Gemini SEO: what Gemini reads before it names a product

Gemini grounds its answers in Google's search index, so a page Google cannot index is a page Gemini cannot cite, and Google-Extended controls whether it may use your content at all.

Does Gemini name you, or your competitor?

Paste your URL. See what ChatGPT, Perplexity and Google AI answer when your buyers ask, who they name instead of you, and which pages they read. Then we do the work to get you onto those pages.

Free scan, no signup, no card.

What Gemini cited in our test

Answers measured: 110
Sources per answer: 9.2
Answers citing Reddit: 20%
Answers citing YouTube: 3%

11 of 110 answers cited no source at all. Across the rest, Gemini cited 274 different domains. We did not re-ask Gemini the same question in this test, so we have no stability number for it.

The domains it cited most

Domain	Answers citing it
reddit.com	22 of 110
semrush.com	7 of 110
startupsubmit.app	6 of 110
zapier.com	5 of 110
discoveredlabs.com	4 of 110
okara.ai	4 of 110
redship.io	4 of 110

Next to the other engines

Engine	Sources per answer	Cites Reddit
ChatGPT	5.8	6% (7 of 110)
Perplexity	10.1	31% (32 of 103)
Gemini	9.2	20% (22 of 110)
Google AI Mode	3.4	40% (44 of 111)

Source: welaunch's test of 434 live AI answers to buyer questions in seven software categories, 27 Sep 2026, US. Brand and domain counts are answers, not citations: an answer citing reddit.com three times counts once. welaunch.sh was named or cited in 0 of these 110 answers.

How Gemini picks its sources

Google-Extended controls whether content Google crawls may be used to train Gemini models and for grounding, meaning content from the Google Search index given to the model at prompt time, in Gemini Apps.
Source: Google crawler documentation
Google-Extended has no user agent of its own. Crawling is done by Google's existing crawlers, and the token is only a robots.txt control. Google says it does not affect inclusion or ranking in Google Search.
Source: Google crawler documentation
It cited 9.2 sources per answer on average; 11 of 110 answers cited none.
Source: welaunch's test of 434 live AI answers to buyer questions in seven software categories, 27 Sep 2026, US
reddit.com was its most cited domain, in 22 of 110 answers (20%).
Source: welaunch's test of 434 live AI answers to buyer questions in seven software categories, 27 Sep 2026, US
It drew on fewer domains than Perplexity (274 against 664 over a similar number of answers), so the pages it trusts repeat more often.
Source: welaunch's test of 434 live AI answers to buyer questions in seven software categories, 27 Sep 2026, US

Which Gemini crawlers to allow

Googlebot Google's crawler. Gemini's grounding reads from the Google Search index it builds.
Google-Extended A robots.txt token, not a separate crawler. Controls Gemini training and grounding in Gemini Apps. Does not affect Google Search.

Add these lines to robots.txt:

User-agent: Googlebot
Allow: /

User-agent: Google-Extended
Allow: /

A crawler given its own group stops reading your User-agent: * rules, so repeat any private paths you disallow there. Then check the firewall: a CDN rule can block a crawler that robots.txt allows. Check both free with the AI crawler checker.

What to fix first for Gemini

01
Do not block Google-Extended by accident
Some robots.txt templates and CDN presets block every AI token. Blocking Google-Extended keeps your content out of Gemini's grounding in Gemini Apps while Google Search carries on as normal, so the loss is easy to miss.
02
Make sure Google indexes the pages that answer
Grounding reads the Google Search index. Pricing, comparison and alternatives pages that are not indexed cannot be cited. Check with a site: search.
03
Get onto the pages it keeps citing
Gemini repeated a smaller set of domains than Perplexity. Find which list articles and directories it cites for your buyer questions and work on those.
04
Use Reddit where it is already cited
reddit.com was in 20% of its answers, below Google AI Mode and Perplexity but well above ChatGPT.

Gemini questions

What is Google-Extended?

A robots.txt token Google provides to control whether your content may be used to train Gemini models and to ground answers in Gemini Apps. It has no crawler of its own; Googlebot does the crawling.

Does blocking Google-Extended hurt my Google rankings?

Google says it does not affect a site's inclusion in Google Search or its ranking.

Does blocking Google-Extended remove me from Gemini?

It stops Google using your content for grounding in Gemini Apps, which is how Gemini brings in fresh pages at answer time. If you want Gemini to cite you, leave it allowed.

Where does Gemini get its sources?

From the Google Search index at answer time. In our test it cited 9.2 sources per answer, most often reddit.com (20% of answers), then vendor guides and directories.

Does Reddit help with Gemini?

Some. reddit.com was cited in 22 of 110 Gemini answers in our test (20%).

Is Gemini the same as Google AI Mode?

No. They are separate products with different citation habits. In our test Google AI Mode cited Reddit in 40% of answers and Gemini in 20%.

Do Gemini answers change between runs?

We did not re-ask Gemini in this test, so we have no number for it. ChatGPT and Perplexity named the same brand in every run only 28 to 29% of the time, and we would not assume Gemini is steadier.

How do I check if Gemini mentions my product?

Ask it your buyer questions several times and count the answers that name you. The paid welaunch plans ask Gemini 4 times per question every week and keep every answer.

The other engines

ChatGPT Perplexity Google AI Mode

Does ChatGPT recommend you, or your competitor?