Why AI Overviews is useless at sending traffic
To rank in Google AI Overviews in 2026, treat citation as the target instead of a ranking position. 83% of AIO citations come from pages…
Why AI Overviews is useless at sending traffic

To rank in Google AI Overviews in 2026, treat citation as the target instead of a ranking position. 83% of AIO citations come from pages outside the Google top 10. Ship an answer capsule under every H2, semantic HTML comparison tables, third-party citations, and named-author schema.
Look at your Google Search Console right now.
Sort by impressions. Find a query where you rank on page 1 and earn zero clicks. That’s your first receipt.
Ours is “emotional copywriting frameworks for facebook ads ecommerce”. 698 impressions in the last 76 days. Position 8.73. Zero clicks. Zero.
The click didn’t disappear. It got absorbed into the AI Overview firing above our organic result. The click that used to convert to a lead now converts to a sentence in Google’s answer. And unless our brand is inside that sentence, we’re paying for the impression with none of the click-through it used to earn.
83% of AI Overview citations come from pages outside the Google top 10
That’s Seer’s 2026 finding. Ranking position #1 isn’t the winning condition anymore. Being cited is.
Our best-cro-agency-uk-2026 page ranks Google position 22.4. That’s not a page-one ranking. It’s page three. And it earns 1,200 Bing Copilot citations across 90 days. On the query “best Shopify CRO agencies UK” it wins a 62.75% citation share (the single highest-share query in our Bing Webmaster Tools report).
Same page. Two entirely different visibility outcomes. One channel doesn’t care about the other.
AI Mode is a consensus engine, not a ranking engine
Ahrefs studied 75,000 brands across 76 million AI Overviews. They found brand mentions correlate 0.664 with AI Mode citation. Backlinks correlate 0.218.
That’s a 3x gap. On the specific job AIO is doing (recommend brands the model has “heard about”), classical link building is not the lever.
AI Mode isn’t ranking a page. It’s aggregating a consensus. It cites brands that the model has seen mentioned in multiple contexts across its training corpus. That’s a different signal than PageRank.
Which means the tactics are different too.
The four tactics that actually move the needle for AIO
1. Ship the answer capsule. A 40–60 word standalone summary passage directly below each H1 and H2. The Princeton GEO study found quotations lift AI citation likelihood by 41%, statistics by 32%, and inline citations by 30%. This is the highest-lifting structural pattern in the study.
2. Build in third-party trust signals. Muck Rack + Seer’s 25M-link study found pages carrying third-party trust signals are cited by AI engines up to 75 times more often than pages without them. Earned media accounts for 84% of AI citations. Hyperlink every numerical claim inline.
3. Pair every pillar with a YouTube video. AIO reward YouTube heavily (18.8% of AIO top-10 source share is YouTube per Profound). Google owns both YouTube and AIO. A YouTube video answering the same question your written pillar answers is a double-dip citation opportunity.
4. Get to Grade 8–10 reading level. Presence AI’s 2026 benchmark found pages at Grade 8–10 earn roughly 67% of ChatGPT citations. Pages at Grade 14+ earn 18–31%. LLMs are trained on readable web content, not corporate whitepapers.
The compound outcome
Being cited inside an AI Overview lifts downstream organic click-through by 35% on the same query, per Seer’s 2026 update. Cited brands earn 120% more organic clicks per impression than uncited brands. Paid CTR runs 91% higher for cited brands too.
Citation isn’t just a visibility play. It’s a click-clawback mechanism inside the answer surface that’s currently eating publisher clicks. You appear inside the answer, or you fund the answer’s traffic without earning it.
What to do this week
Go to Google Search Console. Sort by impressions. Find your top 10 zero-click queries.
Add an answer capsule under the H2 nearest to that query. 40–60 words. Definitional. Standalone. Contain the primary keyword, the entity you want cited, and one hard number.
Ship it. Update the “last modified” date. Then in 60 days, come back and check whether that page is being cited in the AI Overview for that query. Track it in the Bing Webmaster Tools AI Performance report.
That’s your first receipt.
The full playbook lives at gogochimp.com/blog/how-to-rank-in-google-ai-overviews: the 7-factor comparison table, the 8-step tactical framework, and the first-party Bing/GSC data on the click-clawback mechanic.
If you want the multi-engine framing first (ChatGPT + Perplexity + Copilot alongside AIO), the AI Search Optimisation guide is the umbrella.
Chris
메타데이터
- post_id
- e1845afaf6e2
- slug
- why-ai-overviews-is-useless-at-sending-traffic-e1845afaf6e2
- url
- https://medium.com/@GoGoChimp/why-ai-overviews-is-useless-at-sending-traffic-e1845afaf6e2
- canonical_url
- https://medium.com/@GoGoChimp/why-ai-overviews-is-useless-at-sending-traffic-e1845afaf6e2
- author_url
- https://medium.com/@GoGoChimp
- status
- ok
- fetched_at
- 2026-07-13 06:23:13