For data teams

Stop being the bottleneck for every number.

You are the first data hire, or the whole data team. Every metric request lands on you, and three tools disagree on the answer. Visivo lets you define each number once, in code you review, so the rest of the company can answer its own questions from it.

$30 of credit to start, no card. Viewing is free for everyone you invite.

Three requests that stop landing on you

✕“Whose number is right?”

Define each number once.

Metrics, dimensions and joins live in one semantic layer, in YAML on top of your SQL or dbt™ models. Every chart reads the same definition, so the dashboard and the ad-hoc question agree.

  • Metrics, dimensions and joins in one layer
  • Built on your SQL models or your dbt™ project
  • Lineage from the source to every chart
✕“Can you pull this for me?”

Let people filter and split charts themselves.

Each chart runs its query once. People then filter, split and sort it in the browser without a new query or a new ticket. Invite the whole company: reading dashboards and exploring cost nothing.

  • Filters, splits and sorts run in the browser
  • Unlimited viewers on every plan
  • One trusted definition behind each answer
✕“The dashboard broke again.”

Ship nothing you have not reviewed.

The project is code in Git. Every pull request gets its own build and a GitHub check. Lineage shows what a change touches, and visivo test checks the data before you merge.

  • Pull request builds with a GitHub check
  • Edits in Visivo Cloud wait as drafts until you commit
  • Data tests with visivo test
How the work flows

One definition, from the warehouse to every chart.

  1. 1ConnectPoint Visivo at PostgreSQL, Snowflake, BigQuery, DuckDB or one of the other supported sources, or at your dbt™ project.
  2. 2DefineWrite each metric and dimension once in the semantic layer. Declare how two models join once, and Visivo writes the join for you.
  3. 3BuildWrite YAML, drag charts onto a dashboard in the browser, or ask the built-in AI agent. Browser and agent edits land as drafts.
  4. 4Review and shipCommit to a branch or open a pull request. The check runs, you merge, and everyone sees the same number.
project.visivo.yml
models:
  - name: orders
    sql: select * from orders
    metrics:
      - name: net_revenue
        expression: sum(amount - refunds)
    dimensions:
      - name: order_month
        expression: date_trunc('month', created)

insights:
  - name: revenue-by-month
    props:
      type: bar
      x: ?{ ${ref(orders).order_month} }
      y: ?{ ${ref(orders).net_revenue} }
Built for a team of one

Start on your laptop.
Share when you are ready.

Open source CLI

The CLI is free and open source (GPL-3.0). Run visivo serve and see every change on a local dashboard as you save.

Agents edit, you review

Point Claude Code or Cursor at your local project over MCP, or ask the built-in agent. Either one writes drafts. You review, run and commit.

Secrets stay out of YAML

Local runs read your .env. Visivo Cloud keeps passwords in a write-only secret store. Your project files hold a reference, never the password.

Pay for usage, never seats

No per-seat price on any plan. Free starts with $30 of credit for the AI agent and cloud builds, and needs no card.

Install command copied