Skip to content

Automatic translation — the Portuguese version is the reference. View in Portuguese

Updated on 2026-09-27

Home screens — Overview and Architecture

The entry dashboard, with the numbers it shows and what each one measures, and the Architecture screen, which explains where each product fits in the data path.

On this page (4)

Interface inventory: the text here is the text on screen.

Read the entry dashboard (/overview)#

Where it is: menu Home → Overview. Subtitle: "A summary of your workspace, connected sources and forecasted cost."

The four numbers at the top:

NumberWhat it measures
Connected sourceshow many exist, and how many have already completed a load
Dashboardshow many exist, how many have data, and how many are published
Infrastructure cost per montha forecast based on volume and the dashboards' mode
Current planthe plan and the subscription status

The cost here is a forecast

The supporting line itself says: "Estimate based on volume and dashboard mode — the actual amount is in Usage and invoices." For what has already been charged, go to Credits.

The Activity over the last 14 days section shows real runs, not sample data — the screen says so. It includes Runs, Success rate, Rows loaded, Data processed, Estimated cost, Active pipelines, Most recent data, and Average duration, plus the Runs per day, Rows loaded per day, and Volume per pipeline charts.

Some cards only appear when there's a reason:

  • Continue where you read — picks up the dashboard you were viewing
  • You have a pipeline setup halfway done — with the Resume setup button, and the notice that no data has been loaded and nothing you typed was lost
  • Your favorite dashboards and Recent alerts — which disappear entirely when they're empty, instead of showing an empty list

Understand the data path — Architecture (/pipeline)#

Where it is: menu Home → Architecture. Title Architecture, subtitle "From data at the source to the decision: where it comes in, what happens at each stage and which product handles it. Click a block to see what already exists in it."

This screen is not the pipeline list

The /pipeline address, singular, is the architecture diagram. The pipeline list is in ETL/ELT → Pipelines — see Pipelines screens. Mixing the two up is common, and the diagram is here precisely to make clear where each thing lives.

The diagram has five zones:

  1. Orchestration and scheduling — "Decides WHEN each load runs, reruns what failed and logs everything." It drives block 2 and the transformations in block 4.
  2. Data sources — the systems where the data lives today: Databases and Files and platforms.
  3. ETL/ELT — "Brings data from outside into your datalake."
  4. Datalake — storage and processing, in three layers:
    • Bronze — Raw data: "As they came from the source. Nothing is corrected here — it's the faithful record."
    • Silver — Processed data: "Right types, duplicates removed, dates standardized, sensitive columns flagged."
    • Gold — Analytics layer: "Aggregated metrics, the way the dashboard and the AI read them."
  5. Consumption and intelligence — where processed data becomes decisions.

Clicking any block opens a panel with what already exists in your environment, and the Open the screen button takes you straight there. When a block is empty, it says what's missing — for example: "No raw table yet — run a pipeline to create Bronze." or "No access group created — without groups, everyone sees everything with personal data masked."

At the bottom, the three guarantees the product commits to: Per-customer isolation, Cost upfront, and Audit.

Back to the app#

Next task#

Related links