Korva is a scroll of the internet’s best ideas.
YouTube reviews feed AI answers
The AEO race
Answer engine optimization follows the logic of old-school SEO, where thousands of pages were pumped out. Google penalized that, and people stopped, but AEO has brought it back in part. Victor tracks this by having an agent read each competitor's sitemap on a schedule. Competitors are publishing 100 to 200 pages a week, much of it AI slop.
What content must do
Slop at scale isn't what wins. The AI crawler takes in what a site says, and the content has to be valuable enough to shape what gets written when someone asks for the best tool for a task. The most important sources are YouTube, Reddit, and the social narrative around the product.
YouTube and press
YouTube reviews are long form, so they rank for the long tail, and they have a high citation rate in ChatGPT. Press coverage works the same way. A TechCrunch feature or a Product Hunt launch matters less for a traffic wave than for citations: the more credible external sources write about a product, the more AEO traffic it sees.
How do AI assistants decide what to cite?
Assistants with web access run searches when a question needs fresh or specific information. They read the top results, pull passages from them, and write an answer that links the pages used. This design is called retrieval-augmented generation, a term from a 2020 paper by Patrick Lewis and colleagues at Facebook AI Research. Pages that appear in the underlying search results are the pool the answer draws from.
Why did Google crack down on mass-produced pages?
Google's Panda update in February 2011 targeted thin, low-quality content that had been churned out to rank. In March 2024 the company added a spam policy against scaled content abuse, which covers pages produced in bulk mainly to manipulate rankings, whether written by people or automation. Penalties can reduce or remove a site's visibility in results.
Why do long reviews catch long-tail searches?
Chris Anderson named the long tail in a 2004 Wired article about demand spread across many niche items. In search, the long tail is the mass of specific, rarely repeated queries. A long review covers many features, comparisons, and use cases, so it matches many different phrasings. A short page matches only a few.
More from The Twenty Minute VC (20VC): Venture Capital | Startup Funding | The Pitch
