Quickstart

This walkthrough takes you from a new workspace to a working data transformation in roughly 10 minutes.

Start your free trialYou'll need a PlaidCloud workspace to follow along. The free trial gets you one in a few minutes.

What You’ll Build

A small workflow that:

1. Imports a CSV into a project table
2. Filters and transforms the data with a couple of steps
3. Publishes the result so other tools can consume it

Note

This is intentionally lightweight — it’s a tour of the moving parts, not a deep dive. Once you’ve finished, the Concepts page explains why things are organized the way they are, and the Guides section covers each task in detail.

Start With the Sample Project

If you’d rather see a finished result before you build one, take the shortcut. On a brand-new workspace, a Load sample button appears. If your workspace offers more than one guided sample, you’ll first choose which one to load from a short list:

• Retail Sales Analysis — retail and sales
• Campaign Performance by Channel — marketing
• Inventory Status Report — supply chain and operations
• Budget vs Actual by Department — finance

Each is a small, realistic dataset built entirely from PlaidCloud’s low-code transform steps — no hand-written SQL — so you can open the finished workflow afterward and see exactly how it works. Pick one (or, if only one sample is available, PlaidCloud loads it directly), and PlaidCloud seeds the project and runs its workflow, showing you each stage as it goes — Importing a file, Cleaning it, branching into a couple of parallel summaries, then combining them into one final result table — before opening the result for you. Once the run finishes, PlaidCloud also publishes the result table to reporting and builds a dashboard from it automatically: a KPI summary, a trend over time, a breakdown by the sample’s main category, and — for the retail and inventory samples — a breakdown by US state. The first run takes about a minute while it warms your warehouse; every run after that is near-instant, and the screen tells you so, so the wait isn’t a surprise. You reach a real first result and a finished dashboard in a few minutes without building anything, and every piece it created is yours to open, change, and rerun. If loading or running the sample hits a problem, PlaidCloud tells you and offers to try again.

The sample project, its workflow, and each dashboard it builds are shared with the whole workspace the moment they’re created — a teammate with viewer access sees the same finished sample you do, with no separate sharing step required.

Every empty screen also guides you toward the next step rather than leaving you at a blank grid. Projects, Workflows, Tables, Connections, and Dashboards each greet you with a short explanation and a primary action — New Project, New Workflow, Import Table Bundle, and so on — so there’s always an obvious way forward. A Need help? link on these screens and on the welcome panel takes you straight to support if you get stuck. If you’ve been added to a workspace as a viewer, these screens explain what you’re looking at without offering actions you can’t take. And if a viewer lands in a workspace that has no visible content yet, a welcome greets them with a short how-to introduction, a guided tour of the app, and quick access to help — so there’s always a next step.

This guided first run appears in every new workspace, on every plan. If you already know your way around, choose Dismiss on the welcome panel — a quick confirmation reminds you the guidance is still available — and it steps out of the way. To bring it back at any time, open the user menu in the upper-right corner and choose Restart Onboarding.

Start With Your Own CSV

To start from your own data instead of a sample, choose Bring your own data. It sits beside Load sample on an empty Projects screen, and Home offers it too. You need permission to create projects and tables and to write to document storage to see it. Once you have projects, New > Project from CSV File on the Projects screen opens the same window, and to add a CSV to a project you already have, see Import a CSV File.

Drop a CSV file onto the box, or choose Choose a CSV file to pick one; only files ending in .csv are accepted. PlaidCloud uploads the file to your Personal Temp Drive, or to the workspace’s Temp Drive if you don’t have one, creates a project named after it, imports it into a table named after it — Sales Data.csv becomes the project Sales Data with the table sales_data — and opens the new project’s Tables list. A Suggested next step panel then opens, and once PlaidCloud has taken a first look at the table it becomes Your data at a glance, showing how many rows the table has and, for its first six columns, each column’s type, the share of its values that are empty, and how many distinct values it holds.

