Skip to content

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

Updated on 2026-10-05

What's new

What changed in the product, in the words of the people who use it — new screens, renamed items, and new behaviors, from newest to oldest.

On this page (3)

What changed in the product, from newest to oldest. Each item says where it lives and points to the page with the step-by-step.

October 2026#

October 5

Data and pipelines (ETL/ELT)

  • Micro-batch: pipeline every 5 or 15 minutes (Business and Scale plans). On the Pipeline schedule card, the Cadence goes from Batch to Micro-batch (minutes); each round brings only what changed. If the pipeline stops running at the promised pace, it shows as Behind. See Micro-batch.
  • Spark notebooks and clusters. Write Python, PySpark and SQL cells on a cluster you turn on and off (ETL/ELT → Clusters and ETL/ELT → Notebooks). See Notebooks and clusters.
  • Notebook as a Silver or Gold step of the pipeline: the Notebook button next to New Silver / New Gold. See Notebook step.
  • CI/CD: Development and Production. Build in Development, use Send to production, someone else approves and, if something goes wrong, Roll back. With your own Git repository, if you want (ETL/ELT → CI/CD; add-on). See CI/CD.
  • Meta Ads and Google Ads with your business's own access. The connectors use the token of your own ad account. See Meta Ads and Google Ads.

Automation

  • Flows: box-by-box automations — when something happens, run the pipeline, refresh the dashboard, decide, ask for approval and notify by email or WhatsApp (AI → Flows). See Flows.

Dashboards (BI)

  • Import model for all workspaces. Dashboards read each table's extract; older visuals migrate in an assisted way. See Migrate to the Import model.

Account and privacy

  • Account deletion with 30 days to change your mind. The deletion request opens a grace period: nothing is deleted, the environment keeps working and Cancel the deletion undoes everything. See Governance and LGPD.
  • Ingestia Academy at academia.ingestia.io: free courses and tracks; paid certification, with a verifiable credential and a free voucher on request. See Ingestia Academy.

Earlier in October

Dashboards (BI)

  • The dashboard reads each table's extract (Import mode). Swapping or adding a field in a visual shows up instantly, with no new query. Refresh now rebuilds the extracts — and it's what brings a new table column into the dashboard. See Data modes.
  • Migrate old visuals to the extract, one by one, checking before and after: Data → Import model → Migrate visuals. See Migrate to the Import model.
  • Layout per device and TV: the same dashboard with its own arrangement for tablet, phone, and TV (full screen, refreshes on its own). See Layout per device and TV and TV link.

Data and pipelines (ETL/ELT)

  • "DAGs" is now called Runs (ETL/ELT → Runs): the runs of each pipeline, with the progress and the error of each step.
  • New connection in 4 steps, with a test before saving. See Connections.
  • Guided mode on the first ingestion, for people who have never built a pipeline. See First ingestion.
  • Schedule in Brasília time on the pipeline screen. See Schedule and time zone.
  • Run limits (retries, wait, timeout, and concurrent runs) now really take effect. See Pipelines.
  • How to add a column to a Silver or Gold table (e.g., bringing clicks into a per-state table). See the tutorial.

Help (Aiô)

  • Aiô reads your environment to answer: pipelines, runs, failures, the columns of any table, how a Silver/Gold table is built, and the fields of each visual in a dashboard. When it can do something for you (run, schedule, recover days), it proposes it and you confirm. See Help assistant.

Related links