Search Console connects to Data Studio through a free, Google-built connector — but there’s a choice on the setup screen that catches almost everyone, and picking wrong means rebuilding your report.
This covers the connection, the Site vs URL Impression decision, how to verify it worked, and why some of your query data will be missing no matter what you do.
What you need first
- A Google account with access to the Search Console property
- The property already verified in Search Console
- Nothing else — the connector is free and native
Connect Search Console, step by step
- Go to datastudio.google.com, signed in with the account that has Search Console access.
- Create → Data source.
- Choose “Search Console” — marked By Google, so you know it’s the free native one.
- Pick your site. Domain properties and URL-prefix properties both appear; pick the one you actually use.
- Choose the table. This is the important bit — see below.
- Choose the search type — Web, Image, Video, News, Discover or Google News.
- Connect, then Create report.
Site Impression vs URL Impression — pick carefully
A single Search Console data source can use one or the other, never both. This is the decision people get wrong.
| Table | What you get | What you lose |
|---|---|---|
| Site Impression | Site-wide totals by query, country, device and date | No landing page dimension — you cannot break anything down by URL |
| URL Impression | Page-level data, including landing page | Query data is much more limited at this level |
If you want queries and pages in the same report — which most SEO reports do — you need two data sources, one of each, added to the same report. That’s normal and supported; it just isn’t obvious from the setup screen.
How to check it worked
Add a scorecard for Clicks over the last 28 days, then compare with the Performance report in Search Console itself.
Expect them to be close but not identical — Search Console’s own interface and its API round and aggregate slightly differently. A wildly different number usually means the wrong property (a domain property and a URL-prefix property for the same site report different totals) or a search type you didn’t intend.
Why some of your query data is missing
This is the single most common Search Console question, and it isn’t a Data Studio fault.
Google deliberately withholds rare queries. In their words, “to protect user privacy, Search Analytics doesn’t show all data” — and in Data Studio, those withheld queries get aggregated into a single row with a blank query value.
So when your clicks by query don’t add up to your total clicks, nothing is broken. The gap is the withheld long tail. Two practical consequences:
- Never present query-level totals as if they’re complete. They aren’t, and the gap can be large on smaller sites.
- Filter out the blank row in query tables, or label it clearly, so nobody reads it as a real query.
Search Console also only retains 16 months of data. Data Studio queries it live, so if you need longer history you have to store it yourself — a Sheet, or BigQuery via the Search Console bulk data export.
What to build first
- Clicks, impressions, CTR and average position as scorecards, with a previous-period comparison
- Clicks over time, to see the shape
- Top queries (Site Impression source) — sorted by clicks
- Top landing pages (URL Impression source) — the second data source earning its place
- A branded vs non-branded split, which is where this report starts being genuinely useful
That last one needs a calculated field. A simple version: a CASE statement flagging any query containing your brand name as Branded, everything else as Non-branded. It turns a flat query list into a story about whether you’re growing demand or just harvesting it.
Joining Search Console to GA4
The natural next step, and worth knowing the catch before you start: Search Console gives full URLs while GA4 gives paths, so a blend on landing page won’t match without work.
Fix it on one side with a calculated field — either CONCAT("https://yourdomain.com", Landing Page) on the GA4 side, or strip the domain off the Search Console side. Remember GA4 stores paths without a trailing slash, which is the detail that silently breaks the join.
And keep the five-source blend cap in mind — with two Search Console sources already in play, you have less headroom than you think.
Do you ever need a paid connector for this?
Rarely. The native connector is free, maintained by Google, and covers what most SEO reporting needs.
The exceptions are history and scale: if you need more than 16 months, or you’re reporting across many properties and want them consolidated, a paid connector or a BigQuery export starts to earn its cost. Windsor.ai, Supermetrics and Coupler.io all handle Search Console, usually alongside the paid ad platforms you’re already paying them for.
If Search Console is all you’re reporting on, the free connector is the right answer and always will be.
Frequently asked questions
Is the Search Console connector free?
Yes. It’s built by Google, listed under “By Google”, and costs nothing.
Why can’t I see queries and landing pages together?
A Search Console data source uses either the Site Impression table or the URL Impression table, not both. Add two data sources to the same report — one of each.
Why don’t my query clicks add up to my total clicks?
Google withholds rare queries to protect user privacy, and those get grouped into a row with a blank query value. The difference is that withheld long tail, not a broken connector.
How far back does the data go?
Search Console retains 16 months. Data Studio reads it live, so that’s your ceiling unless you export the data to Sheets or BigQuery yourself.
Should I use the domain property or the URL-prefix property?
Whichever matches how you actually work. They report different totals for the same site, so pick one and stay consistent, or your month-on-month comparisons will be meaningless.
Why does my report show fewer clicks than Search Console’s interface?
Small differences come from rounding and aggregation between the interface and the API. Large differences usually mean a different property or search type.