Below that is a starter bar chart. Its bars are the values of a column with a short list of repeated values, such as a region, and each bar totals or averages a number column for that value, or counts its rows. When there is more than one sensible chart to draw, PlaidCloud’s AI picks one and titles it; it sees only the column names, types and counts, never a value from your file. If no column has a short list of repeated values, the panel says there is no starter chart.

Where reporting is turned on and you have permission to publish tables, PlaidCloud also publishes the table for reporting and builds a starter dashboard from the chart, and Open dashboard opens it in a new tab. The dashboard’s charts are named after the columns they show, such as “Total revenue by region”; the AI’s title appears only on the panel’s chart. Like any new dashboard, it starts out visible to everyone in the project; see Control Who Can See a Dashboard to change that.

The panel also offers Clean this data or Summarize this data, each opening a ready-to-review step, or Not now to skip.

If the import fails, PlaidCloud removes the project it just created and tells you what went wrong. PlaidCloud deletes the uploaded file from the temp drive once the import has finished, whether or not it succeeded; failing that, the temp drive deletes it after a day. If the workspace has no temp drive, the window says there’s nowhere to store your file, and a workspace admin needs to contact support.

Track Your Progress on Home

Once you’re inside a workspace, Home shows one Getting started card that tracks your setup step by step, with a progress meter (“N of M”) built in. What you see depends on your role: admins get five steps — try the sample, connect your data, build your first workflow, publish a dashboard, and invite your team; builders get three — try the sample, build your first workflow, and publish a dashboard; viewers don’t see the card at all.

The card always leads with whatever you haven’t done yet as its main action, and every other remaining step is its own clickable row that takes you straight there. A step you finish is ticked for good — it never un-ticks — and finished steps collapse out of the way so the card stays focused on what’s left.

Note

“Publish a dashboard” only counts a dashboard you published yourself — a teammate publishing one doesn’t check your box.

If you’d rather not see the card, choose Hide and it collapses into a small “Getting started · N/M” button in the toolbar; click that button any time to bring the full card back. Restart Onboarding from the user menu brings it back too. To turn onboarding off entirely instead, use Dismiss onboarding; it stays off until you choose Restart Onboarding.

Finish every step and the card retires itself: it disappears from Home, and a one-time “You’re all set” celebration greets you — once per workspace, however many devices or browsers you use to get there.

Your progress, the card’s hidden or shown state, tours, tips, and anything you’ve dismissed on Home all follow your account rather than your browser. Sign in from a new device, a different browser, or an incognito window, and you pick up right where you left off instead of starting over.

The rest of this quickstart builds the same kind of workflow by hand, so you can see each moving part.

1. Open a Project

After signing in, you’ll land in your workspace. Open the Projects tab and create a new project — give it a descriptive name like “Quickstart”. A project is where your data, workflows, and dimensions live together.

See Managing projects for more on the project lifecycle.

2. Import a CSV

Inside the project, open the Workflows tab and create a new workflow. Add an Import → CSV step. Either upload a small CSV from your local machine or point at a CSV in a connected document store.

Run that step. The CSV lands as a table in your project. Open the Tables tab to see it.

Reference: Import CSV step · Guide: Where are workflows?

3. Add a Transform Step

Back in the workflow, add a Tables → Table Lookup step (or any of the table steps). Configure source and target, choose which columns to keep, and apply a simple filter.

When you run the step, the output is a new table you can preview.

Reference: Workflow steps

4. Publish the Result

Add a final Publish step (under Data → Publish) so the table becomes available to dashboards, BI tools, or external consumers.

Guide: Publishing data

5. Run the Whole Workflow

Click Run on the workflow. Watch the log as each step executes. If anything errors, the Managing step errors guide covers debugging.

Where to Go Next

• Understand the model — Concepts explains what a workspace, project, workflow, table, and dimension actually are.
• Build something real — Tutorials walk through end-to-end scenarios (loading and transforming data, building an allocation, publishing a dashboard).
• Browse by task — Guides covers specific things you might want to do.
• Look something up — Reference has every workflow step, expression, and connector.