How to Connect AWS Athena to Google Sheets Without CSV Exports

Why the CSV round trip breaks

The export is stale when it is made, and the query lives in one person’s editor, so a second analyst’s version of the same KPI sends a second number into a second Sheet. The part that is specific to Athena is the bill. Athena charges per terabyte of data scanned, $5 per TB on the AWS pricing page (checked October 2026), and a rerun scans the same data again unless the query opts into Athena’s result reuse, which serves a stored result up to a maximum age you set and can return a result that is stale until that age runs out.

How much a scan costs depends on how the data is stored. AWS’s own example is a 3 TB uncompressed text table: a query on one column scans all of it and costs $15. The same data compressed and converted to Parquet, a columnar format, costs $1.25 for that query, because Athena reads only the column the query names.

Which route fits: download, pull, sync or script

Every route that reads Athena runs a query, and Athena bills the scan. So the route does not change the price of one run. It changes how many runs you get and who decides when they happen. The vendors below describe their own products on their own pages (checked October 2026).

The runs a route produces cost real money. A Data Mart that scans 50 GB reads 0.0488 TB (at 1,024 GB to the TB), which at $5 per TB is $0.244 a run. Over a day, rounded:

These are scan charges only, and the 50 GB is an example. The rule that follows is short: let the load schedule of the table set the refresh schedule, not the other way round. A table that loads once a day needs one run a day, whichever route you pick, and anything faster pays for the same data again. Pull only when someone needs a number between loads, and point that pull at a small summary table so each click scans little.

The rest of this guide builds the push route: one query, published once, delivered on a schedule. Your data stays in S3 and Athena stays the query engine. OWOX Data Marts sends the SQL to Athena, reads the result and writes it into the Sheet. It does not transform the data: the SQL is the analyst’s.

What you need

• An AWS account with Athena enabled, a Glue database or tables to query, and an S3 bucket or folder for query results.
• An IAM user with an access key.
• A Google account, or a Google Cloud service account, for the Sheets destination.
• An OWOX Data Marts project. Reports and report triggers also need a license key, created in an OWOX Data Marts Cloud project: without a valid one, the docs say every report run finishes as Restricted.

Step 1: connect Athena as a storage

In OWOX Data Marts, open Storages, click + New Storage and pick AWS Athena. The form asks for five things, all described in the storage guide:

• Title: any name your team will recognise.
• Region: for example us-east-1.
• Output Bucket: the S3 bucket for query results, or a folder in it such as my-athena-results/owox/.
• Access Key ID and Secret Access Key of an IAM user.

The IAM policy in the guide has three parts: Athena actions to list catalogs and databases and to start, stop and read query executions; Glue actions on databases and tables; and S3 actions to list, read, write and delete in the buckets. The published policy applies them to all resources, so scope it down to your own buckets and databases before it goes to production, and give the key to a dedicated user.

One behaviour is worth knowing before the first run. OWOX Data Marts deletes the results of the queries that read your data once it has read them. The small results of its access checks, SQL validation and last-updated lookups stay in the bucket under athena-check-access/, athena-dry-run/ and athena-data-last-updated/, and an S3 lifecycle rule can clear them.

A Data Mart cannot be validated or published until the storage is fully configured, so finish this step first.

Step 2: publish the query as a Data Mart

Click + New Data Mart, name it, pick the Athena storage and open Data Setup. Set Definition Type to SQL, write the query, wait for the validation and click Save, then publish. The SQL below is the one behind the demo Data Mart in the screenshot, running on a synthetic e-commerce dataset:

SELECT s.date,
    s.session_id,
    s.source_medium,
    s.device_category,
    c.customer_id,
    c.customer_segment,
    c.region AS customer_region,
    c.registration_date,
    date_diff('day', c.registration_date, s.date) AS days_since_customer_registration,
    o.order_id,
    p.product_name,
    p.category AS product_category,
    o.quantity,
    COALESCE(p.price * o.quantity, 0) AS revenue,
    COALESCE((p.price - p.cost) * o.quantity, 0) AS gross_profit,
    IF(o.order_id IS NOT NULL, 1, 0) AS is_purchase
FROM ecommerce_demo.sessions s
    LEFT JOIN ecommerce_demo.orders o ON s.session_id = o.session_id
    LEFT JOIN ecommerce_demo.customers c ON s.customer_id = c.customer_id
    LEFT JOIN ecommerce_demo.products p ON o.product_id = p.product_id

