Stop being the interface to your warehouse

Connect BigQuery, Snowflake, Databricks, Postgres, MySQL, MSSQL, MongoDB, or a data lake through Athena. Papercrane's agent writes the query, shows you every one, and dashboards refresh on a schedule instead of every time someone looks.

Browse the integrations

Read access you approve, in your own words. Free to connect and ask. Paid plans start when you publish a dashboard. See pricing

Six engines, one tax

Every warehouse and database has its own way of billing your curiosity, in dollars, in seconds, or in someone's afternoon. The fix Papercrane applies is identical everywhere: write the real query, show it to you, and stop running it every time a viewer opens a tab.

What each engine actually meters

These are six different billing models. Below is what each one charges for, and what changes when a dashboard refreshes on a schedule instead of on every view.

What each engine meters

Representative examples

Engine	What it meters	The old way	Papercrane's way
BigQuery	Bytes scanned per query	A live dashboard rescans on every open: roughly $168/month for a typical reporting workload	Scheduled refresh scans once per interval: roughly $14/month for the same dashboard
Snowflake	Seconds the warehouse stays awake	Extract refreshes and live queries keep it awake most of the day: around 46 credits	Scheduled refreshes wake it once, briefly: around 1 credit
Databricks	Engineering time to reach an answer	A question becomes a ticket that moves through the bronze, silver, gold pipeline	A question becomes one conversation, answered from the same tables
Postgres, MySQL, MariaDB, MSSQL	Whoever holds the login	Every ad hoc question interrupts a person	A scoped read only login lets the agent query on its own schedule
MongoDB	Time spent hand writing aggregation pipelines	Each new question needs a new pipeline someone builds by hand	The agent writes the pipeline and shows you every stage
Athena	S3 scanned per query, repeated on every ask	The same question rescans the lake each time it's asked again	A published dashboard scans once per refresh and serves everyone from that result

Figures are representative examples from real worked cases on each tool's own page. Open any row for the full arithmetic against your own numbers.

Connect the warehouse you already have

Six ways in, one dashboard on the other side. Pick the one that matches your stack.

BigQuery

What does a question cost in BigQuery?

Snowflake

Dashboards that let your Snowflake warehouse sleep

Databricks

Everything lands in Databricks. Answers still leave as screenshots.

MongoDB

Dashboards from MongoDB, without writing the pipelines

AWS Athena

Where does an Athena query go after it runs?

SQL Database

Can you pull me a quick list from the database?

Migrating off Tableau, Looker, or Qlik? Every one of those BI tools was pointed at a warehouse just like these. See Analytics & SEO.

When you don't need us

If your team already has a staffed BI function running governed models, analysts building in Looker or a dbt metrics layer every day, keep it; that governance is worth its seats. Sigma is a genuinely good answer for a team that lives in a warehouse workbook all day. Papercrane is worth adding when a question falls outside that governed model, or when the answer needs to reach someone who doesn't have a warehouse seat at all.