Skip to content

Insights

The Insights tab lets you create reusable, narrative reports tied to a Data Mart. Each Insight is a Markdown document with embedded data placeholders. When you run an Insight, OWOX executes the underlying SQL queries against your data warehouse and renders the results inline — producing a ready-to-read analysis document you can share or deliver on a schedule.

Use Insights to turn raw Data Mart output into recurring, narrative-style reports for stakeholders — without writing new queries every time.


Before creating an Insight, make sure the following conditions are met:

  • The Data Mart is Published. Insights can only run against a published Data Mart. Draft Data Marts will block execution.
  • The Data Mart has an Output Schema. The schema is required for the AI Assistant to understand the available fields and generate accurate queries. The schema is populated automatically once the Data Mart is saved and its source is valid.
  • Your role is Editor or higher. Viewers can read and preview Insights, but creating, editing, running, and deleting Insights requires the Editor role.

An Insight is composed of two parts: a template (the Markdown document) and one or more Data Artifacts (named SQL queries that supply data to the template).

Insights editor with three panels: AI Assistant on the left, Markdown template editor in the center, and rendered output preview on the right

The template is a Markdown document. You control how data is embedded using the tags.

  • {{table}} — renders the full output of the Data Mart’s Input Source as a Markdown table
  • {{table source="data_artifact_id"}} — renders the result of a specific Data Artifact identified by data_artifact_id
  • {{value}} — inserts a single scalar value from the default source (row 1, column 1)
  • {{value source="data_artifact_id" path=".columnName[row]"}} — inserts a single scalar value from a specific Data Artifact and cell

When the Insight is run, each {{table}} tag is replaced with a live Markdown table containing up to 100 rows from the corresponding query result.

Use {{value}} when you need to pull one number or label out of a query result — ideal for KPIs and summary sentences.

There are three ways to address the cell.

No parameters — returns the value at row 1, column 1 of the default source:

{{value}}

path=".columnName" — returns row 1 of the named column; add [n] to target a specific row (1-based):

{{value source="kpis" path=".total_spend"}}
{{value source="kpis" path=".cost_per_click[2]"}}

column + row — same result as path, using separate parameters; column accepts a name or a 1-based index:

{{value source="kpis" column="cost_per_click" row="2"}}

path and column/row are mutually exclusive.

Example — a summary sentence using values from a kpis Data Artifact:

## This Week at a Glance
Total spend: **{{value source="kpis" path=".total_spend"}}**
Top channel: **{{value source="kpis" column="top_channel"}}**
Cost per click: **{{value source="kpis" path=".cost_per_click[1]"}}**

Insight template editor showing {{value}} tags rendered as inline scalar metrics inside a narrative Markdown document

A Data Artifact is a named SQL query attached to an Insight. Artifacts are the data sources that feed {{table source="data_artifact_id"}} tags in the template.

  • Each artifact has a unique data_artifact_id (alphanumeric and underscores only, e.g., campaigns, top_channels).
  • A single Insight can have up to 5 Data Artifacts.
  • SQL is validated as you type and you can preview results before saving.
  • Artifact SQL is executed against the Data Mart’s connected storage (e.g., BigQuery, Athena) when the Insight runs.

Data Artifacts panel listing attached SQL queries with their keys, titles, and validation status indicators


The AI Assistant panel (within the Insight editor) is a chat interface for iteratively building and refining an Insight. You can ask it to:

  • Write or revise the Markdown template
  • Create or edit a Data Artifact’s SQL
  • Explain what an existing query does
  • Attach a new data source to the template

The assistant proposes changes as suggested actions. Click Apply to write the suggestion to the editor. No changes are saved to the template until you explicitly apply them.

AI Assistant chat panel with a user message, AI response, and an Apply button to write the suggested changes to the Insight template

  1. Open a Data Mart and click the Insights tab.
  2. Click + New Insight.
  3. Give the Insight a title.
  4. Write or paste your Markdown template in the editor. Use {{table}} to embed data.
  5. Add Data Artifacts if needed (see Managing Data Artifacts below).
  6. Click Run to generate the first output.

Insights list view for a Data Mart showing existing Insights with their titles, last modified dates, and a Run button for each


Data Artifacts are managed within the Insight editor.

  1. In the Insight editor, click + Data Artifact.
  2. Enter a title and a unique key (used in {{table source="data_artifact_id"}}).
  3. Write the SQL query. The editor validates syntax in real time.
  4. Click Preview to execute a test run and inspect the output columns and rows.
  5. Click Create Data Artifact to attach the artifact to the Insight.

Data Artifact editor with a SQL query input field, real-time syntax validation, and a row preview table below the editor

  1. In the Data Artifacts list, click the artifact’s title.
  2. Edit the SQL or title as needed.
  3. Click Save.

The artifact key cannot be changed after creation. If you need a different key, delete the artifact and create a new one.

  1. In the Data Artifacts list, open the three-dot menu next to the artifact and click Delete.
  2. Confirm deletion.

Deleting an artifact removes its binding from the Insight. Any {{table source="data_artifact_id"}} tags referencing it will fail to render on the next run.


Once an Insight is working, you can deliver its rendered output to stakeholders automatically via Reports.

Supported delivery channels:

  • Email
  • Slack
  • Microsoft Teams
  • Google Chat

To set up a Report:

  1. In the Insight editor, click Send & Schedule (or open the Destinations tab).
  2. Select the delivery channel and configure the destination.
  3. Set a schedule (daily, weekly, monthly, or interval-based).
  4. Click Create & Run report to save and immediately deliver the report, or Create new report to save it for scheduled delivery only.

On each scheduled run, OWOX will execute the Insight and send the rendered Markdown output to the configured channel.

Reports use the same Insight template and Data Artifacts. Keeping the template up to date automatically updates all scheduled reports that reference it.

Send & Schedule panel showing delivery channel options — Email, Slack, Microsoft Teams, and Google Chat — with a schedule picker

Example of a delivered Insight report rendered as a formatted message in a connected channel


ActionRequired Role
View Insights and preview outputViewer
Create, edit, delete InsightsEditor
Add or remove Data ArtifactsEditor
Run or cancel an InsightEditor
Create or manage ReportsEditor
LimitValue
Data Artifacts per Insight5
Rows returned per Data Artifact on run100
Artifact key formatAlphanumeric and underscores ([a-zA-Z0-9_]), max 64 characters