This is the step that stops two Sheets disagreeing. The query has an owner and one place it lives. A Data Mart can also be a table or a view that already exists in your storage: use those when the logic is heavy and build it in Athena on a schedule, so reports read a small finished result.

Step 3: describe the output

Once the query is saved, the Output Schema is generated: column names and the types Athena returns. You can mark primary keys and add an alias (a business-friendly name) and a description for each field.

The fields in the screenshot have none yet, which is the common state of a first Data Mart. Fill them in before you share the Sheet: the next person should not have to read the SQL to learn what revenue includes. The Data Mart management page covers ownership and access.

Step 4: connect Google Sheets as a destination

A destination is set up once and reused by every report. In Destinations, click + New Destination, choose Google Sheets and pick how to authenticate, as described in the Google Sheets destination guide:

• Google OAuth: connect a Google account and OWOX refreshes the tokens automatically.
• Service account: paste the JSON key of a Google Cloud service account. The guide recommends this for automated, unattended workflows.

Enable the Google Sheets API in the Cloud project for the service account route. If you want OWOX to create new spreadsheets inside a Drive folder, enable the Google Drive API too: the Sheets API cannot place a file in a folder.

Step 5: create the report

Open the Data Mart’s Destinations tab. Each destination has its own block with a + New Report button.

In the form, name the report and paste the link to the spreadsheet tab, with its gid in the URL, or click + New Sheet to have one created. For an existing spreadsheet, share it with the email shown under Share document with and give it Editor access. With OAuth you can skip that if your Google account can already edit the Sheet. Choose the columns the reader needs, not all of them, and click Create & Run report.

A report can also narrow what it delivers without touching the SQL. Output controls let you filter rows, sort and cap the row count per report.

Step 6: put it on a schedule

A report you run by hand is still an export. On the Data Mart’s Triggers tab, click New Trigger, choose Report Run, pick the report and set the schedule: daily, weekly, monthly or an interval, with a time zone. The form offers presets such as every hour and every 6 hours.

A report trigger does not wait for a connector import. If a connector loads the data the report reads, schedule the report after the import usually finishes. Every run, scheduled or manual, is listed in Run History, so a failure is visible without opening the Sheet.

Keeping the push cheap

Google Sheets is a push destination: OWOX Data Marts sends the data on each run, and the Sheet stays static between runs. A report is therefore one Athena query per run, however many people open the Sheet afterwards. If readers need fresher data than a daily run gives, build a smaller summary table in Athena and point the Data Mart at it, so each run scans less. S3 storage and requests for the query results are billed separately by AWS.

For a deeper read on how Athena works and what else it bills, see AWS Athena 101.

Let people pull their own slice

A scheduled report fits a fixed deliverable. Some people want to choose their own columns and filters. For them there is the OWOX Data Marts extension for Google Sheets: business users open it inside the Sheet, pick a published Data Mart, filter, schedule and refresh the data themselves, without an AWS login. The analyst still decides what is published.

If a question needs fields from two Data Marts, joinable Data Marts let you declare the relationship once, on shared keys, instead of writing one large query. Both Data Marts must sit on the same storage, and Athena is a supported one. Spreadsheet reporting shows what the business user’s side looks like.

Amazon Athena

Amazon AthenaHow to Connect AWS Athena to Data StudioOctober 20, 2026→Amazon AthenaAWS Athena 101: Serverless SQL on S3 ExplainedOctober 1, 2026→Amazon AthenaAWS Redshift vs Athena: Key Differences ExplainedOctober 21, 2026→

See all articles →