set one condition: region = West. The node's label updates on the canvas immediately — it should read something like "Filter — region = West" — so the workflow states what it does without needing to be opened again to remember.

Step 3 — add a Browse node and look at the result

Connect a Browse node to see what came out the other side: the filtered rows, the inferred schema, a column profile, and — since this reads from PostgreSQL — the actual SQL that ran.

SELECT *
FROM sales_orders
WHERE region = $1
LIMIT 100

Step 4 — save it

Name the workflow and save it. That's the entire difference between "something on a canvas" and something that exists after you close the tab — the next time this question comes up, the workflow is already there to rerun, not rebuilt from an empty canvas again.

Why start this small

A three-node workflow doesn't look impressive, and that's the point of doing it first. It proves the whole loop — connect, configure, look at the result, save — without also having to learn Summarize, Sort, and how they interact. Once that loop feels familiar, adding a node to total or rank the result is a small addition to something that already works, not a bigger thing built from nothing.

Key takeaways

• The smallest real workflow is three nodes: a source, one condition, and a place to see the result.
• The canvas states what a workflow does in its own labels — nothing has to be remembered or documented separately.
• Saving a workflow, even a small one, means it exists to rerun later — the same idea behind not rebuilding a report from scratch every time, starting from the very first workflow you build.

Next

Once connecting and filtering feels familiar, the next step is usually bringing in your own data — see how to import a CSV file into a reusable dataset. For a few more guided starting points, the tutorials section has step-by-step walkthroughs beyond this one.