Automate Snowflake reports in Google Sheets with OWOX Data Marts

What goes wrong without a shared definition

The spreadsheet is rarely the problem. The path into it is.

• Exports drift. Each one is a snapshot, and nobody can say which snapshot a chart was built from.
• Logic forks. One analyst counts refunded orders, another does not. Both sheets say “revenue”.
• Every question is a ticket. The analyst becomes the refresh button.
• Live connections are paid for by the query. In Snowflake a warehouse bills for every second it is awake, so a sheet that re-queries on open keeps it awake.

All four have the same fix: put the definition in one place, and deliver from there.

How the pieces fit

Three objects do the work.

A Data Mart is a SQL query, a table or a view that an analyst publishes under a business name, with documented fields. It lives on your Snowflake storage; nothing is copied elsewhere.

A destination is where results are delivered. Google Sheets here, though the same Data Mart can also go to Data Studio, Slack or email.

A report connects them: this Data Mart, these columns, that tab, on this schedule.

The split matters because it separates two kinds of change. The analyst changes the logic in the Data Mart. The reader changes what they do with the numbers in the sheet. Neither can break the other’s work.

Step 1: connect Snowflake, on its own warehouse

Add a storage of type Snowflake. The setup guide asks for an account identifier, a warehouse name and one of two authentication methods: a key pair, which it recommends, or a username with a programmatic access token.

The guide also recommends a dedicated warehouse for OWOX Data Marts, and for a reporting workload that is worth doing properly:

CREATE WAREHOUSE reporting_wh
  WAREHOUSE_SIZE = 'XSMALL'
  AUTO_SUSPEND = 60
  AUTO_RESUME = TRUE;

An X-Small warehouse uses 1 credit per hour and bills per second with a 60-second minimum. With auto-suspend at a minute, a report that runs once a day costs about a minute of X-Small time a day, and it has its own line when you look at credits by warehouse. Snowflake pricing explained has the rates and the query for that.

Step 2: publish the query as a Data Mart

Create a Data Mart on the Snowflake storage and choose its input: SQL, an existing table or an existing view. A report of sessions, orders and revenue by channel and day looks like this:

SELECT
  s.date,
  s.source_medium,
  COUNT(DISTINCT s.session_id) AS sessions,
  COUNT(DISTINCT o.order_id) AS orders,
  SUM(p.price * o.quantity) AS revenue
FROM ecommerce.public.sessions s
LEFT JOIN ecommerce.public.orders o
  ON s.session_id = o.session_id
LEFT JOIN ecommerce.public.products p
  ON o.product_id = p.product_id
WHERE s.date >=
  DATEADD(day, -90, CURRENT_DATE())
GROUP BY s.date, s.source_medium;

The query is checked against Snowflake before it is saved, and the output schema is read from it. That schema is where you make the Data Mart usable by someone who has never seen the tables: an alias and a one-line description for each field, so “revenue” says what it includes.

OWOX Data Marts does not transform the data or rewrite the SQL. If the joins are heavy, build the result as a table in Snowflake on a schedule and publish that table, so each report reads a finished result.

Ad platform data follows the same pattern. If campaigns are loaded into Snowflake by a connector, as in Facebook Ads to Snowflake, the connector’s table is already a Data Mart and can be reported from directly.

Step 3: add Google Sheets as a destination

Create a destination of type Google Sheets once for the team. The destination guide gives two ways to connect it: Google OAuth with your own account, or the JSON key of a Google Cloud service account. The spreadsheets that reports write to are then shared with that account.

Step 4: create the report and choose its columns

On the Data Mart’s Destinations tab, create a report under the Google Sheets destination. Give it a title, paste the link to the tab it should fill, or create a new sheet from the form, and pick the columns.

Choosing columns per report is how one Data Mart serves several audiences. The performance team takes campaign and ad set detail. The leadership summary takes five fields. Both read the same definition.

When the report runs, the rows are written into the sheet. The report owns only the columns it writes: headers and data in that range are replaced on each refresh, and whatever sits to the right of it, formulas, comments, a pivot, is preserved.

Step 5: schedule it

On the Triggers tab, add a report run trigger with the cadence the data deserves: daily before the team starts, weekly on Monday, or more often if the source really changes that fast. Run History then lists each run with its time and status, which is where you look first if someone says the sheet looks old.

Letting people pull the data themselves

Scheduled reports cover fixed deliverables. For everything else there is the OWOX Data Marts extension for Google Sheets. A business user opens it in a spreadsheet, picks a published Data Mart, chooses columns, adds filters and sorting, and runs it.

The extension can keep that pull on a schedule too, so the sheet a person built for themselves stays current.

They never see the SQL and never need a Snowflake login. They can only reach what an analyst has published, which is the control that makes self-service safe. There is an Excel add-in for teams that work there.

Structuring the sheet for the people who read it

A few habits keep a reporting spreadsheet usable after the third month.

• One imported tab per report, untouched. Build charts and summaries on separate tabs that reference it.
• Put your own columns to the right. They survive every refresh.
• Name tabs for the reader. “Channels, daily” beats “Sheet7”.
• Say where it comes from. A line with the Data Mart’s name and refresh time saves the question.

For the wider picture of where this layer sits, Snowflake as a reusable reporting layer covers the analyst’s side and self-service marketing analytics on Snowflake covers the marketing team’s. The same route on Databricks is described in Databricks to Google Sheets.

SnowflakeHow to load Facebook Ads data into Snowflake with OWOX Data MartsOctober 1, 2026→SnowflakeHow to load Google Ads data into Snowflake with OWOX Data MartsOctober 1, 2026→SnowflakeHow to load TikTok Ads data into Snowflake with OWOX Data MartsOctober 1, 2026→

See all articles →