Search Console Bulk Data Export: An Agency BigQuery Guide

Key takeaways

Search Console report tables are limited to 1,000 rows of representative examples.
The API returns up to 50,000 rows per day per search type per property, in pages of up to 25,000.
The bulk data export sends performance data to BigQuery every day, excluding anonymized queries.
It starts from the day it is set up, so set it up early for large clients.

The bulk data export is a reporting tool for larger sites, and it changes what an agency can promise in monthly analysis. We have used large-site data on projects such as a 70,000-page directory, where clicks were spread across far more pages than any interface table shows. This guide is part of the reporting stack we run under agency brands through white label SEO; dashboards built on top are covered in white label SEO reports in Looker Studio.

Why does Search Console only show 1,000 rows?

The report tables in the interface show up to 1,000 rows of representative examples of the data, while the totals at the top still reflect the full dataset. Exports from the report to Google Sheets, Excel or CSV carry the same limit. For more rows, Google points to the Search Console API or the bulk export to BigQuery.

That is stated in the help article on how to export data directly from a Search Console report. For a site with a few hundred pages, 1,000 rows covers almost everything. For a catalog or directory, it means most URLs and long-tail queries never appear in a table, which makes page-level analysis and cannibalization checks unreliable.

How many rows can you get from the Search Console API?

Google says the API exposes performance data limited to 50,000 rows per day per search type per property. Each request returns up to 25,000 rows, so agencies page through results by increasing the start row until a response comes back empty. The API is free, has usage quotas and works well for medium-sized sites.

The daily cap is in the help article on how to export data using the Search Console API, and the paging method is described in the API guide Getting all your data, which also says data is typically available after 2 to 3 days. The usage limits page lists Search Analytics quotas such as 1,200 queries per minute per site and per user. Connectors used by reporting tools rely on this API, so they share its limits.

What is the Search Console bulk data export?

It is a scheduled, daily transfer of a property's Search Console performance data into a BigQuery dataset the agency or client controls. It includes all performance data available for the property except anonymized queries, and keeps it in BigQuery for as long as you choose, so the long tail of a large site is not cut off by interface tables.

Google's overview, About bulk data export, describes the scope and the anonymized-query exclusion. The export creates tables described in the table guidelines and reference:

Table	What it holds	Use it for
searchdata_site_impression	Performance data aggregated by property	Query trends, brand vs non-brand, totals
searchdata_url_impression	Performance data aggregated by URL	Page-level analysis, templates, cannibalization
ExportLog	A record of each completed export	Checking that daily exports are running

Rare queries are flagged with is_anonymized_query and their query text is empty, so query totals will not match click totals exactly. The tables are partitioned by data_date, the day the data refers to in Pacific Time.

How do you set up the bulk data export?

Create or choose a Google Cloud project with billing and BigQuery enabled, grant Search Console's service account the BigQuery Job User and BigQuery Data Editor roles, then enter the project ID in the bulk data export settings of the property. The first export arrives within 48 hours and does not include history from before setup.

Those steps are in Start a new bulk data export. The same page says the dataset name always starts with searchconsole, data accumulates indefinitely unless you set an expiration of 14 days or longer, and changing the schema makes the export fail. For older data, it points back to the API or the reports, so pull a historical backfill through the API on the day you set up the export.

Decide who owns the Cloud project: ideally the client, with the agency as a user.
Set a budget alert on the billing account before the first export.
Prefer partition expiration over table expiration if you want to limit retention.
Pull the history the API still offers on the same day, as a backfill.
Check the ExportLog table weekly for missing days.

How much does the bulk data export cost?

Search Console does not charge for the export, but the data lives in Google Cloud and is subject to BigQuery storage and query costs. Google says there is a free usage level and usage above the free quota is charged. Queries that scan only the dates they need are the main way to keep costs low.

The cost note is in the setup article, and the query guidelines recommend a WHERE clause on the date-partitioned tables to limit the data scanned, and remind users that every query is charged. Explain to the client in writing who pays the Cloud bill before you turn the export on.

Which reports should agencies build on the export?

Build the reports the interface cannot give: performance by page template or folder across every URL, long-tail query growth, queries split across several URLs, brand versus non-brand trends over years and position changes after releases. Keep the monthly client report simple and use the raw tables for analysis and quarterly reviews.

Template reports

Clicks and impressions by folder or page type across all URLs.

Cannibalization

Queries where several URLs share impressions.

Long tail

New queries appearing each month on large catalogs.

Long history

Year-over-year comparisons beyond the interface window.

Google's sample queries include average position, computed from sum_top_position divided by impressions plus one, because position in the tables is zero-based. Use the same formula in every client dashboard so numbers are comparable. For big sites, this work is part of our white label enterprise SEO. What to show clients each month is in white label SEO reporting. To set up exports for your larger clients, book a 30-minute call.

Frequently asked questions

Do small sites need the bulk data export?

Usually not. If the site has fewer pages and queries than the 1,000-row tables show, the interface and API are enough.

Can the export include data from before setup?

No. Google says the first export covers the day of the export; use the API for earlier data.

Who needs access to set it up?

Someone who can configure the property in Search Console and grant roles in the Cloud project.

Can Looker Studio read the BigQuery tables?

Yes, Looker Studio can connect to BigQuery, which is useful for large-site dashboards.

Will totals in BigQuery match the interface?

Close but not always exact, because anonymized queries have no query text and aggregation differs by table.

About the author

Juan José Giraldo is a senior SEO specialist with 5+ years of experience. He works on technical and local SEO for 600+ US car dealerships and supervises the SEO of 18 US e-commerce stores. He runs White Label SEO LATAM from Medellín, Colombia, and delivers white label SEO for agencies in the US, UK, Canada and Australia. Read more · LinkedIn.

Search Console bulk data export: when agencies outgrow the 1,000-row interface.