BI.P.EYE logo
Start free trial
How to start

Test the full product in your own free trial

Follow these steps with a sample file of 1,000 Shopify orders, or with your own data. By the end you will have run a pipeline, built a model, asked questions in a dashboard and in Claude, and seen how embedding works.

Start free trial
30-day free trialFull access to every feature. No credit card.
Sample data includedShopify_orders.json: 1,000 orders with their products, about 4.7 MB.
Help when you need itEmail [email protected]
The route
1Sign up→ 2Add data→ 3Run the pipeline→ 4Your warehouse→ 5Schedule→ 6Build a model→ ADashboardor BClaude→ 7Embed
1Sign up

Create an account and start the free trial

  1. Go to bipeye.com/auth/signup
  2. Sign up with your email, or with Google
  3. The 30-day free trial starts right away. No credit card.
  4. Click Create your first dataset →
Full access to every feature for 30 days, with unlimited datasets. Nothing to install.
2Add data

Add your first dataset from a file, app or warehouse

  1. Open Datasets and click + Dataset
  2. Pick a source: a JSON File or CSV File, or an app or warehouse you added in Credentials. For this guide, choose JSON File.
  3. Upload Shopify_orders.json and wait for Ready
  4. Click Continue to pipeline settings
The same flow works for every source: databases (Postgres, SQL Server, MySQL, MongoDB), warehouses (Snowflake, BigQuery), Amazon S3, Shopify and any HTTP API.
3Pipeline

Clean and shape the data, without code

  1. The preview shows the raw rows, as a table or as JSON (1,000 orders)
  2. Click + between steps to add Filters, Select Columns or Flatten Array
  3. Click Test ETL to preview the result
  4. Click Save & Run
Nested JSON, like the list of products inside each order, becomes a clean table with one click on Flatten Array.

Choose the columns and where the data is stored

  1. Select Columns: keep only the fields you need, and rename them
  2. Data Warehouse step: choose the destination. The BI.P.EYE warehouse, your own Postgres, or your Snowflake.
  3. Upsert (optional): update existing rows by key columns, such as orderName + date, instead of adding duplicates
Destination
Upsert on orderName + date
Select Columns

The result: three clean datasets from one file

In the sample account, one upload became three datasets. Each finished run is added to the default model automatically.

Shopify_orders.json
1,000 rows

The raw upload. One row per order.

Orders flat only
1,000 rows

Selected columns with upsert on. Ready for order-level charts.

Product Flat
1,552 rows

The products list flattened. One row per product sold.

4Connect your own systems

Add credentials, use your own warehouse

  1. Open Credentials and create a connection: MySQL, PostgreSQL, SQL Server, Snowflake, BigQuery, MongoDB, Amazon S3, Shopify and more
  2. Fill in the connection details. For Snowflake: account identifier, warehouse, role, and sign-in with a Key pair (recommended) or password.
  3. Allow the BI.P.EYE IPs shown at the top (IP whitelist), then click Test Connection and Save
  4. In any pipeline, open the Data Warehouse step and choose your own Postgres or Snowflake as the destination
The same credentials work as a source and as a destination, so your clean data can stay in your own warehouse.
5Automate

Run datasets in order, on a schedule

  1. Dependencies: in a dataset, set Run when finished to the dataset that should run before it
  2. See the whole chain in Datasets → Pipeline
  3. Schedule: set Interval on the first dataset, for example Run daily at 06:00 (your local time)
  4. When it runs, every dependent dataset runs after it, in order
Data refreshes on its own. Each step waits for the one before it, and failures trigger alerts.
EXAMPLE SETUP
  • Shopify_orders.jsonInterval: daily 06:00
  • Orders flat onlyAfter Shopify_orders.json
  • Product FlatAfter Orders flat only
6Model

Connect the datasets in a model

  1. Open Models. Every dataset is already in the Default model.
  2. Click + New model and pick the datasets that belong together, for example orders with products
  3. On the map, drag the dot from the orders dataset (fact) onto a dimension such as Products, and pick the matching fields
  4. To use the model in Claude, set Available in Claude to Everyone. Every model starts Off.
The model is the semantic layer. Dashboards, dashboard chat and Claude all use its relationships, so the numbers match everywhere.
Then choose a path

Ask questions in a dashboard, or in Claude

Both read the same model. Try one, or both.

PATH A

Dashboard

Build charts in the UI, or turn on chat and ask in plain language.
PATH B

Claude

Connect the BI.P.EYE MCP connector and ask about the same data in Claude.
APath A · Dashboard

Build a dashboard, by hand or with chat

  1. Open Dashboards → + Dashboard → Blank Dashboard
  2. By hand: click + Add Chart and build charts in the UI
  3. With chat: click Set up chat, turn on Enable chat on this dashboard, choose your model and Save
Each viewer sees only the data their permissions allow, also in the chat.

Ask in plain language, get a chart back

Try these on the sample data
Please show orders over time Show top products by country and appName
  • The answer shows the workflow it built, the datasets it used and the join
  • Save to dashboard keeps the chart. Advanced edit opens the full editor.
  • Type a follow-up to change the chart
BPath B · Claude

Ask about the data directly in Claude

  1. In BI.P.EYE: Models → set Available in Claude to Everyone
  2. In Claude, open claude.ai/directory/bipeye and click Connect
  3. Sign in with the same BI.P.EYE account and approve access
  4. Start a new chat and ask a question

Full setup guide: bipeye.com/claude.html

Claude queries the BI.P.EYE models, never your databases directly. Large results come back as a download link.
Claude + BI.P.EYE connector
QUESTIONS TO TRY ON THE SAMPLE DATA
Which models can you see in BI.P.EYE? What are the top 10 products by total sales? Show total sales by country and by month Give me all orders with their products as a table I can download
list_modelsdescribe_modelquery_model

Ask about the model that links orders and products, so Claude can join them. The connector is read-only and signs in with OAuth.

7Embed

Put the dashboard inside your product

  1. Try it first in Playground in the left menu
  2. Open OpenAPI → API Keys and create a key
  3. Embed the dashboard with an iframe or the Embed API, on a website, a customer portal or an internal system
  4. Limit each key by IP, allowed domain and row-level rules, so every customer sees only their own data

Developer docs: docs.bipeye.com

White-label, in your own branding, for managers, teams and mobile.
Illustration of an embedded dashboard

Common questions

What is included in the trial?

Full access to every feature for 30 days, with unlimited datasets and no credit card. Pipelines, models, dashboards, dashboard chat, the Claude connector and embedding are all included.

Can I use my own data instead of the sample file?

Yes. Upload any CSV or JSON file, or connect a database, warehouse, Amazon S3, Shopify or any HTTP API from Credentials. The steps are the same.

Where is my data stored?

In the BI.P.EYE warehouse, in an isolated dedicated warehouse, or in your own Postgres or Snowflake. You choose the destination in each pipeline.

Who can see what?

Team roles control access inside your company, and row-level rules control what each embedded customer sees. Dashboard chat and Claude follow the same permissions.

What happens after the trial?

Pricing has no per-user fees and follows data volume and usage. See pricing, or email [email protected].

Can your team help us set it up?

Yes. The team also helps with data modeling, pipelines and dashboards. Email [email protected].

Ready to try it on your data?

30 days, full access, no credit card.

Start free trial Talk to the team